Radiomics analysis for the prediction of locoregional recurrence of locally advanced oropharyngeal cancer and hypopharyngeal cancer

PURPOSE

By radiomic analysis of the postcontrast CT images, this study aimed to predict locoregional recurrence (LR) of locally advanced oropharyngeal cancer (OPC) and hypopharyngeal cancer (HPC).

METHODS

A total of 192 patients with stage III-IV OPC or HPC from two independent cohort were randomly split into a training cohort with 153 cases and a testing cohort with 39 cases. Only primary tumor mass was manually segmented. Radiomic features were extracted using PyRadiomics, and then the support vector machine was used to build the radiomic model with fivefold cross-validation process in the training data set. For each case, a radiomics score was generated to indicate the probability of LR.

RESULTS

There were 94 patients with LR assigned in the progression group and 98 patients without LR assigned in the stable group. There was no significant difference of TNM staging, treatment strategies and common risk factors between these two groups. For the training data set, the radiomics model to predict LR showed 83.7% accuracy and 0.832 (95% CI 0.72, 0.87) area under the ROC curve (AUC). For the test data set, the accuracy and AUC slightly declined to 79.5% and 0.770 (95% CI 0.64, 0.80), respectively. The sensitivity/specificity of training and test data set for LR prediction were 77.6%/89.6%, and 66.7%/90.5%, respectively.

CONCLUSIONS

The image-based radiomic approach could provide a reliable LR prediction model in locally advanced OPC and HPC. Early identification of those prone to post-treatment recurrence would be helpful for appropriate adjustments to treatment strategies and post-treatment surveillance.

Magnetic Resonance Imaging Features Associated with a High and Low Expression of Tumor-Infiltrating Lymphocytes: A Stratified Analysis According to Molecular Subtypes

A total of 457 patients, including 241 HR+/HER2- patients, 134 HER2+ patients, and 82 TN patients, were studied. The percentage of TILs in the stroma adjacent to the tumor cells was assessed using a 10% cutoff. The low TIL percentages were 82% in the HR+ patients, 63% in the HER2+ patients, and 56% in the TN patients (p < 0.001). MRI features such as morphology as mass or non-mass enhancement (NME), shape, margin, internal enhancement, presence of peritumoral edema, and the DCE kinetic pattern were assessed. Tumor sizes were smaller in the HR+/HER2- group (p < 0.001); HER2+ was more likely to present as NME (p = 0.031); homogeneous enhancement was mostly seen in HR+ (p < 0.001); and the peritumoral edema was present in 45% HR+, 71% HER2+, and 80% TN (p < 0.001). In each subtype, the MR features between the high- vs. low-TIL groups were compared. In HR+/HER2-, peritumoral edema was more likely to be present in those with high TILs (70%) than in those with low TILs (40%, p < 0.001). In TN, those with high TILs were more likely to present a regular shape (33%) than those with low TILs (13%, p = 0.029) and more likely to present the circumscribed margin (19%) than those with low TILs (2%, p = 0.009).

Deep Learning-based Automatic Diagnosis of Breast Cancer on MRI Using Mask R-CNN for Detection Followed by ResNet50 for Classification

RATIONALE AND OBJECTIVES

Diagnosis of breast cancer on MRI requires, first, the identification of suspicious lesions; second, the characterization to give a diagnostic impression. We implemented Mask Reginal-Convolutional Neural Network (R-CNN) to detect abnormal lesions, followed by ResNet50 to estimate the malignancy probability.

MATERIALS AND METHODS

Two datasets were used. The first set had 176 cases, 103 cancer, and 73 benign. The second set had 84 cases, 53 cancer, and 31 benign. For detection, the pre-contrast image and the subtraction images of left and right breasts were used as inputs, so the symmetry could be considered. The detected suspicious area was characterized by ResNet50, using three DCE parametric maps as inputs. The results obtained using slice-based analyses were combined to give a lesion-based diagnosis.

RESULTS

In the first dataset, 101 of 103 cancers were detected by Mask R-CNN as suspicious, and 99 of 101 were correctly classified by ResNet50 as cancer, with a sensitivity of 99/103 = 96%. 48 of 73 benign lesions and 131 normal areas were identified as suspicious. Following classification by ResNet50, only 16 benign and 16 normal areas remained as malignant. The second dataset was used for independent testing. The sensitivity was 43/53 = 81%. Of the total of 121 identified non-cancerous lesions, only 6 of 31 benign lesions and 22 normal tissues were classified as malignant.

CONCLUSION

ResNet50 could eliminate approximately 80% of false positives detected by Mask R-CNN. Combining Mask R-CNN and ResNet50 has the potential to develop a fully-automatic computer-aided diagnostic system for breast cancer on MRI.

Prediction of poor outcome in stroke patients using radiomics analysis of intraparenchymal and intraventricular hemorrhage and clinical factors

PURPOSE

To build three prognostic models using radiomics analysis of the hemorrhagic lesions, clinical variables, and their combination, to predict the outcome of stroke patients with spontaneous intracerebral hemorrhage (sICH).

MATERIALS AND METHODS

Eighty-three sICH patients were included. Among them, 40 patients (48.2%) had poor prognosis with modified Rankin scale (mRS) of 5 and 6 at discharge, and the prognostic model was built to differentiate mRS ≤ 4 vs. 5 + 6. The region of interest (ROI) of intraparenchymal hemorrhage (IPH) and intraventricular hemorrhage (IVH) were separately segmented. Features were extracted using PyRadiomics, and the support vector machine was applied to select features and build radiomics models based on IPH and IPH + IVH. The clinical models were built using multivariate logistic regression, and then the radiomics scores were combined with clinical variables to build the combined model.

RESULTS

When using IPH, the AUC for radiomics, clinical, and combined model was 0.78, 0.82, and 0.87, respectively. When using IPH + IVH, the AUC was increased to 0.80, 0.84, and 0.90, respectively. The combined model had a significantly improved AUC compared to the radiomics by DeLong test. A clinical prognostic model based on the ICH score of 0-1 only achieved AUC of 0.71.

CONCLUSIONS

The combined model using the radiomics score derived from IPH + IVH and the clinical factors could achieve a high accuracy in prediction of sICH patients with poor outcome, which may be used to assist in making the decision about the optimal care.

Differential diagnosis of prostate cancer and benign prostatic hyperplasia based on DCE-MRI using bi-directional CLSTM deep learning and radiomics

Dynamic contrast-enhanced MRI (DCE-MRI) is routinely included in the prostate MRI protocol for a long time; its role has been questioned. It provides rich spatial and temporal information. However, the contained information cannot be fully extracted in radiologists' visual evaluation. More sophisticated computer algorithms are needed to extract the higher-order information. The purpose of this study was to apply a new deep learning algorithm, the bi-directional convolutional long short-term memory (CLSTM) network, and the radiomics analysis for differential diagnosis of PCa and benign prostatic hyperplasia (BPH). To systematically investigate the optimal amount of peritumoral tissue for improving diagnosis, a total of 9 ROIs were delineated by using 3 different methods. The results showed that bi-directional CLSTM with ± 20% region growing peritumoral ROI achieved the mean AUC of 0.89, better than the mean AUC of 0.84 by using the tumor alone without any peritumoral tissue (p = 0.25, not significant). For all 9 ROIs, deep learning had higher AUC than radiomics, but only reaching the significant difference for ± 20% region growing peritumoral ROI (0.89 vs. 0.79, p = 0.04). In conclusion, the kinetic information extracted from DCE-MRI using bi-directional CLSTM may provide helpful supplementary information for diagnosis of PCa.

Diagnosis of architectural distortion on digital breast tomosynthesis using radiomics and deep learning

Purpose

To implement two Artificial Intelligence (AI) methods, radiomics and deep learning, to build diagnostic models for patients presenting with architectural distortion on Digital Breast Tomosynthesis (DBT) images.

Materials and Methods

A total of 298 patients were identified from a retrospective review, and all of them had confirmed pathological diagnoses, 175 malignant and 123 benign. The BI-RADS scores of DBT were obtained from the radiology reports, classified into 2, 3, 4A, 4B, 4C, and 5. The architectural distortion areas on craniocaudal (CC) and mediolateral oblique (MLO) views were manually outlined as the region of interest (ROI) for the radiomics analysis. Features were extracted using PyRadiomics, and then the support vector machine (SVM) was applied to select important features and build the classification model. Deep learning was performed using the ResNet50 algorithm, with the binary output of malignancy and benignity. The Gradient-weighted Class Activation Mapping (Grad-CAM) method was utilized to localize the suspicious areas. The predicted malignancy probability was used to construct the ROC curves, compared by the DeLong test. The binary diagnosis was made using the threshold of ≥ 0.5 as malignant.

Results

The majority of malignant lesions had BI-RADS scores of 4B, 4C, and 5 (148/175 = 84.6%). In the benign group, a substantial number of patients also had high BI-RADS ≥ 4B (56/123 = 45.5%), and the majority had BI-RADS ≥ 4A (102/123 = 82.9%). The radiomics model built using the combined CC+MLO features yielded an area under curve (AUC) of 0.82, the sensitivity of 0.78, specificity of 0.68, and accuracy of 0.74. If only features from CC were used, the AUC was 0.77, and if only features from MLO were used, the AUC was 0.72. The deep-learning model yielded an AUC of 0.61, significantly lower than all radiomics models (p<0.01), which was presumably due to the use of the entire image as input. The Grad-CAM could localize the architectural distortion areas.

Conclusion

The radiomics model can achieve a satisfactory diagnostic accuracy, and the high specificity in the benign group can be used to avoid unnecessary biopsies. Deep learning can be used to localize the architectural distortion areas, which may provide an automatic method for ROI delineation to facilitate the development of a fully-automatic computer-aided diagnosis system using combined AI strategies.

Accuracy of suprascapular notch cross-sectional area by MRI in the diagnosis of suprascapular nerve entrapment syndrome: a retrospective pilot study

BACKGROUND

Previous studies have demonstrated that morphological changes in the suprascapular notch are closely associated with suprascapular nerve entrapment syndrome (SNES). Thus, we hypothesized that the suprascapular notch cross-sectional area (SSNCSA) could be a good diagnostic parameter to assess SNES.

METHODS

We acquired suprascapular notch data from 10 patients with SNES and 10 healthy individuals who had undergone shoulder magnetic resonance imaging (S-MRI) and had no evidence of SNES. T2-weighted coronal magnetic resonance images were acquired from the shoulder. We analyzed the SSNCSA at the shoulder on S-MRI using our image-analysis program (INFINITT PACS). The SSNCSA was measured as the suprascapular notch, which was the most affected site in coronal S-MRI images.

RESULTS

The mean SSNCSA was 64.50 ± 8.93 mm2 in the control group and 44.94 ± 10.40 mm2 in the SNES group. Patients with SNES had significantly lower SSNCSA (P < 0.01) than those in the control group. Receiver operating curve analysis showed that the best cut-off of the SSNCSA was 57.49 mm2, with 80.0% sensitivity, 80.0% specificity, and an area under the curve of 0.92 (95% CI [0.79, 1.00]).

CONCLUSIONS

The SSNCSA was found to have acceptable diagnostic properties for detecting SNES. We hope that these results will help diagnose SNES objectively.

The value of the glenohumeral joint cross-sectional area as a morphological parameter of glenohumeral osteoarthritis

Glenohumeral joint (GHJ) space narrowing has been demonstrated to be an important morphologic parameter of glenohumeral osteoarthritis (GHO). However, the morphology of GHJ space is irregular because of degeneration of subchondral bone and articular cartilage. Thus, we devised GHJ cartilage cross-sectional area (GHJCCSA) as a new diagnostic morphological parameter to assess the irregular morphologic change of GHJ. GHJ samples were acquired from 33 patients with GHO and from 33 normal controls without evidence of GHO based on shoulder magnetic resonance imaging. T2-weighted coronal MRIs were collected at the GHJ level for all individuals. GHJCCSA and GHJ cartilage thickness (GHJCT) at the GHJ were measured on MRIs using a graphic measuring system. The GHJCCSA was measured as the whole cartilage cross-sectional area of the GHJ. The average GHJCCSA was 115.28 ± 17.36 mm2 in normal individuals and 61.77 ± 13.74 mm2 in the GHO group. The mean GHJCT was 2.06 ± 0.35 mm in normal individuals and 1.50 ± 0.28 mm in the GHO group. GHO patients had significantly lower GHJCCSA (P < .001) and GHJCT (P < .001) than normal individuals. Receiver operator characteristics curve analysis revealed that the optimal cutoff score of the GHJCCSA was 82.21 mm2, with a sensitivity of 97.0%, a specificity of 97.0%, and an area under the curve of 0.99 (95% CI: 0.97-1.00). Although GHJCCSA and GHJCT were both significantly associated with GHO, the GHJCCSA was a more sensitive measurement parameter.

The value of the sacroiliac joint area as a new morphological parameter of ankylosing spondylitis

A narrowed sacroiliac joint (SIJ) space has been considered to be a major morphologic parameter of ankylosing spondylitis (AS). Previous studies revealed that the sacroiliac joint thickness (SIJT) correlated with AS in patients. However, irregular narrowing is different from thickness. Thus, we devised a method using the sacroiliac joint cross-sectional area (SIJA) as a new morphological parameter for use in evaluating AS. We hypothesized that the SIJA is a key morphologic parameter in diagnosing AS. SIJ samples were collected from 107 patients with AS, and from 85 control subjects who underwent SIJ-view X-rays that revealed no evidence of AS. We measured the SIJT and SIJA at the SIJ margin on X-rays using our picture archiving and communications system. The SIJT was measured at the narrowest point between the sacrum and the ilium. The SIJA was measured as the entire cross-sectional joint space area of the SIJ in the X-ray images. The average SIJT was 3.09 ± 0.61 mm in the control group, and 1.59 ± 0.52 mm in the AS group. The average SIJA was 166.74 ± 39.98 mm2 in the control group, and 68.65 ± 24.11 mm2 in the AS group. AS patients had significantly lower SIJT (P < .001) and SIJA (P < .001) than the control subjects. Receiver operating characteristics curve analysis showed that the best cutoff point for the SIJT was 2.33 mm, with 92.5% sensitivity, 94.1% specificity, and an area under the curve of 0.97 (95% confidence interval: 0.95-0.99). The optimal cutoff point for the SIJA was 106.19 mm2, with 93.5% sensitivity, 95.3% specificity, and an area under the curve of 0.98 (95% confidence interval: 0.97-1.00). Although the SIJT and SIJA were both significantly associated with AS, the SIJA parameter was a more sensitive measurement. We concluded that the SIJA is an easy-to-use, fast, cheap, and useful new morphological parameter for predicting AS.

The Added Value of Intraventricular Hemorrhage on the Radiomics Analysis for the Prediction of Hematoma Expansion of Spontaneous Intracerebral Hemorrhage

Background: Among patients undergoing head computed tomography (CT) scans within 3 h of spontaneous intracerebral hemorrhage (sICH), 28% to 38% have hematoma expansion (HE) on follow-up CT. This study aimed to predict HE using radiomics analysis and investigate the impact of intraventricular hemorrhage (IVH) compared with the conventional approach based on intraparenchymal hemorrhage (IPH) alone. Methods: This retrospective study enrolled 127 patients with baseline and follow-up non-contrast CT (NCCT) within 4~72 h of sICH. IPH and IVH were outlined separately for performing radiomics analysis. HE was defined as an absolute hematoma growth > 6 mL or percentage growth > 33% of either IPH (HEP) or a combination of IPH and IVH (HEP+V) at follow-up. Radiomic features were extracted using PyRadiomics, and then the support vector machine (SVM) was used to build the classification model. For each case, a radiomics score was generated to indicate the probability of HE. Results: There were 57 (44.9%) HEP and 70 (55.1%) non-HEP based on IPH alone, and 58 (45.7%) HEP+V and 69 (54.3%) non-HEP+V based on IPH + IVH. The majority (>94%) of HE patients had poor early outcomes (death or modified Rankin Scale > 3 at discharge). The radiomics model built using baseline IPH to predict HEP (RMP) showed 76.4% accuracy and 0.73 area under the ROC curve (AUC). The other model using IPH + IVH to predict HEP+V (RMP+V) had higher accuracy (81.9%) with AUC = 0.80, and this model could predict poor outcomes. The sensitivity/specificity of RMP and RMP+V for HE prediction were 71.9%/80.0% and 79.3%/84.1%, respectively. Conclusion: The proposed radiomics approach with additional IVH information can improve the accuracy in prediction of HE, which is associated with poor clinical outcomes. A reliable radiomics model may provide a robust tool to help manage ICH patients and to enroll high-risk ICH cases into anti-expansion or neuroprotection drug trials.

A new diagnostic morphological parameter for the Carpal tunnel syndrome: The palmaris longus tendon cross-sectional area

Carpal tunnel syndrome (CTS) is correlated with increased intracarpal canal pressure (ICP). The effect of palmaris longus tendon (PLT) loading on ICP is documented in previous researches. PLT loading induces the greatest absolute increase in ICP. Therefore, to analyze the connection between the PLT and CTS, we newly made the measurement of the PLT cross-sectional area (PLTCSA). We assumed that PLTCSA is a reliable diagnostic parameter in the CTS. PLTCSA measurement data were acquired from 21 patients with CTS, and from 21 normal subjects who underwent wrist magnetic resonance imaging (W-MRI). We measured the PLTCSA at the level of pisiform on W-MRI. The PLTCSA was measured on the outlining of PLT. The two different cutoff values in the analysis were determined using receiver operating characteristic (ROC) analysis. The mean PLTCSA was 2.34 ± 0.82 mm2 in the normal group and 3.97 ± 1.18 mm2 in the CTS group. ROC curve analysis concluded that the best cutoff point for the PLTCSA was 2.81 mm2, with 76.2% sensitivity, 71.4% specificity, and area under the curve of 0.88 (95% CI, 0.78-0.98). PLTCSA is a sensitive, new, objective morphological parameter for evaluating CTS.

Dynamic change of COVID-19 lung infection evaluated using co-registration of serial chest CT images

Purpose

To evaluate the volumetric change of COVID-19 lesions in the lung of patients receiving serial CT imaging for monitoring the evolution of the disease and the response to treatment.

Materials and methods

A total of 48 patients, 28 males and 20 females, who were confirmed to have COVID-19 infection and received chest CT examination, were identified. The age range was 21-93 years old, with a mean of 54 ± 18 years. Of them, 33 patients received the first follow-up (F/U) scan, 29 patients received the second F/U scan, and 11 patients received the third F/U scan. The lesion region of interest (ROI) was manually outlined. A two-step registration method, first using the Affine alignment, followed by the non-rigid Demons algorithm, was developed to match the lung areas on the baseline and F/U images. The baseline lesion ROI was mapped to the F/U images using the obtained geometric transformation matrix, and the radiologist outlined the lesion ROI on F/U CT again.

Results

The median (interquartile range) lesion volume (cm³) was 30.9 (83.1) at baseline CT exam, 18.3 (43.9) at first F/U, 7.6 (18.9) at second F/U, and 0.6 (19.1) at third F/U, which showed a significant trend of decrease with time. The two-step registration could significantly decrease the mean squared error (MSE) between baseline and F/U images with p < 0.001. The method could match the lung areas and the large vessels inside the lung. When using the mapped baseline ROIs as references, the second-look ROI drawing showed a significantly increased volume, p < 0.05, presumably due to the consideration of all the infected areas at baseline.

Conclusion

The results suggest that the registration method can be applied to assist in the evaluation of longitudinal changes of COVID-19 lesions on chest CT.

Prediction of Intraparenchymal Hemorrhage Progression and Neurologic Outcome in Traumatic Brain Injury Patients Using Radiomics Score and Clinical Parameters

(1) Background: Radiomics analysis of spontaneous intracerebral hemorrhages on computed tomography (CT) images has been proven effective in predicting hematoma expansion and poor neurologic outcome. In contrast, there is limited evidence on its predictive abilities for traumatic intraparenchymal hemorrhage (IPH). (2) Methods: A retrospective analysis of 107 traumatic IPH patients was conducted. Among them, 45 patients (42.1%) showed hemorrhagic progression of contusion (HPC) and 51 patients (47.7%) had poor neurological outcome. The IPH on the initial CT was manually segmented for radiomics analysis. After feature extraction, selection and repeatability evaluation, several machine learning algorithms were used to derive radiomics scores (R-scores) for the prediction of HPC and poor neurologic outcome. (3) Results: The AUCs for R-scores alone to predict HPC and poor neurologic outcome were 0.76 and 0.81, respectively. Clinical parameters were used to build comparison models. For HPC prediction, variables including age, multiple IPH, subdural hemorrhage, Injury Severity Score (ISS), international normalized ratio (INR) and IPH volume taken together yielded an AUC of 0.74, which was significantly (p = 0.022) increased to 0.83 after incorporation of the R-score in a combined model. For poor neurologic outcome prediction, clinical variables of age, Glasgow Coma Scale, ISS, INR and IPH volume showed high predictability with an AUC of 0.92, and further incorporation of the R-score did not improve the AUC. (4) Conclusion: The results suggest that radiomics analysis of IPH lesions on initial CT images has the potential to predict HPC and poor neurologic outcome in traumatic IPH patients. The clinical and R-score combined model further improves the performance of HPC prediction.

A Multiparametric Method Based on Clinical and CT-Based Radiomics to Predict the Expression of p53 and VEGF in Patients With Spinal Giant Cell Tumor of Bone

Purpose

This project aimed to assess the significance of vascular endothelial growth factor (VEGF) and p53 for predicting progression-free survival (PFS) in patients with spinal giant cell tumor of bone (GCTB) and to construct models for predicting these two biomarkers based on clinical and computer tomography (CT) radiomics to identify high-risk patients for improving treatment.

Material and Methods

A retrospective study was performed from April 2009 to January 2019. A total of 80 patients with spinal GCTB who underwent surgery in our institution were identified. VEGF and p53 expression and clinical and general imaging information were collected. Multivariate Cox regression models were used to verify the prognostic factors. The radiomics features were extracted from the regions of interest (ROIs) in preoperative CT, and then important features were selected by the SVM to build classification models, evaluated by 10-fold crossvalidation. The clinical variables were processed using the same method to build a conventional model for comparison.

Results

The immunohistochemistry of 80 patients was obtained: 49 with high-VEGF and 31 with low-VEGF, 68 with wild-type p53, and 12 with mutant p53. p53 and VEGF were independent prognostic factors affecting PFS found in multivariate Cox regression analysis. For VEGF, the Spinal Instability Neoplastic Score (SINS) was greater in the high than low groups, p &lt; 0.001. For p53, SINS (p = 0.030) and Enneking stage (p = 0.017) were higher in mutant than wild-type groups. The VEGF radiomics model built using 3 features achieved an area under the curve (AUC) of 0.88, and the p53 radiomics model built using 4 features had an AUC of 0.79. The conventional model built using SINS, and the Enneking stage had a slightly lower AUC of 0.81 for VEGF and 0.72 for p53.

Conclusion

p53 and VEGF are associated with prognosis in patients with spinal GCTB, and the radiomics analysis based on preoperative CT provides a feasible method for the evaluation of these two biomarkers, which may aid in choosing better management strategies.

Body Map of Droplet Distributions During Oropharyngeal Suction to Protect Health Care Workers From Airborne Diseases

PURPOSE

Health care workers (HCWs), and in particular anesthesia providers, often must perform aerosol-generating medical procedures (AGMPs). However, no studies have analyzed droplet distributions on the bodies of HCWs during AGMPs. Therefore, the purpose of this study was to assess and analyze droplet distributions on the bodies of HCWs during suction of oral cavities with and without oral airways and during extubations.

DESIGN

Using a quasi-experiemental design, we assumed the HCWs perform suction and extubation on intubated patients, and we prepared an intubated mannequin mimicking a patient. This study performed the oral suction and extubation on the intubated mannequin (with or without oral airways in place) and analyzed the droplet distributions.

METHODS

We prepared a mannequin intubated with an 8.0 mm endotracheal tube, assuming the situation of general anesthesia. We designed the body mapping gown, and divided it into 10 areas including the head, neck, chest, abdomen, upper arms, forearms, and hands. We classified experiments into group O when suctions were performed on the mannequin with an oral airway, and into group X when the suctions were performed on the mannequin without an oral airway. An experienced board-certified anesthesiologist performed 10 oral suctions on each mannequin, and 10 extubations. We counted the droplets on the anesthesiologist's gown according to the divided areas after each procedure.

FINDINGS

The mean droplet count after suction was 6.20 ± 2.201 in group O and 13.6 ± 4.300 in group X, with a significant difference between the two groups (P < .001). The right and left hands were the most contaminated areas in group O (2.8 ± 1.033 droplets and 2.0 ± 0.943 droplets, respectively). The abdomen, right hand, left forearm, and left hand showed many droplets in group X. (1.3 ± 1.337 droplets, 3.1 ± 1.792 droplets, 3.2 ± 3.910 droplets, and 4.3 ± 2.214 droplets, respectively). The chest, abdomen, and left hand presented significantly more droplets in group X than in group O. The trunk area (chest and abdomen) was exposed to more droplets during extubations than during suctions.

CONCLUSIONS

During suctions, more droplets are splattered from mannequins without oral airways than from those with oral airways. The right and left hands were the most contaminated areas in group O. Moreover, the abdomen, right hand, left forearm, and left hand presented a lot of droplets in group X. In addition, extubations contaminate wider areas (the head, neck, chest and abdomen) of an HCW than suctions.

Usage of image registration and three-dimensional visualization tools on serial computed tomography for the analysis of patients with traumatic intraparenchymal hemorrhages

The aim of this study was to apply registration and three-dimensional (3D) display tools to assess the evolution of intraparenchymal hemorrhage (IPH) in patients with traumatic brain injury (TBI). We identified 109 TBI patients who had two computed tomography (CT) scans within 4 days retrospectively. The IPH was manually outlined. The registration was performed in 39 lesions from 29 patients with lesion volume < 1.5 cm on both baseline and follow-up CT. The center of mass (COM) of each lesion was calculated, and the distance between baseline and follow-up CT was used to evaluate the registration effect. The mean distances of COM before registration in the XYZ, XY, and YZ coordinates were 20.5 ± 10.2 mm, 17.8 ± 9.4 mm, and 15.9 ± 9.4 mm, respectively, which decreased significantly (p < 0.001) to 7.9 ± 4.9, 7.8 ± 5.0, and 6.1 ± 4.1 mm after registration. A 3D short video displaying the rendering view of all lesions in 34 randomly selected patients from baseline and follow-up scans were presented side-by-side for comparison. The detection rate of new IPH lesions increased in 3D videos (100%) as compared with axial CT slices (78.6-92.9%). A very high interrater agreement (k = 0.856) on perceiving IPH lesion progression upon viewing 3D video was noted, and the absolute volume increase was significantly higher (p < 0.001) for progressive lesions (median 7.36 cc) over non-progressive lesions (median 0.01 cc). Compared to patients with spontaneous hemorrhagic stroke, evaluation of multiple small traumatic hemorrhages in TBI is more challenging. The applied image analysis and visualization methods may provide helpful tools for comparing changes between serial CT scans.

Automatic Detection and Segmentation of Breast Cancer on MRI Using Mask R-CNN Trained on Non-Fat-Sat Images and Tested on Fat-Sat Images

RATIONALE AND OBJECTIVES

Computer-aided methods have been widely applied to diagnose lesions on breast magnetic resonance imaging (MRI). The first step was to identify abnormal areas. A deep learning Mask Regional Convolutional Neural Network (R-CNN) was implemented to search the entire set of images and detect suspicious lesions.

MATERIALS AND METHODS

Two DCE-MRI datasets were used, 241 patients acquired using non-fat-sat sequence for training, and 98 patients acquired using fat-sat sequence for testing. All patients have confirmed unilateral mass cancers. The tumor was segmented using fuzzy c-means clustering algorithm to serve as the ground truth. Mask R-CNN was implemented with ResNet-101 as the backbone. The neural network output the bounding boxes and the segmented tumor for evaluation using the Dice Similarity Coefficient (DSC). The detection performance, and the trade-off between sensitivity and specificity, was analyzed using free response receiver operating characteristic.

RESULTS

When the precontrast and subtraction image of both breasts were used as input, the false positive from the heart and normal parenchymal enhancements could be minimized. The training set had 1469 positive slices (containing lesion) and 9135 negative slices. In 10-fold cross-validation, the mean accuracy = 0.86 and DSC = 0.82. The testing dataset had 1568 positive and 7264 negative slices, with accuracy = 0.75 and DSC = 0.79. When the obtained per-slice results were combined, 240 of 241 (99.5%) lesions in the training and 98 of 98 (100%) lesions in the testing datasets were identified.

CONCLUSION

Deep learning using Mask R-CNN provided a feasible method to search breast MRI, localize, and segment lesions. This may be integrated with other artificial intelligence algorithms to develop a fully automatic breast MRI diagnostic system.

Diagnosis of Breast Cancer Using Radiomics Models Built Based on Dynamic Contrast Enhanced MRI Combined With Mammography

Objective

To build radiomics models using features extracted from DCE-MRI and mammography for diagnosis of breast cancer.

Materials and Methods

266 patients receiving MRI and mammography, who had well-enhanced lesions on MRI and histologically confirmed diagnosis were analyzed. Training dataset had 146 malignant and 56 benign, and testing dataset had 48 malignant and 18 benign lesions. Fuzzy-C-means clustering algorithm was used to segment the enhanced lesion on subtraction MRI maps. Two radiologists manually outlined the corresponding lesion on mammography by consensus, with the guidance of MRI maximum intensity projection. Features were extracted using PyRadiomics from three DCE-MRI parametric maps, and from the lesion and a 2-cm bandshell margin on mammography. The support vector machine (SVM) was applied for feature selection and model building, using 5 datasets: DCE-MRI, mammography lesion-ROI, mammography margin-ROI, mammography lesion+margin, and all combined.

Results

In the training dataset evaluated using 10-fold cross-validation, the diagnostic accuracy of the individual model was 83.2% for DCE-MRI, 75.7% for mammography lesion, 64.4% for mammography margin, and 77.2% for lesion+margin. When all features were combined, the accuracy was improved to 89.6%. By adding mammography features to MRI, the specificity was significantly improved from 69.6% (39/56) to 82.1% (46/56), p<0.01. When the developed models were applied to the independent testing dataset, the accuracy was 78.8% for DCE-MRI and 83.3% for combined MRI+Mammography.

Conclusion

The radiomics model built from the combined MRI and mammography has the potential to provide a machine learning-based diagnostic tool and decrease the false positive diagnosis of contrast-enhanced benign lesions on MRI.

Clinical Significance of Preoperative CT and MR Imaging Findings in the Prediction of Postoperative Recurrence of Spinal Giant Cell Tumor of Bone

Objectives

To explore the predictive value of preoperative imaging in patients with spinal giant cell tumor of bone (GCTB) for postoperative recurrence and risk stratification.

Methods

Clinical data for 62 cases of spinal GCTB diagnosed and treated at our hospital from 2008 to 2018 were identified. All patients were followed up for more than 2 years according to the clinical guidelines after surgery. Medical history data including baseline demographic and clinical characteristics, computed tomography (CT) and magnetic resonance imaging (MRI) findings of recurrent and non‐recurrent patients were compared. Two musculoskeletal radiologists read the images and were blinded to the clinical data. The imaging features associated with postoperative recurrence were analyzed by multivariate logistic regression, and receiver operating characteristic (ROC) curve analysis was performed to determine the optimal cutoff value of the largest lesion diameter predicting recurrence after surgery.

Results

According to whether the disease recurred within the follow‐up period, patients were divided into the recurrence group and the non‐recurrence group. Of 62 patients (29 males and 33 females), 17 had recurrence and 45 did not. The recurrence rate was 27.4%. The mean follow‐up time was 73.66 (± 32.92) months. The three major treatments were total en bloc spondylectomy (n = 26), intralesional spondylectomy (n = 20), and curettage(n = 16). A total of 16 CT and MRI features were analyzed. A univariate analysis showed no significant difference in age, sex, treatment, multi‐vertebral body involvement, location, boundary, expansile mass, residual bone crest, paravertebral soft tissue mass, CT value, and MRI signal on T1‐weighted imaging (WI), T2‐WI, and T2‐WI fat suppression (FS) sequences (P > 0.05). The largest lesion diameter [(4.68 ± 1.79) vs (5.92 ± 2.17) cm, t = 2.287, P = 0.026] and the vertebral compression fracture (51% vs 82%, χ² = 5.005, P = 0.025) were significantly different between the non‐recurrence and recurrence groups. Logistic regression analysis showed that both largest lesion diameter (odds ratio [OR], 1.584; 95% confidence interval [CI], 1.108–2.264; P = 0.012) and compression fracture (OR, 8.073; 95%CI, 1.481–11.003; P = 0.016) were independent predictors of postoperative recurrence. When we set the cutoff value for the largest lesion diameter at 4.2 cm, the sensitivity and specificity for distinguishing the recurrence and non‐recurrence of GCTB were 94.1% and 42.2%, respectively, and the area under the curve (AUC) was 0.671. The combined model achieved a sensitivity, specificity and accuracy of 47.1%, 97.8% and 83.9%, respectively.

Conclusions

In spinal GCTB, maximum lesion diameter and the vertebral compression fracture are associated with tumor recurrence after surgery, which may provide helpful information for planning personalized treatment.

BI-RADS Reading of Non-Mass Lesions on DCE-MRI and Differential Diagnosis Performed by Radiomics and Deep Learning

Background

A wide variety of benign and malignant processes can manifest as non-mass enhancement (NME) in breast MRI. Compared to mass lesions, there are no distinct features that can be used for differential diagnosis. The purpose is to use the BI-RADS descriptors and models developed using radiomics and deep learning to distinguish benign from malignant NME lesions.

Materials and Methods

A total of 150 patients with 104 malignant and 46 benign NME were analyzed. Three radiologists performed reading for morphological distribution and internal enhancement using the 5th BI-RADS lexicon. For each case, the 3D tumor mask was generated using Fuzzy-C-Means segmentation. Three DCE parametric maps related to wash-in, maximum, and wash-out were generated, and PyRadiomics was applied to extract features. The radiomics model was built using five machine learning algorithms. ResNet50 was implemented using three parametric maps as input. Approximately 70% of earlier cases were used for training, and 30% of later cases were held out for testing.

Results

The diagnostic BI-RADS in the original MRI report showed that 104/104 malignant and 36/46 benign lesions had a BI-RADS score of 4A–5. For category reading, the kappa coefficient was 0.83 for morphological distribution (excellent) and 0.52 for internal enhancement (moderate). Segmental and Regional distribution were the most prominent for the malignant group, and focal distribution for the benign group. Eight radiomics features were selected by support vector machine (SVM). Among the five machine learning algorithms, SVM yielded the highest accuracy of 80.4% in training and 77.5% in testing datasets. ResNet50 had a better diagnostic performance, 91.5% in training and 83.3% in testing datasets.

Conclusion

Diagnosis of NME was challenging, and the BI-RADS scores and descriptors showed a substantial overlap. Radiomics and deep learning may provide a useful CAD tool to aid in diagnosis.

Multi-parametric MRI (mpMRI) for treatment response assessment of radiation therapy

Magnetic resonance imaging (MRI) plays an important role in the modern radiation therapy (RT) workflow. In comparison with computed tomography (CT) imaging, which is the dominant imaging modality in RT, MRI possesses excellent soft-tissue contrast for radiographic evaluation. Based on quantitative models, MRI can be used to assess tissue functional and physiological information. With the developments of scanner design, acquisition strategy, advanced data analysis, and modeling, multiparametric MRI (mpMRI), a combination of morphologic and functional imaging modalities, has been increasingly adopted for disease detection, localization, and characterization. Integration of mpMRI techniques into RT enriches the opportunities to individualize RT. In particular, RT response assessment using mpMRI allows for accurate characterization of both tissue anatomical and biochemical changes to support decision-making in monotherapy of radiation treatment and/or systematic cancer management. In recent years, accumulating evidence have, indeed, demonstrated the potentials of mpMRI in RT response assessment regarding patient stratification, trial benchmarking, early treatment intervention, and outcome modeling. Clinical application of mpMRI for treatment response assessment in routine radiation oncology workflow, however, is more complex than implementing an additional imaging protocol; mpMRI requires additional focus on optimal study design, practice standardization, and unified statistical reporting strategy to realize its full potential in the context of RT. In this article, the mpMRI theories, including image mechanism, protocol design, and data analysis, will be reviewed with a focus on the radiation oncology field. Representative works will be discussed to demonstrate how mpMRI can be used for RT response assessment. Additionally, issues and limits of current works, as well as challenges and potential future research directions, will also be discussed.

Improving CBCT quality to CT level using deep learning with generative adversarial network

PURPOSE

To improve image quality and computed tomography (CT) number accuracy of daily cone beam CT (CBCT) through a deep learning methodology with generative adversarial network.

METHODS

One hundred and fifty paired pelvic CT and CBCT scans were used for model training and validation. An unsupervised deep learning method, 2.5D pixel-to-pixel generative adversarial network (GAN) model with feature mapping was proposed. A total of 12 000 slice pairs of CT and CBCT were used for model training, while ten-fold cross validation was applied to verify model robustness. Paired CT-CBCT scans from an additional 15 pelvic patients and 10 head-and-neck (HN) patients with CBCT images collected at a different machine were used for independent testing purpose. Besides the proposed method above, other network architectures were also tested as: 2D vs 2.5D; GAN model with vs without feature mapping; GAN model with vs without additional perceptual loss; and previously reported models as U-net and cycleGAN with or without identity loss. Image quality of deep-learning generated synthetic CT (sCT) images was quantitatively compared against the reference CT (rCT) image using mean absolute error (MAE) of Hounsfield units (HU) and peak signal-to-noise ratio (PSNR). The dosimetric calculation accuracy was further evaluated with both photon and proton beams.

RESULTS

The deep-learning generated sCTs showed improved image quality with reduced artifact distortion and improved soft tissue contrast. The proposed algorithm of 2.5 Pix2pix GAN with feature matching (FM) was shown to be the best model among all tested methods producing the highest PSNR and the lowest MAE to rCT. The dose distribution demonstrated a high accuracy in the scope of photon-based planning, yet more work is needed for proton-based treatment. Once the model was trained, it took 11-12 ms to process one slice, and could generate a 3D volume of dCBCT (80 slices) in less than a second using a NVIDIA GeForce GTX Titan X GPU (12 GB, Maxwell architecture).

CONCLUSION

The proposed deep learning algorithm is promising to improve CBCT image quality in an efficient way, thus has a potential to support online CBCT-based adaptive radiotherapy.

Pre-operative MRI Radiomics for the Prediction of Progression and Recurrence in Meningiomas

Objectives: A subset of meningiomas may show progression/recurrence (P/R) after surgical resection. This study applied pre-operative MR radiomics based on support vector machine (SVM) to predict P/R in meningiomas.

Methods: From January 2007 to January 2018, 128 patients with pathologically confirmed WHO grade I meningiomas were included. Only patients who had undergone pre-operative MRIs and post-operative follow-up MRIs for more than 1 year were studied. Pre-operative T2WI and contrast-enhanced T1WI were analyzed. On each set of images, 32 first-order features and 75 textural features were extracted. The SVM classifier was utilized to evaluate the significance of extracted features, and the most significant four features were selected to calculate SVM score for each patient.

Results: Gross total resection (Simpson grades I–III) was performed in 93 (93/128, 72.7%) patients, and 19 (19/128, 14.8%) patients had P/R after surgery. Subtotal tumor resection, bone invasion, low apparent diffusion coefficient (ADC) value, and high SVM score were more frequently encountered in the P/R group (p < 0.05). In multivariate Cox hazards analysis, bone invasion, ADC value, and SVM score were high-risk factors for P/R (p < 0.05) with hazard ratios of 7.31, 4.67, and 8.13, respectively. Using the SVM score, an AUC of 0.80 with optimal cutoff value of 0.224 was obtained for predicting P/R. Patients with higher SVM scores were associated with shorter progression-free survival (p = 0.003).

Conclusions: Our preliminary results showed that pre-operative MR radiomic features may have the potential to offer valuable information in treatment planning for meningiomas.

Automatic breast tissue segmentation in MRIs with morphology snake and deep denoiser training via extended Stein's unbiased risk estimator

Accurate segmentation of the breast tissue is a significant challenge in the analysis of breast MR images, especially analysis of breast images with low contrast. Most of the existing methods for breast segmentation are semi-automatic and limited in their ability to achieve accurate results. This is because of difficulties in removing landmarks from noisy magnetic resonance images (MRI). Especially, when tumour is imaged for scanning, how to isolate the tumour region from chest will directly affect the accuracy for tumour to be detected. Due to low intensity levels and the close connection between breast and chest portion in MRIs, this study proposes an innovative, fully automatic and fast segmentation approach which combines histogram with inverse Gaussian gradient for morphology snakes, along with extended Stein's unbiased risk estimator (eSURE) applied for unsupervised learning of deep neural network Gaussian denoisers, aimed at accurate identification of landmarks such as chest and breast.

Prediction of the early recurrence in spinal giant cell tumor of bone using radiomics of preoperative CT: Long-term outcome of 62 consecutive patients

•

Characteristics of 62 patients with spinal GCTB who underwent surgery.

•

A prognostic classification model was built based on features selected by SVM.

•

The combined histogram and texture features could predict recurrence of GCTB.

Objectives

To determine if radiomics analysis based on preoperative computed tomography (CT) can predict early postoperative recurrence of giant cell tumor of bone (GCTB) in the spine.

Methods

In a retrospective review, 62 patients with pathologically confirmed spinal GCTB from March 2008 to February 2018, with a minimum follow-up of 24 months, were identified. The mean follow-up was 73.7 months (range, 28.7–152.1 months). The clinical information including age, gender, lesion location, multi-vertebral involvement, and surgical methods, were obtained. CT images acquired before the operation were retrieved for radiomics analysis. For each case, the tumor regions of interest (ROI) was manually outlined, and a total of 107 radiomics features were extracted. The features were selected via the sequential selection process by using the support vector machine (SVM), then used to construct classification models with Gaussian kernels. The differentiation between recurrence and non-recurrence groups was evaluated by ROC analysis, using 10-fold cross-validation.

Results

Of the 62 patients, 17 had recurrence with a recurrence rate of 27.4%. None of the clinical information was significantly different between the two groups. Patients receiving curettage had a higher recurrence rate (6/16 = 37.5%) compared to patients receiving TES (6/26 = 23.1%) or intralesional spondylectomy (5/20 = 25%). The final radiomics model was built using 10 selected features, which achieved an accuracy of 89% with AUC of 0.78.

Conclusions

The radiomics model developed based on pre-operative CT can achieve a high accuracy to predict the recurrence of spinal GCTB. Patients who have a high risk of early recurrence should be treated more aggressively to minimize recurrence.

Abstract PS5-32: Overall survival outcomes of combination anastrozole and fulvestrant in molecular subsets of hormone receptor-positive breast cancer

Background: The combination of anastrozole and fulvestrant (A+F) is associated with improvement in overall survival (OS) as compared to anastrozole (A) in patients (pts) with hormone receptor-positive (HR+) breast cancer (BC), specifically in pts with endocrine naïve disease or in pts with long disease-free interval (Mehta et al. NEJM 2019). We performed an IRB-approved retrospective assessment of the time-to-treatment failure (TTF) and overall survival (OS) outcomes in pts treated at our institution with A+F, and assessed the impact of A+F in various molecularly defined subsets of pts. Methods: We reviewed charts of 118 pts with advanced HR+ BC who received A+F. Pts with brain metastases were not excluded. Performance status (PS) ranged from 0-3. We compared TTF and OS outcomes in pts with and without molecular aberrations including HER2 overexpression/amplification, or mutations in PIK3CA, ESR1 or BRCA1/2. HER2 status was tested through tumor. PIK3CA and ESR1 were tested on tumor or ctDNA by next generation sequencing. Germline/somatic BRCA1/2 status was assessed either through ctDNA or germline testing. Kaplan-Meir survival curves were constructed. Primary statistical analysis was log-rank test, followed by Cox regression to estimate the hazard ratio and 95% confidence intervals. All p values are 2-sided. Results: Overall 68 patients had endocrine-sensitive (ES) disease and 50 patients had acquired endocrine-resistant (ER) disease. Median TTF was 26 months in the ES group vs. 10 months in the ER group (HR: 0.47;95% Cl: 0.30 - 0.75, p= 0.001). Median OS was 57 months in the ES group vs. 38 months in the ER group (HR: 0.51; 95% Cl: 0.30 - 0.86, p= 0.01). Among the 31 pts tested for PIK3CA mutation, 21 pts were negative (control) and 10 pts were positive (PIK3CA mutant). Median TTF was 23 months for control vs.14 months for PIK3CA mutant (HR: 0.89; 95% CI: 0.40 -1.95, p = 0.76). Median OS was not reached for control vs. 44 months for PIK3CA mutant (HR: 0.50; 95% Cl: 0.16 - 1.56, p = 0.23). Among the 35 pts tested for ESR1 mutation, 27 pts were negative (control) and 8 patients were positive (ESR1 mutant). Median TTF was 26 months for control vs. 10 months for ESR1 mutant (HR: 0.59; 95% CI: 0.23 - 1.52, p = 0.20). Median OS was 66 months for control vs. 65 months for ESR1 mutant (HR: 1.63; 95% Cl: 0.44 - 5.95, p = 0.46). In the 17 pts tested for BRCA mutation (germline or somatic), 6 of them were positive (BRCA mutant) and 11 of them were negative (control). Median TTF was 36 months for control vs. 10 months for BRCA mutant (HR: 0.40; 95% CI: 0.12 - 1.40, p = 0.09). Median OS was 35 months for control but has not been reached in the BRCA mutant group (HR: 1.16; 95% Cl: 0.23 - 5.88, p = 0.93). Of 118 patients, 92 pts were HER2 negative (control) and 26 patients were HER2 positive. Median TTF was 23 months for control vs. 14 months for HER2+ group (HR: 0.80; 95% CI: 0.46 - 1.39, p = 0.42). Median OS was 52 months for control vs. 40 months for HER2+ group (HR: 0.87; 95% Cl: 0.47 - 1.64, p = 0.67). Conclusions: The time-to-treatment failure and overall survival in the endocrine-sensitive disease are 2-fold higher compared to the endocrine-resistant disease. Patients with PIK3CA, ESR1 or BRCA1/2 mutation or with HER2 overexpression/amplification historically have poor survival outcomes, but in hypothesis-generating analysis, we showed statistically similar overall survival outcomes in these patients treated with anastrozole plus fulvestrant, compared to controls without these poor prognostic features. We suggest that anastrozole plus fulvestrant should be the preferred partner with molecularly targeted agents.

Citation Format: Tianyi Tang, Yu Cheng, Ritesh Parajuli, Neha Patel, Min-Ying Su, Chaitali Nangia, Rita S Mehta. Overall survival outcomes of combination anastrozole and fulvestrant in molecular subsets of hormone receptor-positive breast cancer [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS5-32.

Radiomics Approach for Prediction of Recurrence in Non-Functioning Pituitary Macroadenomas

Objectives

A subset of non-functioning pituitary macroadenomas (NFPAs) may exhibit early progression/recurrence (P/R) after surgical resection. The purpose of this study was to apply radiomics in predicting P/R in NFPAs.

Methods

Only patients who had undergone preoperative MRI and postoperative MRI follow-ups for more than 1 year were included in this study. From September 2010 to December 2017, 50 eligible patients diagnosed with pathologically confirmed NFPAs were identified. Preoperative coronal T2WI and contrast-enhanced (CE) T1WI imaging were analyzed by computer algorithms. For each imaging sequence, 32 first-order features and 75 texture features were extracted. Support vector machine (SVM) classifier was utilized to evaluate the importance of extracted parameters, and the most significant three parameters were used to build the prediction model. The SVM score was calculated based on the three selected features.

Results

Twenty-eight patients exhibited P/R (28/50, 56%) after surgery. The median follow-up time was 38 months, and the median time to P/R was 20 months. Visual disturbance, hypopituitarism, extrasellar extension, compression of the third ventricle, large tumor height and volume, failed optic chiasmatic decompression, and high SVM score were more frequently encountered in the P/R group (p < 0.05). In multivariate Cox hazards analysis, symptoms of sex hormones, hypopituitarism, and SVM score were high risk factors for P/R (p < 0.05) with hazard ratios of 10.71, 2.68, and 6.88. The three selected radiomics features were T1 surface-to-volume radio, T1 GLCM-informational measure of correlation, and T2 NGTDM-coarseness. The radiomics predictive model shows 25 true positive, 16 true negative, 6 false positive, and 3 false negative cases, with an accuracy of 82% and AUC of 0.78 in differentiating P/R from non-P/R NFPAs. For SVM score, optimal cut-off value of 0.537 and AUC of 0.87 were obtained for differentiation of P/R. Higher SVM scores were associated with shorter progression-free survival (p < 0.001).

Conclusions

Our preliminary results showed that objective and quantitative MR radiomic features can be extracted from NFPAs. Pending more studies and evidence to support the findings, radiomics analysis of preoperative MRI may have the potential to offer valuable information in treatment planning for NFPAs.

Diagnosis of Benign and Malignant Breast Lesions on DCE-MRI by Using Radiomics and Deep Learning With Consideration of Peritumor Tissue

BACKGROUND

Computer-aided methods have been widely applied to diagnose lesions detected on breast MRI, but fully-automatic diagnosis using deep learning is rarely reported.

PURPOSE

To evaluate the diagnostic accuracy of mass lesions using region of interest (ROI)-based, radiomics and deep-learning methods, by taking peritumor tissues into consideration.

STUDY TYPE

Retrospective.

POPULATION

In all, 133 patients with histologically confirmed 91 malignant and 62 benign mass lesions for training (74 patients with 48 malignant and 26 benign lesions for testing).

FIELD STRENGTH/SEQUENCE

3T, using the volume imaging for breast assessment (VIBRANT) dynamic contrast-enhanced (DCE) sequence.

ASSESSMENT

3D tumor segmentation was done automatically by using fuzzy-C-means algorithm with connected-component labeling. A total of 99 texture and histogram parameters were calculated for each case, and 15 were selected using random forest to build a radiomics model. Deep learning was implemented using ResNet50, evaluated with 10-fold crossvalidation. The tumor alone, smallest bounding box, and 1.2, 1.5, 2.0 times enlarged boxes were used as inputs.

STATISTICAL TESTS

The malignancy probability was calculated using each model, and the threshold of 0.5 was used to make a diagnosis.

RESULTS

In the training dataset, the diagnostic accuracy was 76% using three ROI-based parameters, 84% using the radiomics model, and 86% using ROI + radiomics model. In deep learning using the per-slice basis, the area under the receiver operating characteristic (ROC) was comparable for tumor alone, smallest and 1.2 times box (AUC = 0.97-0.99), which were significantly higher than 1.5 and 2.0 times box (AUC = 0.86 and 0.71, respectively). For per-lesion diagnosis, the highest accuracy of 91% was achieved when using the smallest bounding box, and that decreased to 84% for tumor alone and 1.2 times box, and further to 73% for 1.5 times box and 69% for 2.0 times box. In the independent testing dataset, the per-lesion diagnostic accuracy was also the highest when using the smallest bounding box, 89%.

DATA CONCLUSION

Deep learning using ResNet50 achieved a high diagnostic accuracy. Using the smallest bounding box containing proximal peritumor tissue as input had higher accuracy compared to using tumor alone or larger boxes.

LEVEL OF EVIDENCE

3 Technical Efficacy: Stage 2.

Evaluation of breast stiffness measured by ultrasound and breast density measured by MRI using a prone-supine deformation model

Background

This study evaluated breast tissue stiffness measured by ultrasound elastography and the percent breast density measured by magnetic resonance imaging to understand their relationship.

Methods

Magnetic resonance imaging and whole breast ultrasound were performed in 20 patients with suspicious lesions. Only the contralateral normal breasts were analyzed. Breast tissue stiffness was measured from the echogenic homogeneous fibroglandular tissues in the central breast area underneath the nipple. An automatic, computer algorithm-based, segmentation method was used to segment the whole breast and fibroglandular tissues on three dimensional magnetic resonanceimaging. A finite element model was applied to deform the prone magnetic resonance imaging to match the supine ultrasound images, by using the inversed gravity loaded transformation. After deformation, the tissue level used in ultrasound elastography measurement could be estimated on the deformed supine magnetic resonance imaging to measure the breast density in the corresponding tissue region.

Results

The mean breast tissue stiffness was 2.3 ± 0.8 m/s. The stiffness was not correlated with age (r = 0.29). Overall, there was no positive correlation between breast stiffness and breast volume (r = - 0.14), or the whole breast percent density (r = - 0.09). There was also no correlation between breast stiffness and the local percent density measured from the corresponding region (r = - 0.12).

Conclusions

The lack of correlation between breast stiffness measured by ultrasound and the whole breast or local percent density measured by magnetic resonance imaging suggests that breast stiffness is not solely related to the amount of fibroglandular tissue. Further studies are needed to investigate whether they are dependent or independent cancer risk factors.

Radiomics approach for prediction of recurrence in skull base meningiomas

PURPOSE

A subset of skull base meningiomas (SBM) may show early progression/recurrence (P/R) as a result of incomplete resection. The purpose of this study is the implementation of MR radiomics to predict P/R in SBM.

METHODS

From October 2006 to December 2017, 60 patients diagnosed with pathologically confirmed SBM (WHO grade I, 56; grade II, 3; grade III, 1) were included in this study. Preoperative MRI including T2WI, diffusion-weighted imaging (DWI), and contrast-enhanced T1WI were analyzed. On each imaging modality, 13 histogram parameters and 20 textural gray level co-occurrence matrix (GLCM) features were extracted. Random forest algorithms were utilized to evaluate the importance of these parameters, and the most significant three parameters were selected to build a decision tree for prediction of P/R in SBM. Furthermore, ADC values obtained from manually placed ROI in tumor were also used to predict P/R in SBM for comparison.

RESULTS

Gross-total resection (Simpson Grades I-III) was performed in 33 (33/60, 55%) patients, and 27 patients received subtotal resection. Twenty-one patients had P/R (21/60, 35%) after a postoperative follow-up period of at least 12 months. The three most significant parameters included in the final radiomics model were T1 max probability, T1 cluster shade, and ADC correlation. In the radiomics model, the accuracy for prediction of P/R was 90%; by comparison, the accuracy was 83% using ADC values measured from manually placed tumor ROI.

CONCLUSIONS

The results show that the radiomics approach in preoperative MRI offer objective and valuable clinical information for treatment planning in SBM.

Combinatorial targeting of cancer bone metastasis using mRNA engineered stem cells

Background

Bone metastases are common and devastating to cancer patients. Existing treatments do not specifically target the disease sites and are therefore ineffective and systemically toxic. Here we present a new strategy to treat bone metastasis by targeting both the cancer cells (“the seed”), and their surrounding niche (“the soil”), using stem cells engineered to home to the bone metastatic niche and to maximise local delivery of multiple therapeutic factors.

Methods

We used mesenchymal stem cells engineered using mRNA to simultaneously express P-selectin glycoprotein ligand-1 (PSGL-1)/Sialyl-Lewis X (SLEX) (homing factors), and modified versions of cytosine deaminase (CD) and osteoprotegerin (OPG) (therapeutic factors) to target and treat breast cancer bone metastases in two mouse models, a xenograft intratibial model and a syngeneic model of spontaneous bone metastasis.

Findings

We first confirmed that MSC engineered using mRNA produced functional proteins (PSGL-1/SLEX, CD and OPG) using various in vitro assays. We then demonstrated that mRNA-engineered MSC exhibit enhanced homing to the bone metastatic niche likely through interactions between PSGL-1/SLEX and P-selectin expressed on tumour vasculature. In both the xenograft intratibial model and syngeneic model of spontaneous bone metastasis, engineered MSC can effectively kill tumour cells and preserve bone integrity. The engineered MSC also exhibited minimal toxicity in vivo, compared to its non-targeted chemotherapy counterpart (5-fluorouracil).

Interpretation

Our combinatorial targeting of both the cancer cells and the niche represents a simple, safe and effective way to treat metastatic bone diseases, otherwise difficult to manage with existing strategies. It can also be applied to other cell types (e.g., T cells) and cargos (e.g., genome editing components) to treat a broad range of cancer and other complex diseases.

Fund

National Institutes of Health, National Cancer Institute of the National Institutes of Health, Department of Defense, California Institute of Regenerative Medicine, National Science Foundation, Baylx Inc., and Fondation ARC pour la recherche sur le cancer.

Automated artificial intelligence quantification of fibroglandular tissue on breast MRI

e12071

Background: The objective of this study is to examine if a convolutional neural network can be utilized to automate breast fibroglandular tissue segmentation, a risk factor for breast cancer, on MRIs. Methods: This institutional review board approved study assessed retrospectively acquired MRI T1 pre-contrast image data for 238 patients. Ground truth parameters were derived through manual segmentation. A hybrid 3D/2D U-Net architecture was developed for fibroglandular tissue segmentation. The network was trained with T1 pre-contrast MRI data and their corresponding ground-truth labels. The analysis was started with image pre-processing. Each MRI volume was re-sampled and normalized using z-scores. Convolution operations reduced 3D volumes into a 2D slice in the contracting arm of the U-Net architecture. Results: A 5-fold cross validation was performed and the Dice similarity coefficient was used to assess the accuracy of fibroglandular tissue segmentation. Cross-validation results showed that the automated hybrid CNN approach resulted in a Dice similarity coefficient of 0.848 and a Pearson correlation of 0.961 in comparison to the ground-truth for fibroglandular breast tissue segmentation, which demonstrates high accuracy. Conclusions: The results demonstrate significant application of deep learning in accurately automating segmentation of breast fibroglandular tissue.

Automated breast cancer lesion detection on breast MRI using artificial intelligence

e14612

Background: Over 40,000 women in the US will die from breast cancer. Early detection of cancer is crucial and is a potential avenue to improve survival. The objective of this research study is to develop a convoluted neural network (CNN), a subset of artificial intelligence, in order to enhance computerized detection of breast lesions on MRIs. Methods: This is an institutional review board approved retrospective study with post contrast MRI data from 238 patients. Breast tumor segmentation was automated with a hybrid 3D/2D CNN designed adapted from U-net, a popular neural network architecture in biomedical image analysis. T1 post-contrast MRI volumes were used to train the network. The data set was separated into training (80%) and validation (20%) sets. Re-sampling and normalization using z-scores were applied to each volume before training. Contracting and expanding arms of the model consist of successive convolutions followed by batch normalization and ReLU operations. Ground truth was established through manual segmentation and previously conducted readings of the images used to train our network. Results: A 5-fold cross validation was performed for analysis. The Dice similarity coefficient was used to assess segmentation accuracy. The hybrid 3D/2D U-Net architecture yielded a Dice score of 0.753 and a Pearson correlation of 0.548 for the breast tumor segmentation. Conclusions: These results demonstrated the feasibility for artificial intelligence applications in accurately identifying the presence of lesions on breast MRI images.

Machine learning for prediction of chemoradiation therapy response in rectal cancer using pre-treatment and mid-radiation multi-parametric MRI

PURPOSE

To predict the neoadjuvant chemoradiation therapy (CRT) response in patients with locally advanced rectal cancer (LARC) using radiomics and deep learning based on pre-treatment MRI and a mid-radiation follow-up MRI taken 3-4 weeks after the start of CRT.

METHODS

A total of 51 patients were included, 45 with pre-treatment, 41 with mid-radiation therapy (RT), and 35 with both MRI sets. The multi-parametric MRI protocol included T2, diffusion weighted imaging (DWI) with b-values of 0 and 800 s/mm², and dynamic-contrast-enhanced (DCE) MRI. After completing CRT and surgery, the specimen was examined to determine the pathological response based on the tumor regression grade. The tumor ROI was manually drawn on the post-contrast image and mapped to other sequences. The total tumor volume and mean apparent diffusion coefficient (ADC) were measured. Radiomics using GLCM texture and histogram parameters, and deep learning using a convolutional neural network (CNN), were performed to differentiate pathologic complete response (pCR) vs. non-pCR, and good response (GR) vs. non-GR.

RESULTS

Tumor volume decreased and ADC increased significantly in the mid-RT MRI compared to the pre-treatment MRI. For predicting pCR vs. non-pCR, combining ROI and radiomics features achieved an AUC of 0.80 for pre-treatment, 0.82 for mid-RT, and 0.86 for both MRI together. For predicting GR vs. non-GR, the AUC was 0.91 for pre-treatment, 0.92 for mid-RT, and 0.93 for both MRI together. In deep learning using CNN, combining pre-treatment and mid-RT MRI achieved a higher accuracy compared to using either dataset alone, with AUC of 0.83 for predicting pCR vs. non-pCR.

CONCLUSION

Radiomics based on pre-treatment and early follow-up multi-parametric MRI in LARC patients receiving CRT could extract comprehensive quantitative information to predict final pathologic response.

Differentiation of spinal metastases originated from lung and other cancers using radiomics and deep learning based on DCE-MRI

PURPOSE

To differentiate metastatic lesions in the spine originated from primary lung cancer and other cancers using radiomics and deep learning, compared to traditional hot-spot ROI analysis.

METHODS

In a retrospective review of clinical spinal MRI database with a dynamic contrast enhanced (DCE) sequence, a total of 61 patients without prior cancer diagnosis and later confirmed to have metastases (30 lung; 31 non-lung cancers) were identified. For hot-spot analysis, a manual ROI was placed to calculate three heuristic parameters from the wash-in, maximum, and wash-out phases in the DCE kinetics. For each case, the 3D tumor mask was generated by using the normalized-cut algorithm. Radiomics analysis was performed to extract histogram and texture features from three DCE parametric maps. Deep learning was performed using these maps as inputs into a conventional convolutional neural network (CNN), as well as using all 12 sets of DCE images into a convolutional long short term memory (CLSTM) network.

RESULTS

For hot-spot ROI analysis, mean wash-out slope was 0.25 ± 10% for lung metastases and -9.8 ± 12.9% for other tumors. CHAID classification using a wash-out slope of -6.6% followed by wash-in enhancement ratio of 98% achieved a diagnostic accuracy of 0.79. Radiomics analysis using features representing tumor heterogeneity only reached the highest accuracy of 0.71. Classification using CNN achieved a mean accuracy of 0.71 ± 0.043, whereas a CLSTM improved accuracy to 0.81 ± 0.034.

CONCLUSIONS

DCE-MRI machine-learning analysis methods have potential to predict lung cancer metastases in the spine, which may be used to guide subsequent workup for confirmed diagnosis.

Automatic and fast segmentation of breast region-of-interest (ROI) and density in MRIs

Accurate segmentation of the breast region of interest (BROI) and breast density (BD) is a significant challenge during the analysis of breast MR images. Most of the existing methods for breast segmentation are semi-automatic and limited in their ability to achieve accurate results. This is because of difficulties in removing landmarks from noisy magnetic resonance images (MRI) due to similar intensity levels and the close connection to BROI. This study proposes an innovative, fully automatic and fast segmentation approach to identify and remove landmarks such as the heart and pectoral muscles. The BROI segmentation is carried out with a framework consisting of three major steps. Firstly, we use adaptive wiener filtering and k-means clustering to minimize the influence of noises, preserve edges and remove unwanted artefacts. The second step systematically excludes the heart area by utilizing active contour based level sets where initial contour points are determined by the maximum entropy thresholding and convolution method. Finally, a pectoral muscle is removed by using morphological operations and local adaptive thresholding on MR images. Prior to the elimination of the pectoral muscle, the MR image is sub divided into three sections: left, right, and central based on the geometrical information. Subsequently, a BD segmentation is achieved with 4 level fuzzy c-means (FCM) thresholding on the denoised BROI segmentation. The proposed method is validated using the 1350 breast images from 15 female subjects. The pixel-based quantitative analysis showed excellent segmentation results when compared with manually drawn BROI and BD. Furthermore, the presented results in terms of evaluation matrices: Acc, Sp, AUC, MR, P, Se and DSC demonstrate the high quality of segmentations using the proposed method. The average computational time for the segmentation of BROI and BD is 1 minute and 50 seconds.

Feasibility and Diagnostic Performance of Voxelwise Computed Diffusion-Weighted Imaging in Breast Cancer

BACKGROUND

Conventional diffusion-weighted imaging (DWI) with high b-values may improve lesion conspicuity, but with a low signal intensity and thus a low signal-to-noise ratio (SNR). The voxelwise computed DWI (vcDWI) may generate high-quality images with a strong lesion signal and low background.

PURPOSE

To evaluate the feasibility and diagnostic performance of vcDWI.

STUDY TYPE

Retrospective.

POPULATION

In all, 67 patients with 72 lesions, 33 malignant and 39 benign.

FIELD STRENGTH/SEQUENCE

3T, including T₂ /T₁ , DWI with two b-values, and dynamic contrast-enhanced MRI (DCE-MRI).

ASSESSMENT

Computed DWI (cDWI) with high b-values of 1500, 2000, 2500 s/mm² (cDWI₁₅₀₀ , cDWI₂₀₀₀ , cDWI₂₅₀₀ ) and vcDWI were generated from measured DWI (mDWI). The mDWI, cDWIs and vcDWI were evaluated by three readers independently to determine lesion conspicuity, background signal suppression, overall image quality using 1-5 rating scales, as well as to give BI-RADS scores. The mean apparent diffusion coefficient (ADC) value for each lesion was measured.

STATISTICAL TESTS

Agreement among the three readers was evaluated by the intraclass correlation coefficient. Receiver operating characteristic (ROC) analysis was performed to compare the diagnostic performance based on reading of mDWI, cDWIs, vcDWI, and the measured ADC values.

RESULTS

vcDWI provided the best lesion conspicuity compared with mDWI and cDWIs (P < 0.005). For overall image quality, vcDWI was significantly better than cDWI (P < 0.005), but not significantly better compared with mDWI for two readers (P = 0.037 and P = 0.013) and significantly worse for the third reader (P < 0.005). Background signal suppression was the best on cDWI₂₅₀₀ , and better on vcDWI than on mDWI, cDWI₁₅₀₀ , and cDWI₂₀₀₀ . The AUC value for differential diagnosis was 0.868 for mDWI, 0.862 for cDWI₁₅₀₀ , 0.781 for cDWI₂₀₀₀ , 0.704 for cDWI₂₅₀₀ , 0.946 for vcDWI, 0.704 for ADC value, and 0.961 for DCE-MRI. DATA CONCLUSION: vcDWI was implemented without increasing scanning time, and it provided excellent lesion conspicuity for detection of breast lesions and assisted in differentiating malignant from benign breast lesions.

LEVEL OF EVIDENCE

4 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2018.

Quantitative analysis of peri-tumor fat in different molecular subtypes of breast cancer

BACKGROUND AND PURPOSES

The aim of this study was to develop morphological analytic methods to analyze the tumor-fat interface and in different peritumoral shells away from the tumor, and to compare the results among three molecular subtypes of breast cancer.

MATERIALS AND METHODS

A total of 102 women (mean age 48.5 y/o) with solitary well-defined breast cancers were analyzed, including 46 human epidermal growth factor receptor 2 (HER2) (+), 46 HER2(-) hormonal receptor (HR) (+), and 10 triple negative (TN) breast cancers. The tumor lesion, the breast, the fibroglandular and fatty tissue were segmented using well-established methods. The whole breast fat percentage and the peri-tumor interface fat percentage were measured. Three shells (SH1, SH2, SH3) surrounding the convex hall of the three dimensional (3D) tumor were defined and in each shell the volumetric percentage of fat was calculated. The peri-tumor interface fat percentage and the volumetric percentage of fat in the three peri-tumoral shells were compared among different subtypes.

RESULTS

In the TN group, the fat percentage on the tumor boundary was 43 ± 20% and 78 ± 12% for two dimensional (2D) and 3D measurement, respectively, which were the highest among the three subtypes but not significantly different. The fat percentage in SH2 and SH3 in the TN group was 82 ± 7% and 85 ± 7%, which was significantly higher compared to the two other two subtypes. The results remained after controlling for the whole breast fat percentage.

CONCLUSIONS

This study provided a feasible method for quantitative analysis of peri-tumoral tissue characteristics. Because of small patient number, the finding that TN tumors had the highest peri-tumor fat content among the three subtypes needs to be further verified with a large cohort study.

Role of dynamic contrast-enhanced MRI in evaluating the association between contralateral parenchymal enhancement and survival outcome in ER-positive, HER2-negative, node-negative invasive breast cancer

BACKGROUND

Background parenchymal enhancement (BPE) on dynamic contrast-enhanced (DCE)-MRI has been associated with breast cancer risk, both based on qualitative and quantitative assessments.

PURPOSE

To investigate whether BPE of the contralateral breast on preoperative DCE-MRI is associated with therapy outcome in ER-positive, HER2-negative, node-negative invasive breast cancer.

STUDY TYPE

Retrospective.

POPULATION

In all, 289 patients with unilateral ER-positive, HER2-negative, node-negative breast cancer larger than 5 mm.

FIELD STRENGTH/SEQUENCE

3T, T₁ -weighted DCE sequence.

ASSESSMENT

BPE of the contralateral breast was assessed qualitatively by two dedicated radiologists and quantitatively (using region-of-interest and automatic breast segmentation).

STATISTICAL TESTS

Cox regression analysis was used to determine associations with recurrence-free survival (RFS) and distant metastasis-free survival (DFS). Interobserver variability for parenchymal enhancement was assessed using kappa statistics and intraclass correlation coefficient (ICC).

RESULTS

The median follow-up time was 75.8 months. Multivariate analysis showed receipt of total mastectomy (hazard ratio [HR]: 5.497) and high Ki-67 expression level (HR: 5.956) were independent factors associated with worse RFS (P < 0.05). Only a high Ki-67 expression level was associated with worse DFS (HR: 3.571, P = 0.045). BPE assessments were not associated with outcome (RFS [qualitative BPE: P = 0.75, 0.92 for readers 1 and 2; quantitative BPE: P = 0.38-0.99], DFS, [qualitative BPE: P = 0.41, 0.16 for readers 1 and 2; quantitative BPE: P = 0.68-0.99]). For interobserver variability, there was good agreement between qualitative (κ = 0.700) and good to perfect agreement for most quantitative parameters of BPE.

DATA CONCLUSION

Contralateral BPE showed no association with survival outcome in patients with ER-positive, HER2-negative, node-negative invasive breast cancer. A high Ki-67 expression level was associated with both worse recurrence-free and distant metastasis-free survival.

LEVEL OF EVIDENCE

3 Technical Efficacy: Stage 4 J. Magn. Reson. Imaging 2018;48:1678-1689.

[Decolorization of skin and hair-derived melanin by three ligninolytic enzymes]

Objective: To compare the decolorization efficiency of lignin peroxidase (LiP), manganese peroxidase (MnP) and laccase on eumelanin and pheomelanin, and to investigate the effect of topical administration of LiP solution on hyperpigmented guinea pigs skin induced by 308 nm excimer light. Methods: Pheomelanin-enriched specimens were prepared from human hair and cutaneous melanoma tissue using alkaline lysis method.Synthetic eumelanin was purchased from a commercial supplier.The same amount (0.02%) of melanin was incubated with the equal enzyme activity (0.2 U/ml) of ligninolytic enzymes for 3 h respectively.The absorbance at 475 nm (A(475)) in the enzyme-catalyzed solution was measured using ELISA microplate reader.The experimental hyperpigmentation model was established in the dorsal skin of brownish guinea pigs using 308 nm excimer light radiation.LiP and heat-inactivated LiP solution were topically applied at each site.Meanwhile, 3% hydroquinone and vehicle cream were used as control.The skin color (L value) was recorded using a CR-10 Minolta chromameter.Corneocytes were collected using adhesive taping method.The amount and distribution of melanin in the corneocytes and skin tissues was visualized by Fontana-Masson staining. Results: All three ligninolytic enzymes showed various degree of eumelanin and pheomelanin decolorization activity.The decolorization activity of LiP, MnP and laccase was 40%-70%, 22%-42% and 9%-21%, respectively.The similar lightening was shown in the skin treated with LiP solution and 3% hydroquinone.The amount of melanin granules in the corneocytes was 199±11 by LiP, which was less than that in untreated control (923±12) and heat-inactive control (989±13). The amount of melanin was decreased in the whole epidermis treated with hydroquinone, the epidermis thickness was increased as well. In contrast, melanin of LiP group was decreased only in the superficial epidermis, the epidermis thickness seemed to be normal. Conclusion: LiP exerts a potent decolorization activity for hair- or skin-derived pheomelanin as well as eumelanin.It remains to be further investigated whether LiP serves as a substitute for hydroquinone in skin lightening products.

Hypothermic Cooling Measured by Thermal Magnetic Resonance Imaging; Feasibility and Implications for Virtual Imaging in the Urogenital Pelvis

OBJECTIVE

To study the combination of thermal magnetic resonance imaging (MRI) and novel hypothermic cooling, via an endorectal cooling balloon (ECB), to assess the effective dispersion and temperature drop in pelvic tissue to potentially reduce inflammatory cascade in surgical applications.

METHODS

Three male subjects, before undergoing robot-assisted radical prostatectomy, were cooled via an ECB, rendered MRI compatible for patient safety before ECB hypothermia. MRI studies were performed using a 3T scanner and included T2-weighted anatomic scan for the pelvic structures, followed by a temperature mapping scan. The sequence was performed repeatedly during the cooling experiment, whereas the phase data were collected using an integrated MR-high-intensity focused ultrasound workstation in real time. Pelvic cooling was instituted with a cooling console located outside the MRI magnet room.

RESULTS

The feasibility of pelvic cooling measured a temperature drop of the ECB of 20-25 degrees in real time was achieved after an initial time delay of 10-15 seconds for the ECB to cool. The thermal MRI anatomic images of the prostate and neurovascular bundle demonstrate cooling at this interface to be 10-15 degrees, and also that cooling extends into the prostate itself ~5 degrees, and disperses into the pelvic region as well.

CONCLUSION

An MRI-compatible ECB coupled with thermal MRI is a feasible method to assess effective hypothermic diffusion and saturation to pelvic structures. By inference, hypothermia-induced rectal cooling could potentially reduce inflammation, scarring, and fistula in radical prostatectomy, as well as other urologic tissue procedures of high-intensity focused ultrasound, external beam radiation therapy, radioactive seed implants, transurethral microwave therapy, and transurethral resection of the prostate.

Diagnosis of Spinal Lesions Using Heuristic and Pharmacokinetic Parameters Measured by Dynamic Contrast-Enhanced MRI

RATIONALE AND OBJECTIVES

This study aimed to evaluate the diagnostic performance of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) in differentiation of four spinal lesions by using heuristic and pharmacokinetic parameters analyzed from DCE signal intensity time course.

MATERIALS AND METHODS

DCE-MRI of 62 subjects with confirmed myeloma (n = 9), metastatic cancer (n = 22), lymphoma (n = 7), and inflammatory tuberculosis (TB) (n = 24) in the spine were analyzed retrospectively. The region of interest was placed on strongly enhanced tissues. The DCE time course was categorized as the "wash-out," "plateau," or "persistent enhancement" pattern. The maximum enhancement, steepest wash-in enhancement, and wash-out slope using the signal intensity at 67 seconds after contrast injection as reference were measured. The Tofts 2-compartmental pharmacokinetic model was applied to obtain K^trans and k_ep. Pearson correlation between heuristic and pharmacokinetic parameters was evaluated, and receiver operating characteristic curve analysis was performed for pairwise group differentiation.

RESULTS

The mean wash-out slope was -22% ± 10% for myeloma, 1% ± 0.4% for metastatic cancer, 3% ± 3% for lymphoma, and 7% ± 10% for TB, and it could significantly distinguish myeloma from metastasis (area under the curve [AUC] = 0.884), lymphoma (AUC = 1.0), and TB (AUC = 1.0) with P = .001, and distinguish metastasis from TB (AUC = 0.741) with P = .005. The k_ep and wash-out slope were highly correlated (r = 0.92), and they showed a similar diagnostic performance. The K^trans was significantly correlated with the maximum enhancement (r = 0.71) and the steepest wash-in enhancement (r = 0.85), but they had inferior diagnostic performance compared to the wash-out slope.

CONCLUSIONS

DCE-MRI may provide additional diagnostic information, and a simple wash-out slope had the best diagnostic performance. The heuristic and pharmacokinetic parameters were highly correlated.

3D MRI for Quantitative Analysis of Quadrant Percent Breast Density: Correlation with Quadrant Location of Breast Cancer

RATIONALE AND OBJECTIVES

Breast cancer occurs more frequently in the upper outer (UO) quadrant, but whether this higher cancer incidence is related to the greater amount of dense tissue is not known. Magnetic resonance imaging acquires three-dimensional volumetric images and is the most suitable among all breast imaging modalities for regional quantification of density. This study applied a magnetic resonance imaging-based method to measure quadrant percent density (QPD), and evaluated its association with the quadrant location of the developed breast cancer.

MATERIALS AND METHODS

A total of 126 cases with pathologically confirmed breast cancer were reviewed. Only women who had unilateral breast cancer located in a clear quadrant were selected for analysis. A total of 84 women, including 47 Asian women and 37 western women, were included. An established computer-aided method was used to segment the diseased breast and the contralateral normal breast, and to separate the dense and fatty tissues. Then, a breast was further separated into four quadrants using the nipple and the centroid as anatomic landmarks. The tumor was segmented using a computer-aided method to determine its quadrant location. The distribution of cancer quadrant location, the quadrant with the highest QPD, and the proportion of cancers occurring in the highest QPD were analyzed.

RESULTS

The highest incidence of cancer occurred in the UO quadrant (36 out of 84, 42.9%). The highest QPD was also noted most frequently in the UO quadrant (31 out of 84, 36.9%). When correlating the highest QPD with the quadrant location of breast cancer, only 17 women out of 84 (20.2%) had breast cancer occurring in the quadrant with the highest QPD.

CONCLUSIONS

The results showed that the development of breast cancer in a specific quadrant could not be explained by the density in that quadrant, and further studies are needed to find the biological reasons accounting for the higher breast cancer incidence in the UO quadrant.

Evaluation of the association between quantitative mammographic density and breast cancer occurred in different quadrants

BACKGROUND

To investigate the relationship between mammographic density measured in four quadrants of a breast with the location of the occurred cancer.

METHODS

One hundred and ten women diagnosed with unilateral breast cancer that could be determined in one specific breast quadrant were retrospectively studied. Women with previous cancer/breast surgery were excluded. The craniocaudal (CC) and mediolateral oblique (MLO) mammography of the contralateral normal breast were used to separate a breast into 4 quadrants: Upper-Outer (UO), Upper-Inner (UI), Lower-Outer (LO), and Lower-Inner (LI). The breast area (BA), dense area (DA), and percent density (PD) in each quadrant were measured by using the fuzzy-C-means segmentation. The BA, DA, and PD were compared between patients who had cancer occurring in different quadrants.

RESULTS

The upper-outer quadrant had the highest BA (37 ± 15 cm²) and DA (7.1 ± 2.9 cm²), with PD = 20.0 ± 5.8%. The order of BA and DA in the 4 separated quadrants were: UO > UI > LO > LI, and almost all pair-wise comparisons showed significant differences. For tumor location, 67 women (60.9%) had tumor in UO, 16 (14.5%) in UI, 7 (6.4%) in LO, and 20 (18.2%) in LI quadrant, respectively. The estimated odds and the 95% confidence limits of tumor development in the UO, UI, LO and LI quadrants were 1.56 (1.06, 2.29), 0.17 (0.10, 0.29), 0.07 (0.03, 0.15), and 0.22 (0.14, 0.36), respectively. In these 4 groups of women, the order of quadrant BA and DA were all the same (UO > UI > LO > LI), and there was no significant difference in BA, DA or PD among them (all p > 0.05).

CONCLUSIONS

Breast cancer was most likely to occur in the UO quadrant, which was also the quadrant with highest BA and DA; but for women with tumors in other quadrants, the density in that quadrant was not the highest. Therefore, there was no direct association between quadrant density and tumor occurrence.

A multi-resolution approach for spinal metastasis detection using deep Siamese neural networks

Spinal metastasis, a metastatic cancer of the spine, is the most common malignant disease in the spine. In this study, we investigate the feasibility of automated spinal metastasis detection in magnetic resonance imaging (MRI) by using deep learning methods. To accommodate the large variability in metastatic lesion sizes, we develop a Siamese deep neural network approach comprising three identical subnetworks for multi-resolution analysis and detection of spinal metastasis. At each location of interest, three image patches at three different resolutions are extracted and used as the input to the networks. To further reduce the false positives (FPs), we leverage the similarity between neighboring MRI slices, and adopt a weighted averaging strategy to aggregate the results obtained by the Siamese neural networks. The detection performance is evaluated on a set of 26 cases using a free-response receiver operating characteristic (FROC) analysis. The results show that the proposed approach correctly detects all the spinal metastatic lesions while producing only 0.40 FPs per case. At a true positive (TP) rate of 90%, the use of the aggregation reduces the FPs from 0.375 FPs per case to 0.207 FPs per case, a nearly 44.8% reduction. The results indicate that the proposed Siamese neural network method, combined with the aggregation strategy, provide a viable strategy for the automated detection of spinal metastasis in MRI images.

Morphological and dynamic contrast enhanced MR imaging features for the differentiation of chordoma and giant cell tumors in the Axial Skeleton

PURPOSE

To characterize the morphological and dynamic-contrast-enhanced (DCE) MRI features of chordoma and giant cell tumor (GCT) of bone occurring in the axial skeleton.

MATERIALS AND METHODS

A total of 13 patients with chordoma and 26 patients with GCT who received conventional T1, T2, and DCE-MRI on 3 Tesla MR scanners were retrospectively identified and analyzed. Two radiologists evaluated morphological features independently, including the lesion location, expansile bone changes, vertebral compression, presence of paraspinal soft tissue mass, fibrous septa, and the signal intensity on T1WI and T2WI. The inter-observer agreement was evaluated by kappa test. The DCE kinetics was measured to obtain the initial area under curve (IAUC) and the wash-out slope; also the two-compartmental pharmacokinetic model was applied to obtain K^trans and k_ep . The diagnostic accuracy was evaluated by CHAID decision tree and ROC analysis.

RESULTS

Chordomas were more likely to show soft tissue mass than GCTs (13/13 = 100% versus 15/26 = 58%; P = 0.007), as well as fibrous septa (9/13 = 69% versus 0; P < 0.001). In decision tree analysis, presence of fibrous septa and lesion location yield 31/39 = 79% accuracy. The DCE-MRI pharmacokinetic parameters K^trans and k_ep of GCTs were significantly higher than those of chordomas, 0.13 ± 0.65 versus 0.06 ± 0.04 (1/min) for K^trans , 0.62 ± 0.22 versus 0.17 ± 0.12 (1/min) for k_ep , P < 0.001 for both. If using kep = 0.43/min as the cut-off value, it achieved 100% sensitivity and 92% specificity to differentiate chordoma from GCT, with an overall accuracy of 37/39 = 95%. The IAUC was highly correlated with K^trans (r = 0.94), and the slope was highly correlated with k_ep (r = 0.95).

CONCLUSION

Several morphological features were significantly different between chordoma and GCT, but their diagnostic performance was inferior to that of DCE-MRI.

LEVEL OF EVIDENCE

4 J. Magn. Reson. Imaging 2017;45:1068-1075.

Resolution Improvement in Positron Emission Tomography Using Anatomical Magnetic Resonance Imaging

An ideal imaging system should provide information with high-sensitivity, high spatial, and temporal resolution. Unfortunately, it is not possible to satisfy all of these desired features in a single modality. In this paper, we discuss methods to improve the spatial resolution in positron emission imaging (PET) using a priori information from Magnetic Resonance Imaging (MRI). Our approach uses an image restoration algorithm based on the maximization of mutual information (MMI), which has found significant success for optimizing multimodal image registration. The MMI criterion is used to estimate the parameters in the Sharpness-Constrained Wiener filter. The generated filter is then applied to restore PET images of a realistic digital brain phantom. The resulting restored images show improved resolution and better signal-to-noise ratio compared to the interpolated PET images. We conclude that a Sharpness-Constrained Wiener filter having parameters optimized from a MMI criterion may be useful for restoring spatial resolution in PET based on a priori information from correlated MRI.

Coregistration of Dynamic Contrast Enhanced MRI and Broadband Diffuse Optical Spectroscopy for Characterizing Breast Cancer

A handheld scanning probe based on broadband Diffuse Optical Spectroscopy (DOS) was used in combination with dynamic contrast enhanced MRI (DCE-MRI) to quantitatively characterize locally-advanced breast cancers in six patients. Measurements were performed sequentially using external fiducial markers for co-registration. Tumor patterns were categorized according to MRI morphological data, and 3D DCE-MRI slices were converted into a volumetric matrix with isotropic voxels to generate views that coincided with the DOS scanning plane. Tumor volume and depth at each DOS measurement site were determined, and a tissue optical index (TOI) that reflects both angiogenic and stromal characteristics was derived from broadband DOS data.

In all six cases, optical scans showed significant TOI contrast corresponding to MRI morphological information. Sharp TOI peaks were recovered for well-circumscribed masses. A reduction in TOI was found inside a tumor with a necrotic center. A broadened peak was observed for a diffuse tumor pattern, and an inflammatory septal case provided two TOI peaks that correlated qualitatively with MRI enhancement. These results provide qualitative confirmation of the common signal origin and complementary information content that can be achieved by combining optical and MR imaging for breast cancer detection and clinical management.