Quantifying tumour heterogeneity in 18F-FDG PET/CT imaging by texture analysis

Intra-tumoural heterogeneity characterization through texture and colour analysis for differentiation of non-small cell lung carcinoma subtypes

Radiomics has shown potential in disease diagnosis, but its feasibility for non-small cell lung carcinoma (NSCLC) subtype classification is unclear. This study aims to explore the diagnosis value of texture and colour features from positron emission tomography computed tomography (PET-CT) images in differentiation of NSCLC subtypes: adenocarcinoma (ADC) and squamous cell carcinoma (SqCC). Two patient cohorts were retrospectively collected into a dataset of 341 ¹⁸F-labeled 2-deoxy-2fluoro-d-glucose ([¹⁸F] FDG) PET-CT images of NSCLC tumours (125 ADC, 174 SqCC, and 42 cases with unknown subtype). Quantification of texture and colour features was performed using freehand regions of interest. The relation between extracted features and commonly used parameters such as age, gender, tumour size, and standard uptake value (SUVmax) was explored. To classify NSCLC subtypes, support vector machine algorithm was applied on these features and the classification performance was evaluated by receiver operating characteristic curve analysis. There was a significant difference between ADC and SqCC subtypes in texture and colour features (P  <  0.05); this showed that imaging features were significantly correlated to both SUVmax and tumour diameter (P  <  0.05). When evaluating classification performance, features combining texture and colour showed an AUC of 0.89 (95% CI, 0.78-1.00), colour features showed an AUC of 0.85 (95% CI, 0.71-0.99), and texture features showed an AUC of 0.68 (95% CI, 0.48-0.88). DeLong's test showed that AUC was higher for features combining texture and colour than that for texture features only (P  =  0.010), but not significantly different from that for colour features only (P  =  0.328). HSV colour features showed a similar performance to RGB colour features (P  =  0.473). The colour features are promising in the refinement of NSCLC subtype differentiation, and features combining texture and colour of PET-CT images could result in better classification performance.

Radiomics analysis of pre-treatment [18F]FDG PET/CT for patients with metastatic colorectal cancer undergoing palliative systemic treatment

BACKGROUND

The aim of this study was to assess radiomics features on pre-treatment [¹⁸F]FDG positron emission tomography (PET) as potential biomarkers for response and survival in patients with metastatic colorectal cancer (mCRC).

METHODS

Patients with mCRC underwent [¹⁸F]FDG PET/computed tomography (CT) prior to first- or third-line palliative systemic treatment. Tumour lesions were semiautomatically delineated and standard uptake value (SUV), metabolically active tumour volume (MATV), total lesion glycolysis (TLG), entropy, area under the curve of the cumulative SUV-volume histogram (AUC-CSH), compactness and sphericity were obtained.

RESULTS

Lesions of 47 patients receiving third-line systemic treatment had higher SUV_max, SUV_peak, SUV_mean, MATV and TLG, and lower AUC-CSH, compactness and sphericity compared to 52 patients receiving first-line systemic treatment. Therefore, first- and third-line groups were evaluated separately. In the first-line group, anatomical changes on CT correlated negatively with TLG (ρ = 0.31) and MATV (ρ = 0.36), and positively with compactness (ρ = -0.27) and sphericity (ρ = -0.27). Patients without benefit had higher mean entropy (p = 0.021). Progression-free survival (PFS) and overall survival (OS) were worse with a decreased mean AUC [hazard ratio (HR) 0.86, HR 0.77] and increase in mean MATV (HR 1.15, HR 1.22), sum MATV (HR 1.14, HR 1.19), mean TLG (HR 1.16, HR 1.22) and sum TLG (HT1.12, HR1.18). In the third-line group, AUC-CSH correlated negatively with anatomical change (ρ = 0.21). PFS and OS were worse with an increased mean MATV (HR 1.27, HR 1.68), sum MATV (HR 1.35, HR 2.04), mean TLG (HR 1.29, HR 1.52) and sum TLG (HT 1.27, HR 1.80). SUV_max and SUV_peak negatively correlated with OS (HR 1.19, HR 1.21). Cluster analysis of the 10 radiomics features demonstrated no complementary value in identifying aggressively growing lesions or patients with impaired survival.

CONCLUSION

We demonstrated an association between improved clinical outcome and pre-treatment low tumour volume and heterogeneity as well as high sphericity on [¹⁸F]FDG PET. Future PET imaging research should include radiomics features that incorporate tumour volume and heterogeneity when correlating PET data with clinical outcome.

Could perfusion heterogeneity at dynamic contrast-enhanced MRI be used to predict rectal cancer sensitivity to chemoradiotherapy?

AIM

To evaluate whether perfusion heterogeneity of rectal cancer prior to chemoradiotherapy (CRT) using histogram analysis of dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) quantitative parameters can predict response to treatment.

MATERIALS AND METHODS

Twenty-one patients with histologically proven rectal adenocarcinoma were enrolled prospectively. All patients underwent 1.5 T DCE-MRI before CRT. Tumour volumes were drawn on Ktrans and Ve maps, using T2-weighted (W) images as reference, and the following first-order texture parameters of Ve and Ktrans values were extracted: 25th, 50th, 75th percentile, mean, standard deviation, skewness, and kurtosis. After CRT, patients underwent surgery and according with Rödel's tumour regression grade (TRG), they were classified as poor responders "non-GR" (TRG 0-2) and good responders "GR" (TRG 3-4). Differences between GR and non-GR in DCE-MRI first-order texture parameters were evaluated using the Mann-Whitney test, and their role in the prediction of response was investigated using receiver operating characteristic (ROC) curve analysis.

RESULTS

Sixteen (76%) patients were classified as GR and five (24%) were non-GR. Skewness and kurtosis of Ve was significantly higher in non-GR (4.886±1.320 and 36.402±24.486, respectively) than in GR patients (1.809±1.280, p=0.003 and 6.268±8.130, p= 0.011). Ve skewness <3.635 was able to predict GR with an area under the ROC curve (AUC) of 0.988, sensitivity 93.8%, specificity 80%, and accuracy 90.5%. Ve kurtosis <21.095 was able to predict response with an AUC of 0.963, sensitivity 93.8%, specificity 80%, and accuracy 90.5%. Other parameters were not different between groups or predictors of response.

CONCLUSION

Ve skewness and kurtosis seem to be promising in the prediction of response to CRT in rectal cancer patients.

Metabolic Tumor Volume Metrics in Lymphoma

Although visual assessment using the Deauville criteria is strongly recommended by guidelines for treatment response monitoring in all FDG-avid lymphoma histologies, the high rate of false-positives and concerns about interobserver variability have motivated the development of quantitative tools to facilitate objective measurement of tumor response in both routine and clinical trial settings. Imaging studies using functional quantitative measures play a significant role in profiling oncologic processes. These quantitative metrics allow for objective end points in multicenter clinical trials. However, the standardization of imaging procedures including image acquisition parameters, reconstruction and analytic measures, and validation of these methods are essential to enable an individualized treatment approach. A robust quality control program associated with the inclusion of proper scanner calibration, cross-calibration with dose calibrators and across other scanners is required for accurate quantitative measurements. In this section, we will review the technical and methodological considerations related to PET-derived quantitative metrics and the relevant published data to emphasize the potential value of these metrics in the prediction of patient prognosis in lymphoma.

Metabolic heterogeneity on baseline 18FDG-PET/CT scan is a predictor of outcome in primary mediastinal B-cell lymphoma

An important unmet need in the management of primary mediastinal B-cell lymphoma (PMBCL) is to identify the patients for whom first-line therapy will fail to intervene before the lymphoma becomes refractory. High heterogeneity of intratumoral 18F-fluorodeoxyglucose (18FDG) uptake distribution on positron emission tomography/computed tomography (PET/CT) scans has been suggested as a possible marker of chemoresistance in solid tumors. In the present study, we investigated the prognostic value of metabolic heterogeneity (MH) in 103 patients with PMBCL prospectively enrolled in the International Extranodal Lymphoma Study Group (IELSG) 26 study, aimed at clarifying the role of PET in this lymphoma subtype. MH was estimated using the area under curve of cumulative standardized uptake value-volume histogram (AUC-CSH) method. Progression-free survival at 5 years was 94% vs 73% in low- and high-MH groups, respectively (P = .0001). In a Cox model of progression-free survival including dichotomized MH, metabolic tumor volume, total lesion glycolysis (TLG), international prognostic index, and tumor bulk (mediastinal mass > 10 cm), as well as age as a continuous variable, only TLG (P < .001) and MH (P < .001) retained statistical significance. Using these 2 features to construct a simple prognostic model resulted in early and accurate (positive predictive value, 89%; negative predictive value, ≥90%) identification of patients at high risk for progression at a point that would allow the use of risk-adapted treatments. This may provide an important opportunity for the design of future trials aimed at helping the minority of patients who harbor chemorefractory PMBCL. The study is registered at ClinicalTrials.gov as NCT00944567.

Presurgical Identification of Uterine Smooth Muscle Malignancies through the Characteristic FDG Uptake Pattern on PET Scans

The unidentified presence of uterine smooth muscle malignancies poses a tremendous risk in women planning surgery for presumed benign leiomyomas. We sought to investigate whether preoperative FDG PET may be useful to identify leiomyosarcomas (LMS) and smooth muscle tumors of uncertain malignant potential (STUMP). Methods. We investigated patients with rapidly growing uterine masses which were suspected of being malignant on ultrasound or MRI. Among the 21 patients who underwent FDG PET, we identified 7 LMS, 1 STUMP, and 13 leiomyomas. PET-derived parameters and FDG uptake patterns were analyzed retrospectively. Results. The SUV_max values of LMS/STUMP (range: 3.7–11.8) were significantly higher than those observed in leiomyomas (range: 2.0–9.4; P=0.003) despite a significant overlap. The metabolic tumor/necrosis ratio was significantly higher in LMS/STUMP than in leiomyomas (P < 0.001), with no significant intergroup overlaps. All LMS/STUMP revealed a characteristic pattern of FDG uptake, identifying a specific “hollow ball” sign (corresponding to areas of coagulative tumor necrosis). In contrast, this sign was invariably absent in patients with leiomyomas. Conclusion. The characteristic FDG uptake pattern instead of SUV on PET images allows identifying LMS/STUMP in patients with rapidly growing uterine masses, avoiding the deleterious consequences of regular surgery for presumed benign leiomyomas.

Repeatability and Reproducibility of Radiomic Features: A Systematic Review

Purpose:

An ever-growing number of predictive models used to inform clinical decision making have included quantitative, computer-extracted imaging biomarkers, or “radiomic features.” Broadly generalizable validity of radiomics-assisted models may be impeded by concerns about reproducibility. We offer a qualitative synthesis of 41 studies that specifically investigated the repeatability and reproducibility of radiomic features, derived from a systematic review of published peer-reviewed literature.

Methods and Materials:

The PubMed electronic database was searched using combinations of the broad Haynes and Ingui filters along with a set of text words specific to cancer, radiomics (including texture analyses), reproducibility, and repeatability. This review has been reported in compliance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. From each full-text article, information was extracted regarding cancer type, class of radiomic feature examined, reporting quality of key processing steps, and statistical metric used to segregate stable features.

Results:

Among 624 unique records, 41 full-text articles were subjected to review. The studies primarily addressed non-small cell lung cancer and oropharyngeal cancer. Only 7 studies addressed in detail every methodologic aspect related to image acquisition, preprocessing, and feature extraction. The repeatability and reproducibility of radiomic features are sensitive at various degrees to processing details such as image acquisition settings, image reconstruction algorithm, digital image preprocessing, and software used to extract radiomic features. First-order features were overall more reproducible than shape metrics and textural features. Entropy was consistently reported as one of the most stable first-order features. There was no emergent consensus regarding either shape metrics or textural features; however, coarseness and contrast appeared among the least reproducible.

Conclusions:

Investigations of feature repeatability and reproducibility are currently limited to a small number of cancer types. Reporting quality could be improved regarding details of feature extraction software, digital image manipulation (preprocessing), and the cutoff value used to distinguish stable features.

We offer a qualitative synthesis of 41 studies that specifically investigated the repeatability and reproducibility of radiomic features. The repeatability and reproducibility of radiomic features are sensitive at various degrees to image quality and to software used to extract radiomic features. Investigations of feature repeatability and reproducibility are currently limited to a small number of cancer types. No consensus was found regarding the most repeatable and reproducible features with respect to different settings.

Quantitative biomarkers for liver metastases: comparison of MRI diffusion-weighted imaging heterogeneity index and fluorine-18-fluoro-deoxyglucose standardised uptake value in hybrid PET/MR

AIM

To investigate the ability of apparent diffusion coefficient (ADC) heterogeneity index to discriminate liver metastases (LM) from normal-appearing liver (NAL) tissue as compared to common magnetic resonance imaging (MRI) metrics, and to investigate its correlation with 2-[¹⁸F]-fluoro-2-deoxy-d-glucose (¹⁸F-FDG) positron-emission tomography (PET) standardised uptake value (SUV).

MATERIALS AND METHODS

Thirty-nine liver metastases in 24 oncology patients (13 women, 11 men; mean age 56±13 years) with proven LM from heterogeneous sources were evaluated on a PET/MRI system. Abdominal sequences included Dixon and diffusion-weighted imaging (DWI) protocols with simultaneous PET. Tissue heterogeneity was calculated using the coefficient of variance (CV) of the ADC, and compared in LM and in NAL tissue of the same volume in an adjacent portion of the liver. The correlations between various ADC measures and PET SUV in distinguishing LM from NAL were evaluated.

RESULTS

A good correlation was found between ADCcv and SUVpeak (r=0.712). Moderate inverse correlation was found between ADCmin and SUVpeak (r=-0.536), and a weak inverse correlation between ADCmean and SUVpeak (r=-0.273). There was a significant difference between LM and NAL when ADCcv (p<0.0001) and ADCmin (p=0.001) were used. Receiver operating characteristic (ROC) analysis of SUV, ADCcv, ADCmin, and ADCmean produced an AUC of 0.989, 0.900, 0.742, and 0.623 respectively.

CONCLUSIONS

The ADCcv index is a potential biomarker of LM with better correlation to ¹⁸F-FDG PET SUVpeak than conventional MRI metrics, and may serve to quantitatively discriminate between LM and NAL.

Positron emission tomography-based assessment of metabolic gradient and other prognostic features in sarcoma

Intratumoral heterogeneity biomarkers derived from positron emission tomography (PET) imaging with fluorodeoxyglucose (FDG) are of interest for a number of cancers, including sarcoma. A range of radiomic texture variables, adapted from general methodologies for image analysis, has shown promise in the setting. In the context of sarcoma, our group introduced an alternative model-based approach to the measurement of heterogeneity. In this approach, the heterogeneity of a tumor is characterized by the extent to which the 3-D FDG uptake pattern deviates from a simple elliptically contoured structure. By using a nonparametric analysis of the uptake profile obtained from this spatial model, a variable assessing the metabolic gradient of the tumor is developed. The work explores the prognostic potential of this new variable in the context of FDG-PET imaging of sarcoma. A mature clinical series involving 197 patients, 88 of whom have complete time-to-death information, is used. Texture variables based on the imaging data are also evaluated in this series and a range of appropriate machine learning methodologies are then used to explore the complementary prognostic roles for structure and texture variables. We conclude that both texture-based and model-based variables can be combined to achieve enhanced prognostic assessments of outcome for patients with sarcoma based on FDG-PET imaging information.

Pharmacometabolomics Informs Quantitative Radiomics for Glioblastoma Diagnostic Innovation

Applications of omics systems biology technologies have enormous promise for radiology and diagnostics in surgical fields. In this context, the emerging fields of radiomics (a systems scale approach to radiology using a host of technologies, including omics) and pharmacometabolomics (use of metabolomics for patient and disease stratification and guiding precision medicine) offer much synergy for diagnostic innovation in surgery, particularly in neurosurgery. This synthesis of omics fields and applications is timely because diagnostic accuracy in central nervous system tumors still challenges decision-making. Considering the vast heterogeneity in brain tumors, disease phenotypes, and interindividual variability in surgical and chemotherapy outcomes, we believe that diagnostic accuracy can be markedly improved by quantitative radiomics coupled to pharmacometabolomics and related health information technologies while optimizing economic costs of traditional diagnostics. In this expert review, we present an innovation analysis on a systems-level multi-omics approach toward diagnostic accuracy in central nervous system tumors. For this, we suggest that glioblastomas serve as a useful application paradigm. We performed a literature search on PubMed for articles published in English between 2006 and 2016. We used the search terms "radiomics," "glioblastoma," "biomarkers," "pharmacogenomics," "pharmacometabolomics," "pharmacometabonomics/pharmacometabolomics," "collaborative informatics," and "precision medicine." A list of the top 4 insights we derived from this literature analysis is presented in this study. For example, we found that (i) tumor grading needs to be better refined, (ii) diagnostic precision should be improved, (iii) standardization in radiomics is lacking, and (iv) quantitative radiomics needs to prove clinical implementation. We conclude with an interdisciplinary call to the metabolomics, pharmacy/pharmacology, radiology, and surgery communities that pharmacometabolomics coupled to information technologies (chemoinformatics tools, databases, collaborative systems) can inform quantitative radiomics, thus translating Big Data and information growth to knowledge growth, rational drug development and diagnostics innovation for glioblastomas, and possibly in other brain tumors.

Radiomics of CT Features May Be Nonreproducible and Redundant: Influence of CT Acquisition Parameters

Purpose To identify the reproducible and nonredundant radiomics features (RFs) for computed tomography (CT). Materials and Methods Two phantoms were used to test RF reproducibility by using test-retest analysis, by changing the CT acquisition parameters (hereafter, intra-CT analysis), and by comparing five different scanners with the same CT parameters (hereafter, inter-CT analysis). Reproducible RFs were selected by using the concordance correlation coefficient (as a measure of the agreement between variables) and the coefficient of variation (defined as the ratio of the standard deviation to the mean). Redundant features were grouped by using hierarchical cluster analysis. Results A total of 177 RFs including intensity, shape, and texture features were evaluated. The test-retest analysis showed that 91% (161 of 177) of the RFs were reproducible according to concordance correlation coefficient. Reproducibility of intra-CT RFs, based on coefficient of variation, ranged from 89.3% (151 of 177) to 43.1% (76 of 177) where the pitch factor and the reconstruction kernel were modified, respectively. Reproducibility of inter-CT RFs, based on coefficient of variation, also showed large material differences, from 85.3% (151 of 177; wood) to only 15.8% (28 of 177; polyurethane). Ten clusters were identified after the hierarchical cluster analysis and one RF per cluster was chosen as representative. Conclusion Many RFs were redundant and nonreproducible. If all the CT parameters are fixed except field of view, tube voltage, and milliamperage, then the information provided by the analyzed RFs can be summarized in only 10 RFs (each representing a cluster) because of redundancy.

Selected PET radiomic features remain the same

Purpose

We investigated whether PET radiomic features are affected by differences in the scanner, scan protocol, and lesion location using ¹⁸F-FDG PET/CT and PET/MR scans.

Results

SUV, TMR, skewness, kurtosis, entropy, and homogeneity strongly correlated between PET/CT and PET/MR images. SUVs were significantly higher on PET/MR_{0-2 min} and PET/MR_{0-10 min} than on PET/CT in gynecological cancer (p = 0.008 and 0.008, respectively), whereas no significant difference was observed between PET/CT, PET/MR_{0–2 min}, and PET/MR_{0–10 min} images in oral cavity/oropharyngeal cancer. TMRs on PET/CT, PET/MR_{0–2 min}, and PET/MR_{0–10 min} increased in this order in gynecological cancer and oral cavity/oropharyngeal cancer. In contrast to conventional and histogram indices, 4 textural features (entropy, homogeneity, SRE, and LRE) were not significantly different between PET/CT, PET/MR_{0–2 min}, and PET/MR_{0–10 min} images.

Conclusions

¹⁸F-FDG PET radiomic features strongly correlated between PET/CT and PET/MR images. Dixon-based attenuation correction on PET/MR images underestimated tumor tracer uptake more significantly in oral cavity/oropharyngeal cancer than in gynecological cancer. ¹⁸F-FDG PET textural features were affected less by differences in the scanner and scan protocol than conventional and histogram features, possibly due to the resampling process using a medium bin width.

Methods

Eight patients with gynecological cancer and 7 with oral cavity/oropharyngeal cancer underwent a whole-body ¹⁸F-FDG PET/CT scan and regional PET/MR scan in one day. PET/MR scans were performed for 10 minutes in the list mode, and PET/CT and 0–2 min and 0–10 min PET/MR images were reconstructed. The standardized uptake value (SUV), tumor-to-muscle SUV ratio (TMR), skewness, kurtosis, entropy, homogeneity, short-run emphasis (SRE), and long-run emphasis (LRE) were compared between PET/CT, PET/MR_{0-2 min}, and PET/MR_{0-10 min} images.

Texture analysis of diffusion weighted imaging for the evaluation of glioma heterogeneity based on different regions of interest

The present study aimed to explore the role of texture analysis with apparent diffusion coefficient (ADC) maps based on different regions of interest (ROI) in determining glioma grade. Thirty patients with glioma underwent diffusion-weighted imaging (DWI). ADC values were determined from the following three ROIs: i) whole tumor; ii) solid portion; and iii) peritumoral edema. Texture features were compared between high-grade gliomas (HGGs) and low-grade gliomas (LGGs) using the non-parametric Wilcoxon rank-sum test or the unpaired Student's t-test. Receiver operating characteristic (ROC) curves were constructed to determine the optimum threshold for inhomogeneity values in discrimination of HGGs from LGGs. With a spearman rank correlation model, the aforementioned ADC inhomogeneity values were correlated with the Ki-67 labeling index. With whole tumor ROI, inhomogeneity values proved to be significantly different between HGGs and LGGs (P<0.001). With solid portion ROI, inhomogeneity and median values showed significant difference between HGGs and LGGs (P=0.001 and P=0.043, respectively). With peritumoral edema ROI, entropy and edema volume demonstrated positive results (P=0.016, P<0.001). The whole tumor inhomogeneity parameter performed with better diagnostic accuracy (P=0.048) than selecting the solid portion ROI. The association between inhomogeneity and Ki-67 labeling index was significantly positive in whole tumor and solid portion ROI (R=0.628, P<0.001 and R=0.470, P=0.009). Texture analysis of DWI based on different ROI can provide various significant parameters to evaluate tumor heterogeneity, which were correlated with tumor grade. Particularly, the inhomogeneity value derived from whole tumor ROI provided high diagnostic value and predicting the status of tumor proliferation.

Pathologic stratification of operable lung adenocarcinoma using radiomics features extracted from dual energy CT images

PURPOSE

To evaluate the usefulness of surrogate biomarkers as predictors of histopathologic tumor grade and aggressiveness using radiomics data from dual-energy computed tomography (DECT), with the ultimate goal of accomplishing stratification of early-stage lung adenocarcinoma for optimal treatment.

RESULTS

Pathologic grade was divided into grades 1, 2, and 3. Multinomial logistic regression analysis revealed i-uniformity and 97.5th percentile CT attenuation value as independent significant factors to stratify grade 2 or 3 from grade 1. The AUC value calculated from leave-one-out cross-validation procedure for discriminating grades 1, 2, and 3 was 0.9307 (95% CI: 0.8514-1), 0.8610 (95% CI: 0.7547-0.9672), and 0.8394 (95% CI: 0.7045-0.9743), respectively.

MATERIALS AND METHODS

A total of 80 patients with 91 clinically and radiologically suspected stage I or II lung adenocarcinoma were prospectively enrolled. All patients underwent DECT and F-18-fluorodeoxyglucose (FDG) positron emission tomography (PET)/CT, followed by surgery. Quantitative CT and PET imaging characteristics were evaluated using a radiomics approach. Significant features for a tumor aggressiveness prediction model were extracted and used to calculate diagnostic performance for predicting all pathologic grades.

CONCLUSIONS

Quantitative radiomics values from DECT imaging metrics can help predict pathologic aggressiveness of lung adenocarcinoma.

Clinical Application of Biparametric MRI Texture Analysis for Detection and Evaluation of High-Grade Prostate Cancer in Zone-Specific Regions

OBJECTIVE

The purpose of this study was to investigate the performance of biparametric MRI texture analysis (TA) in detecting and evaluating high-grade prostate cancer in zone-specific regions.

MATERIALS AND METHODS

A retrospective study included 184 consecutively registered biopsy-naive patients in whom prostate cancer was suspected who were undergoing multiparametric prostate MRI. MR images were scored and evaluated by two readers using the Prostate Imaging Reporting and Data System version 2 (PI-RADSv2) and biparametric MRI TA in separate sessions. Interobserver agreement on PI-RADSv2 score and textural parameters of biparametric MRI was evaluated. The logistic regression model based on TA was built for different zones of the prostate. ROC analysis was used to compare the TA-based model with other parameters alone. The correlation of each parameter with Gleason score of high-grade prostate cancer was also assessed.

RESULTS

Reader reliability ranged from moderate to good for PI-RADSv2 (Cohen κ = 0.525-0.616) and from good to excellent for textural metrics (intraclass correlation coefficient, 0.745-0.925). Diagnostic performance was significantly improved by use of the TA-based model (transition zone AUC, 0.87; peripheral zone AUC, 0.89) in comparison with PI-RADSv2 and other texture parameters alone. For the transition zone, entropy had moderate to good correlation with the Gleason score of high-grade prostate cancer (r = 0.562, p = 0.004). In the peripheral zone, entropy (r = 0.614, p = 0.003) and inertia (r = 0.663, p = 0.002) had moderate to good correlations with Gleason score.

CONCLUSION

The results of this clinical study indicate that a TA-based model that includes biparametric MRI can be used for identifying high-grade prostate cancer and that specific parameters extracted from TA may be additional tools for assessing tumor aggressiveness.

Unsupervised consensus cluster analysis of [18F]-fluoroethyl-L-tyrosine positron emission tomography identified textural features for the diagnosis of pseudoprogression in high-grade glioma

Rationale

Timely detection of pseudoprogression (PSP) is crucial for the management of patients with high-grade glioma (HGG) but remains difficult. Textural features of O-(2-[¹⁸F]fluoroethyl)-L-tyrosine positron emission tomography (FET-PET) mirror tumor uptake heterogeneity; some of them may be associated with tumor progression.

Methods

Fourteen patients with HGG and suspected of PSP underwent FET-PET imaging. A set of 19 conventional and textural FET-PET features were evaluated and subjected to unsupervised consensus clustering. The final diagnosis of true progression vs. PSP was based on follow-up MRI using RANO criteria.

Results

Three robust clusters have been identified based on 10 predominantly textural FET-PET features. None of the patients with PSP fell into cluster 2, which was associated with high values for textural FET-PET markers of uptake heterogeneity. Three out of 4 patients with PSP were assigned to cluster 3 that was largely associated with low values of textural FET-PET features. By comparison, tumor-to-normal brain ratio (TNRmax) at the optimal cutoff 2.1 was less predictive of PSP (negative predictive value 57% for detecting true progression, p=0.07 vs. 75% with cluster 3, p=0.04).

Principal Conclusions

Clustering based on textural O-(2-[18F]fluoroethyl)-L-tyrosine PET features may provide valuable information in assessing the elusive phenomenon of pseudoprogression.

A Novel Framework for Automated Segmentation and Labeling of Homogeneous Versus Heterogeneous Lung Tumors in [18F]FDG-PET Imaging

PURPOSE

Determination of intra-tumor high-uptake area using 2-deoxy-2-[¹⁸F]fluoro-D-glucose ([¹⁸F]FDG) positron emission tomography (PET) imaging is an important consideration for dose painting in radiation treatment applications. The aim of our study was to develop a framework towards automated segmentation and labeling of homogeneous vs. heterogeneous tumors in clinical lung [¹⁸F]FDG-PET with the capability of intra-tumor high-uptake region delineation.

PROCEDURES

We utilized and extended a fuzzy random walk PET tumor segmentation algorithm to delineate intra-tumor high-uptake areas. Tumor textural feature (TF) analysis was used to find a relationship between tumor type and TF values. Segmentation accuracy was evaluated quantitatively utilizing 70 clinical [¹⁸F]FDG-PET lung images of patients with a total of 150 solid tumors. For volumetric analysis, the Dice similarity coefficient (DSC) and Hausdorff distance (HD) measures were extracted with respect to gold-standard manual segmentation. A multi-linear regression model was also proposed for automated tumor labeling based on TFs, including cross-validation analysis.

RESULTS

Two-tailed t test analysis of TFs between homogeneous and heterogeneous tumors revealed significant statistical difference for size-zone variability (SZV), intensity variability (IV), zone percentage (ZP), proposed parameters II and III, entropy and tumor volume (p < 0.001), dissimilarity, high intensity emphasis (HIE), and SUV_min (p < 0.01). Lower statistical differences were observed for proposed parameter I (p = 0.02), and no significant differences were observed for SUV_max and SUV_mean. Furthermore, the Spearman rank analysis between visual tumor labeling and TF analysis depicted a significant correlation for SZV, IV, entropy, parameters II and III, and tumor volume (0.68 ≤ ρ ≤ 0.84) and moderate correlation for ZP, HIE, homogeneity, dissimilarity, parameter I, and SUV_min (0.22 ≤ ρ ≤ 0.52), while no correlations were observed for SUV_max and SUV_mean (ρ < 0.08). The multi-linear regression model for automated tumor labeling process resulted in R ² and RMSE values of 0.93 and 0.14, respectively (p < 0.001), and generated tumor labeling sensitivity and specificity of 0.93 and 0.89. With respect to baseline random walk segmentation, the results showed significant (p < 0.001) mean DSC, HD, and SUV_mean error improvements of 21.4 ± 11.5 %, 1.4 ± 0.8 mm, and 16.8 ± 8.1 % in homogeneous tumors and 7.4 ± 4.4 %, 1.5 ± 0.6 mm, and 7.9 ± 2.7 % in heterogeneous lesions. In addition, significant (p < 0.001) mean DSC, HD, and SUV_mean error improvements were observed for tumor sub-volume delineations, namely 5 ± 2 %, 1.5 ± 0.6 mm, and 7 ± 3 % for the proposed Fuzzy RW method compared to RW segmentation.

CONCLUSION

We proposed and demonstrated an automatic framework for significantly improved segmentation and labeling of homogeneous vs. heterogeneous tumors in lung [¹⁸F]FDG-PET images.

Survival prediction in patients undergoing radionuclide therapy based on intratumoral somatostatin-receptor heterogeneity

The NETTER-1 trial demonstrated significantly improved progression-free survival (PFS) for peptide receptor radionuclide therapy (PRRT) in neuroendocrine tumors (NET) emphasizing the high demand for response prediction in appropriate candidates. In this multicenter study, we aimed to elucidate the prognostic value of tumor heterogeneity as assessed by somatostatin receptor (SSTR)-PET/CT. 141 patients with SSTR-expressing tumors were analyzed obtaining SSTR-PET/CT before PRRT (1-6 cycles, ¹⁷⁷Lu somatostatin analog). Using the Interview Fusion Workstation (Mediso), a total of 872 metastases were manually segmented. Conventional PET parameters as well as textural features representing intratumoral heterogeneity were computed. The prognostic ability for PFS and overall survival (OS) were examined. After performing Cox regression, independent parameters were determined by ROC analysis to obtain cut-off values to be used for Kaplan-Meier analysis. Within follow-up (median, 43.1 months), 75 patients showed disease progression (median, 22.2 m) and 54 patients died (median, 27.6 m). Cox analysis identified 8 statistically independent heterogeneity parameters for time-to-progression and time-to-death. Among them, the textural feature Entropy predicted both PFS and OS. Conventional PET parameters failed in response prediction. Imaging-based heterogeneity assessment provides prognostic information in PRRT candidates and outperformed conventional PET parameters. Its implementation in clinical practice can pave the way for individualized patient management.

Diagnostic and prognostic value of amyloid PET textural and shape features: comparison with classical semi-quantitative rating in 760 patients from the ADNI-2 database

We evaluated the performance of amyloid PET textural and shape features in discriminating normal and Alzheimer's disease (AD) subjects, and in predicting conversion to AD in subjects with mild cognitive impairment (MCI) or significant memory concern (SMC). Subjects from the Alzheimer's Disease Neuroimaging Initiative with available baseline ¹⁸F-florbetapir and T1-MRI scans were included. The cross-sectional cohort consisted of 181 controls and 148 AD subjects. The longitudinal cohort consisted of 431 SMC/MCI subjects, 85 of whom converted to AD during follow-up. PET images were normalized to MNI space and post-processed using in-house software. Relative retention indices (SUVr) were computed with respect to pontine, cerebellar, and composite reference regions. Several textural and shape features were extracted then combined using a support vector machine (SVM) to build a predictive model of AD conversion. Diagnostic and prognostic performance was evaluated using ROC analysis and survival analysis with the Cox proportional hazard model. The three SUVr and all the tested features effectively discriminated AD subjects in cross-sectional analysis (all p < 0.001). In longitudinal analysis, the variables with the highest prognostic value were composite SUVr (AUC 0.86; accuracy 81%), skewness (0.87; 83%), local minima (0.85; 79%), Geary's index (0.86; 81%), gradient norm maximal argument (0.83; 82%), and the SVM model (0.91; 86%). The adjusted hazard ratio for AD conversion was 5.5 for the SVM model, compared with 4.0, 2.6, and 3.8 for cerebellar, pontine and composite SUVr (all p < 0.001), indicating that appropriate amyloid textural and shape features predict conversion to AD with at least as good accuracy as classical SUVr.

Radiogenomics of hepatocellular carcinoma: multiregion analysis-based identification of prognostic imaging biomarkers by integrating gene data-a preliminary study

Our objective was to identify prognostic imaging biomarkers for hepatocellular carcinoma in contrast-enhanced computed tomography (CECT) with biological interpretations by associating imaging features and gene modules. We retrospectively analyzed 371 patients who had gene expression profiles. For the 38 patients with CECT imaging data, automatic intra-tumor partitioning was performed, resulting in three spatially distinct subregions. We extracted a total of 37 quantitative imaging features describing intensity, geometry, and texture from each subregion. Imaging features were selected after robustness and redundancy analysis. Gene modules acquired from clustering were chosen for their prognostic significance. By constructing an association map between imaging features and gene modules with Spearman rank correlations, the imaging features that significantly correlated with gene modules were obtained. These features were evaluated with Cox's proportional hazard models and Kaplan-Meier estimates to determine their prognostic capabilities for overall survival (OS). Eight imaging features were significantly correlated with prognostic gene modules, and two of them were associated with OS. Among these, the geometry feature volume fraction of the subregion, which was significantly correlated with all prognostic gene modules representing cancer-related interpretation, was predictive of OS (Cox p  =  0.022, hazard ratio  =  0.24). The texture feature cluster prominence in the subregion, which was correlated with the prognostic gene module representing lipid metabolism and complement activation, also had the ability to predict OS (Cox p  =  0.021, hazard ratio  =  0.17). Imaging features depicting the volume fraction and textural heterogeneity in subregions have the potential to be predictors of OS with interpretable biological meaning.

Translational Radiomics: Defining the Strategy Pipeline and Considerations for Application-Part 2: From Clinical Implementation to Enterprise

Enterprise imaging has channeled various technological innovations to the field of clinical radiology, ranging from advanced imaging equipment and postacquisition iterative reconstruction tools to image analysis and computer-aided detection tools. More recently, the advancement in the field of quantitative image analysis coupled with machine learning-based data analytics, classification, and integration has ushered in the era of radiomics, a paradigm shift that holds tremendous potential in clinical decision support as well as drug discovery. However, there are important issues to consider to incorporate radiomics into a clinically applicable system and a commercially viable solution. In this two-part series, we offer insights into the development of the translational pipeline for radiomics from methodology to clinical implementation (Part 1) and from that point to enterprise development (Part 2). In Part 2 of this two-part series, we study the components of the strategy pipeline, from clinical implementation to building enterprise solutions.

Diagnosis of Distant Metastasis of Lung Cancer: Based on Clinical and Radiomic Features

OBJECTIVES

To analyze the distant metastasis possibility based on computed tomography (CT) radiomic features in patients with lung cancer.

METHODS

This was a retrospective analysis of 348 patients with lung cancer enrolled between 2014 and February 2015. A feature set containing clinical features and 485 radiomic features was extracted from the pretherapy CT images. Feature selection via concave minimization (FSV) was used to select effective features. A support vector machine (SVM) was used to evaluate the predictive ability of each feature.

RESULTS

Four radiomic features and three clinical features were obtained by FSV feature selection. Classification accuracy by the proposed SVM with SGD method was 71.02%, and the area under the curve was 72.84% with only the radiomic features extracted from CT. After the addition of clinical features, 89.09% can be achieved.

CONCLUSION

The radiomic features of the pretherapy CT images may be used as predictors of distant metastasis. And it also can be used in combination with the patient's gender and tumor T and N phase information to diagnose the possibility of distant metastasis in lung cancer.

Intratumoral heterogeneity characterized by pretreatment PET in non-small cell lung cancer patients predicts progression-free survival on EGFR tyrosine kinase inhibitor

Intratumoral heterogeneity has been suggested to be an important resistance mechanism leading to treatment failure. We hypothesized that radiologic images could be an alternative method for identification of tumor heterogeneity. We tested heterogeneity textural parameters on pretreatment FDG-PET/CT in order to assess the predictive value of target therapy. Recurred or metastatic non-small cell lung cancer (NSCLC) subjects with an activating EGFR mutation treated with either gefitinib or erlotinib were reviewed. An exploratory data set (n = 161) and a validation data set (n = 21) were evaluated, and eight parameters were selected for survival analysis. The optimal cutoff value was determined by the recursive partitioning method, and the predictive value was calculated using Harrell’s C-index. Univariate analysis revealed that all eight parameters showed an increased hazard ratio (HR) for progression-free survival (PFS). The highest HR was 6.41 (P<0.01) with co-occurrence (Co) entropy. Increased risk remained present after adjusting for initial stage, performance status (PS), and metabolic volume (MV) (aHR: 4.86, P<0.01). Textural parameters were found to have an incremental predictive value of early EGFR tyrosine kinase inhibitor (TKI) failure compared to that of the base model of the stage and PS (C-index 0.596 vs. 0.662, P = 0.02, by Co entropy). Heterogeneity textural parameters acquired from pretreatment FDG-PET/CT are highly predictive factors for PFS of EGFR TKI in EGFR-mutated NSCLC patients. These parameters are easily applicable to the identification of a subpopulation at increased risk of early EGFR TKI failure. Correlation to genomic alteration should be determined in future studies.

Radiomics approach for preoperative identification of stages III and IIIIV of esophageal cancer

Objective

To predict preoperative staging using a radiomics approach based on computed tomography (CT) images of patients with esophageal squamous cell carcinoma (ESCC).

Methods

This retrospective study included 154 patients (primary cohort: n=114; validation cohort: n=40) with pathologically confirmed ESCC. All patients underwent a preoperative CT scan from the neck to abdomen. High throughput and quantitative radiomics features were extracted from the CT images for each patient. A radiomics signature was constructed using the least absolute shrinkage and selection operator (Lasso). Associations between radiomics signature, tumor volume and ESCC staging were explored. Diagnostic performance of radiomics approach and tumor volume for discriminating between stages I-II and III-IV was evaluated and compared using the receiver operating characteristics (ROC) curves and net reclassification improvement (NRI).

Results

A total of 9,790 radiomics features were extracted. Ten features were selected to build a radiomics signature after feature dimension reduction. The radiomics signature was significantly associated with ESCC staging (P<0.001), and yielded a better performance for discrimination of early and advanced stage ESCC compared to tumor volume in both the primary [area under the receiver operating characteristic curve (AUC): 0.795vs. 0.694, P=0.003; NRI=0.424)] and validation cohorts (AUC: 0.762 vs. 0.624, P=0.035; NRI=0.834).

Conclusions

The quantitative approach has the potential to identify stage I-II and III-IV ESCC before treatment.

Towards precision medicine: from quantitative imaging to radiomics

Radiology (imaging) and imaging-guided interventions, which provide multi-parametric morphologic and functional information, are playing an increasingly significant role in precision medicine. Radiologists are trained to understand the imaging phenotypes, transcribe those observations (phenotypes) to correlate with underlying diseases and to characterize the images. However, in order to understand and characterize the molecular phenotype (to obtain genomic information) of solid heterogeneous tumours, the advanced sequencing of those tissues using biopsy is required. Thus, radiologists image the tissues from various views and angles in order to have the complete image phenotypes, thereby acquiring a huge amount of data. Deriving meaningful details from all these radiological data becomes challenging and raises the big data issues. Therefore, interest in the application of radiomics has been growing in recent years as it has the potential to provide significant interpretive and predictive information for decision support. Radiomics is a combination of conventional computer-aided diagnosis, deep learning methods, and human skills, and thus can be used for quantitative characterization of tumour phenotypes. This paper discusses the overview of radiomics workflow, the results of various radiomics-based studies conducted using various radiological images such as computed tomography (CT), magnetic resonance imaging (MRI), and positron-emission tomography (PET), the challenges we are facing, and the potential contribution of radiomics towards precision medicine.

Investigating the Robustness Neighborhood Gray Tone Difference Matrix and Gray Level Co-occurrence Matrix Radiomic Features on Clinical Computed Tomography Systems Using Anthropomorphic Phantoms: Evidence From a Multivendor Study

OBJECTIVE

The aim of this study was to determine if optimized imaging protocols across multiple computed tomography (CT) vendors could result in reproducible radiomic features calculated from an anthropomorphic phantom.

METHODS

Materials with varying degrees of heterogeneity were placed throughout the lungs of the phantom. Twenty scans of the phantom were acquired on 3 CT manufacturers with chest CT protocols that had optimized protocol parameters. Scans were reconstructed using vendor-specific standards and lung kernels. The concordance correlation coefficient (CCC) was used to calculate reproducibility between features. For features with high CCC values, Bland-Altman analysis was also used to quantify agreement.

RESULTS

The mean Hounsfield unit (HU) was 32.93 HU (141.7 to -26.5 HU) for the rubber insert and 347.2 HU (-320.9 to -347.7 HU) for the wood insert. Low CCC values of less than 0.9 were calculated for all features across all scans.

CONCLUSIONS

Radiomic features that are derived from the spatial distribution of voxel intensities should be particularly scrutinized for reproducibility in a multivendor environment.

Comparison of machine learning methods for classifying mediastinal lymph node metastasis of non-small cell lung cancer from 18F-FDG PET/CT images

BACKGROUND

This study aimed to compare one state-of-the-art deep learning method and four classical machine learning methods for classifying mediastinal lymph node metastasis of non-small cell lung cancer (NSCLC) from 18F-FDG PET/CT images. Another objective was to compare the discriminative power of the recently popular PET/CT texture features with the widely used diagnostic features such as tumor size, CT value, SUV, image contrast, and intensity standard deviation. The four classical machine learning methods included random forests, support vector machines, adaptive boosting, and artificial neural network. The deep learning method was the convolutional neural networks (CNN). The five methods were evaluated using 1397 lymph nodes collected from PET/CT images of 168 patients, with corresponding pathology analysis results as gold standard. The comparison was conducted using 10 times 10-fold cross-validation based on the criterion of sensitivity, specificity, accuracy (ACC), and area under the ROC curve (AUC). For each classical method, different input features were compared to select the optimal feature set. Based on the optimal feature set, the classical methods were compared with CNN, as well as with human doctors from our institute.

RESULTS

For the classical methods, the diagnostic features resulted in 81~85% ACC and 0.87~0.92 AUC, which were significantly higher than the results of texture features. CNN's sensitivity, specificity, ACC, and AUC were 84, 88, 86, and 0.91, respectively. There was no significant difference between the results of CNN and the best classical method. The sensitivity, specificity, and ACC of human doctors were 73, 90, and 82, respectively. All the five machine learning methods had higher sensitivities but lower specificities than human doctors.

CONCLUSIONS

The present study shows that the performance of CNN is not significantly different from the best classical methods and human doctors for classifying mediastinal lymph node metastasis of NSCLC from PET/CT images. Because CNN does not need tumor segmentation or feature calculation, it is more convenient and more objective than the classical methods. However, CNN does not make use of the import diagnostic features, which have been proved more discriminative than the texture features for classifying small-sized lymph nodes. Therefore, incorporating the diagnostic features into CNN is a promising direction for future research.

The effect of post-injection 18F-FDG PET scanning time on texture analysis of peripheral nerve sheath tumours in neurofibromatosis-1

BACKGROUND

Texture features are being increasingly evaluated in 18F-fluorodeoxyglucose positron emission tomography (18F-FDG PET) as adjunctive imaging biomarkers in a number of different cancers. Whilst studies have reported repeatability between scans, there have been no studies that have specifically investigated the effect that the time of acquisition post-injection of 18F-FDG has on texture features. The aim of this study was to investigate if texture features change between scans performed at different time points post-injection.

RESULTS

Fifty-four patients (30 male, 24 female, mean age 35.1 years) with neurofibromatosis-1 and suspected malignant transformation of a neurofibroma underwent 18F-FDG PET/computed tomography (CT) scans at 101.5 ± 15.0 and 251.7 ± 18.4 min post-injection of 350 MBq 18F-FDG to a standard clinical protocol. Following tumour segmentation on both early and late scans, first- (n = 37), second- (n = 25) and high-order (n = 31) statistical features, as well as fractal texture features (n = 6), were calculated and a comparison was made between the early and late scans for each feature. Of the 54 tumours, 30 were benign and 24 malignant on histological analysis or on clinical follow-up for at least 5 years. Overall, 25/37 first-order, 9/25 second-order, 13/31 high-order and 3/6 fractal features changed significantly (p < 0.05) between early and late scans. The corresponding proportions for the 30 benign tumours alone were 22/37, 7/25, 8/31 and 2/6 and for the 24 malignant tumours, 11/37, 6/25, 8/31 and 0/6, respectively.

CONCLUSIONS

Several texture features change with time post-injection of 18F-FDG. Thus, when comparing texture features in intra- and inter-patient studies, it is essential that scans are obtained at a consistent time post-injection of 18F-FDG.

The effects of segmentation algorithms on the measurement of 18F-FDG PET texture parameters in non-small cell lung cancer

BACKGROUND

Measures of tumour heterogeneity derived from 18-fluoro-2-deoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) scans are increasingly reported as potential biomarkers of non-small cell lung cancer (NSCLC) for classification and prognostication. Several segmentation algorithms have been used to delineate tumours, but their effects on the reproducibility and predictive and prognostic capability of derived parameters have not been evaluated. The purpose of our study was to retrospectively compare various segmentation algorithms in terms of inter-observer reproducibility and prognostic capability of texture parameters derived from non-small cell lung cancer (NSCLC) 18F-FDG PET/CT images. Fifty three NSCLC patients (mean age 65.8 years; 31 males) underwent pre-chemoradiotherapy 18F-FDG PET/CT scans. Three readers segmented tumours using freehand (FH), 40% of maximum intensity threshold (40P), and fuzzy locally adaptive Bayesian (FLAB) algorithms. Intraclass correlation coefficient (ICC) was used to measure the inter-observer variability of the texture features derived by the three segmentation algorithms. Univariate cox regression was used on 12 commonly reported texture features to predict overall survival (OS) for each segmentation algorithm. Model quality was compared across segmentation algorithms using Akaike information criterion (AIC).

RESULTS

40P was the most reproducible algorithm (median ICC 0.9; interquartile range [IQR] 0.85-0.92) compared with FLAB (median ICC 0.83; IQR 0.77-0.86) and FH (median ICC 0.77; IQR 0.7-0.85). On univariate cox regression analysis, 40P found 2 out of 12 variables, i.e. first-order entropy and grey-level co-occurence matrix (GLCM) entropy, to be significantly associated with OS; FH and FLAB found 1, i.e., first-order entropy. For each tested variable, survival models for all three segmentation algorithms were of similar quality, exhibiting comparable AIC values with overlapping 95% CIs.

CONCLUSIONS

Compared with both FLAB and FH, segmentation with 40P yields superior inter-observer reproducibility of texture features. Survival models generated by all three segmentation algorithms are of at least equivalent utility. Our findings suggest that a segmentation algorithm using a 40% of maximum threshold is acceptable for texture analysis of 18F-FDG PET in NSCLC.

Clinical applications of textural analysis in non-small cell lung cancer

Lung cancer is the leading cause of cancer mortality worldwide. Treatment pathways include regular cross-sectional imaging, generating large data sets which present intriguing possibilities for exploitation beyond standard visual interpretation. This additional data mining has been termed "radiomics" and includes semantic and agnostic approaches. Textural analysis (TA) is an example of the latter, and uses a range of mathematically derived features to describe an image or region of an image. Often TA is used to describe a suspected or known tumour. TA is an attractive tool as large existing image sets can be submitted to diverse techniques for data processing, presentation, interpretation and hypothesis testing with annotated clinical outcomes. There is a growing anthology of published data using different TA techniques to differentiate between benign and malignant lung nodules, differentiate tissue subtypes of lung cancer, prognosticate and predict outcome and treatment response, as well as predict treatment side effects and potentially aid radiotherapy planning. The aim of this systematic review is to summarize the current published data and understand the potential future role of TA in managing lung cancer.

Screening and Biosensor-Based Approaches for Lung Cancer Detection

Early diagnosis of lung cancer helps to reduce the cancer death rate significantly. Over the years, investigators worldwide have extensively investigated many screening modalities for lung cancer detection, including computerized tomography, chest X-ray, positron emission tomography, sputum cytology, magnetic resonance imaging and biopsy. However, these techniques are not suitable for patients with other pathologies. Developing a rapid and sensitive technique for early diagnosis of lung cancer is urgently needed. Biosensor-based techniques have been recently recommended as a rapid and cost-effective tool for early diagnosis of lung tumor markers. This paper reviews the recent development in screening and biosensor-based techniques for early lung cancer detection.

Breast-Lesion Characterization using Textural Features of Quantitative Ultrasound Parametric Maps

This study evaluated, for the first time, the efficacy of quantitative ultrasound (QUS) spectral parametric maps in conjunction with texture-analysis techniques to differentiate non-invasively benign versus malignant breast lesions. Ultrasound B-mode images and radiofrequency data were acquired from 78 patients with suspicious breast lesions. QUS spectral-analysis techniques were performed on radiofrequency data to generate parametric maps of mid-band fit, spectral slope, spectral intercept, spacing among scatterers, average scatterer diameter, and average acoustic concentration. Texture-analysis techniques were applied to determine imaging biomarkers consisting of mean, contrast, correlation, energy and homogeneity features of parametric maps. These biomarkers were utilized to classify benign versus malignant lesions with leave-one-patient-out cross-validation. Results were compared to histopathology findings from biopsy specimens and radiology reports on MR images to evaluate the accuracy of technique. Among the biomarkers investigated, one mean-value parameter and 14 textural features demonstrated statistically significant differences (p < 0.05) between the two lesion types. A hybrid biomarker developed using a stepwise feature selection method could classify the legions with a sensitivity of 96%, a specificity of 84%, and an AUC of 0.97. Findings from this study pave the way towards adapting novel QUS-based frameworks for breast cancer screening and rapid diagnosis in clinic.

Unsupervised Learning in PET Radiomics

In this study, we investigated large scale radoimics on 116 breast cancer patients. We are particularly interested in unsupervised learning to bicluster patients and features in order to associate such biclusters with the disease characteristics. The results show that radiomics features with wavelet features have a better biclustering ability. And 172 radiomics features have shown a better classification capability.

Characterisation of malignant peripheral nerve sheath tumours in neurofibromatosis-1 using heterogeneity analysis of 18F-FDG PET

PURPOSE

Measurement of heterogeneity in 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography (PET) images is reported to improve tumour phenotyping and response assessment in a number of cancers. We aimed to determine whether measurements of 18F-FDG heterogeneity could improve differentiation of benign symptomatic neurofibromas from malignant peripheral nerve sheath tumours (MPNSTs).

METHODS

18F-FDG PET data from a cohort of 54 patients (24 female, 30 male, mean age 35.1 years) with neurofibromatosis-1 (NF1), and clinically suspected malignant transformation of neurofibromas into MPNSTs, were included. Scans were performed to a standard clinical protocol at 1.5 and 4 h post-injection. Six first-order [including three standardised uptake value (SUV) parameters], four second-order (derived from grey-level co-occurrence matrices) and four high-order (derived from neighbourhood grey-tone difference matrices) statistical features were calculated from tumour volumes of interest. Each patient had histological verification or at least 5 years clinical follow-up as the reference standard with regards to the characterisation of tumours as benign (n = 30) or malignant (n = 24).

RESULTS

There was a significant difference between benign and malignant tumours for all six first-order parameters (at 1.5 and 4 h; p < 0.0001), for second-order entropy (only at 4 h) and for all high-order features (at 1.5 h and 4 h, except contrast at 4 h; p < 0.0001-0.047). Similarly, the area under the receiver operating characteristic curves was high (0.669-0.997, p < 0.05) for the same features as well as 1.5-h second-order entropy. No first-, second- or high-order feature performed better than maximum SUV (SUVmax) at differentiating benign from malignant tumours.

CONCLUSIONS

18F-FDG uptake in MPNSTs is higher than benign symptomatic neurofibromas, as defined by SUV parameters, and more heterogeneous, as defined by first- and high-order heterogeneity parameters. However, heterogeneity analysis does not improve on SUVmax discriminative performance.

Classifying the evolutionary and ecological features of neoplasms

Neoplasms change over time through a process of cell-level evolution, driven by genetic and epigenetic alterations. However, the ecology of the microenvironment of a neoplastic cell determines which changes provide adaptive benefits. There is widespread recognition of the importance of these evolutionary and ecological processes in cancer, but to date, no system has been proposed for drawing clinically relevant distinctions between how different tumours are evolving. On the basis of a consensus conference of experts in the fields of cancer evolution and cancer ecology, we propose a framework for classifying tumours that is based on four relevant components. These are the diversity of neoplastic cells (intratumoural heterogeneity) and changes over time in that diversity, which make up an evolutionary index (Evo-index), as well as the hazards to neoplastic cell survival and the resources available to neoplastic cells, which make up an ecological index (Eco-index). We review evidence demonstrating the importance of each of these factors and describe multiple methods that can be used to measure them. Development of this classification system holds promise for enabling clinicians to personalize optimal interventions based on the evolvability of the patient's tumour. The Evo- and Eco-indices provide a common lexicon for communicating about how neoplasms change in response to interventions, with potential implications for clinical trials, personalized medicine and basic cancer research.

Reproducibility of F18-FDG PET radiomic features for different cervical tumor segmentation methods, gray-level discretization, and reconstruction algorithms

Site‐specific investigations of the role of radiomics in cancer diagnosis and therapy are emerging. We evaluated the reproducibility of radiomic features extracted from ¹⁸Flourine–fluorodeoxyglucose (¹⁸F‐FDG) PET images for three parameters: manual versus computer‐aided segmentation methods, gray‐level discretization, and PET image reconstruction algorithms. Our cohort consisted of pretreatment PET/CT scans from 88 cervical cancer patients. Two board‐certified radiation oncologists manually segmented the metabolic tumor volume (MTV₁ and MTV₂) for each patient. For comparison, we used a graphical‐based method to generate semiautomated segmented volumes (GBSV). To address any perturbations in radiomic feature values, we down‐sampled the tumor volumes into three gray‐levels: 32, 64, and 128 from the original gray‐level of 256. Finally, we analyzed the effect on radiomic features on PET images of eight patients due to four PET 3D‐reconstruction algorithms: maximum likelihood‐ordered subset expectation maximization (OSEM) iterative reconstruction (IR) method, fourier rebinning‐ML‐OSEM (FOREIR), FORE‐filtered back projection (FOREFBP), and 3D‐Reprojection (3DRP) analytical method. We extracted 79 features from all segmentation method, gray‐levels of down‐sampled volumes, and PET reconstruction algorithms. The features were extracted using gray‐level co‐occurrence matrices (GLCM), gray‐level size zone matrices (GLSZM), gray‐level run‐length matrices (GLRLM), neighborhood gray‐tone difference matrices (NGTDM), shape‐based features (SF), and intensity histogram features (IHF). We computed the Dice coefficient between each MTV and GBSV to measure segmentation accuracy. Coefficient values close to one indicate high agreement, and values close to zero indicate low agreement. We evaluated the effect on radiomic features by calculating the mean percentage differences (d¯) between feature values measured from each pair of parameter elements (i.e. segmentation methods: MTV₁‐MTV₂, MTV₁‐GBSV, MTV₂‐GBSV; gray‐levels: 64‐32, 64‐128, and 64‐256; reconstruction algorithms: OSEM‐FORE‐OSEM, OSEM‐FOREFBP, and OSEM‐3DRP). We used |d¯| as a measure of radiomic feature reproducibility level, where any feature scored |d¯| ±SD ≤ |25|% ± 35% was considered reproducible. We used Bland–Altman analysis to evaluate the mean, standard deviation (SD), and upper/lower reproducibility limits (U/LRL) for radiomic features in response to variation in each testing parameter. Furthermore, we proposed U/LRL as a method to classify the level of reproducibility: High— ±1% ≤ U/LRL ≤ ±30%; Intermediate— ±30% < U/LRL ≤ ±45%; Low— ±45 < U/LRL ≤ ±50%. We considered any feature below the low level as nonreproducible (NR). Finally, we calculated the interclass correlation coefficient (ICC) to evaluate the reliability of radiomic feature measurements for each parameter. The segmented volumes of 65 patients (81.3%) scored Dice coefficient >0.75 for all three volumes. The result outcomes revealed a tendency of higher radiomic feature reproducibility among segmentation pair MTV₁‐GBSV than MTV₂‐GBSV, gray‐level pairs of 64‐32 and 64‐128 than 64‐256, and reconstruction algorithm pairs of OSEM‐FOREIR and OSEM‐FOREFBP than OSEM‐3DRP. Although the choice of cervical tumor segmentation method, gray‐level value, and reconstruction algorithm may affect radiomic features, some features were characterized by high reproducibility through all testing parameters. The number of radiomic features that showed insensitivity to variations in segmentation methods, gray‐level discretization, and reconstruction algorithms was 10 (13%), 4 (5%), and 1 (1%), respectively. These results suggest that a careful analysis of the effects of these parameters is essential prior to any radiomics clinical application.

Textural features and SUV-based variables assessed by dual time point 18F-FDG PET/CT in locally advanced breast cancer

Aim

To study the influence of dual time point 18F-FDG PET/CT in textural features and SUV-based variables and their relation among them.

Methods

Fifty-six patients with locally advanced breast cancer (LABC) were prospectively included. All of them underwent a standard 18F-FDG PET/CT (PET-1) and a delayed acquisition (PET-2). After segmentation, SUV variables (SUVmax, SUVmean, and SUVpeak), metabolic tumor volume (MTV), and total lesion glycolysis (TLG) were obtained. Eighteen three-dimensional (3D) textural measures were computed including: run-length matrices (RLM) features, co-occurrence matrices (CM) features, and energies. Differences between all PET-derived variables obtained in PET-1 and PET-2 were studied.

Results

Significant differences were found between the SUV-based parameters and MTV obtained in the dual time point PET/CT, with higher values of SUV-based variables and lower MTV in the PET-2 with respect to the PET-1. In relation with the textural parameters obtained in dual time point acquisition, significant differences were found for the short run emphasis, low gray-level run emphasis, short run high gray-level emphasis, run percentage, long run emphasis, gray-level non-uniformity, homogeneity, and dissimilarity. Textural variables showed relations with MTV and TLG.

Conclusion

Significant differences of textural features were found in dual time point 18F-FDG PET/CT. Thus, a dynamic behavior of metabolic characteristics should be expected, with higher heterogeneity in delayed PET acquisition compared with the standard PET. A greater heterogeneity was found in bigger tumors.

Improved prediction of outcome in Parkinson's disease using radiomics analysis of longitudinal DAT SPECT images

No disease modifying therapies for Parkinson's disease (PD) have been found effective to date. To properly power clinical trials for discovery of such therapies, the ability to predict outcome in PD is critical, and there is a significant need for discovery of prognostic biomarkers of PD. Dopamine transporter (DAT) SPECT imaging is widely used for diagnostic purposes in PD. In the present work, we aimed to evaluate whether longitudinal DAT SPECT imaging can significantly improve prediction of outcome in PD patients. In particular, we investigated whether radiomics analysis of DAT SPECT images, in addition to use of conventional non-imaging and imaging measures, could be used to predict motor severity at year 4 in PD subjects. We selected 64 PD subjects (38 male, 26 female; age at baseline (year 0): 61.9 ± 7.3, range [46,78]) from the Parkinson's Progressive Marker Initiative (PPMI) database. Inclusion criteria included (i) having had at least 2 SPECT scans at years 0 and 1 acquired on a similar scanner, (ii) having undergone a high-resolution 3 T MRI scan, and (iii) having motor assessment (MDS-UPDRS-III) available in year 4 used as outcome measure. Image analysis included automatic region-of-interest (ROI) extraction on MRI images, registration of SPECT images onto the corresponding MRI images, and extraction of radiomic features. Non-imaging predictors included demographics, disease duration as well as motor and non-motor clinical measures in years 0 and 1. The image predictors included 92 radiomic features extracted from the caudate, putamen, and ventral striatum of DAT SPECT images at years 0 and 1 to quantify heterogeneity and texture in uptake. Random forest (RF) analysis with 5000 trees was used to combine both non-imaging and imaging variables to predict motor outcome (UPDRS-III: 27.3 ± 14.7, range [3,77]). The RF prediction was evaluated using leave-one-out cross-validation. Our results demonstrated that addition of radiomic features to conventional measures significantly improved (p < 0.001) prediction of outcome, reducing the absolute error of predicting MDS-UPDRS-III from 9.00 ± 0.88 to 4.12 ± 0.43. This shows that radiomics analysis of DAT SPECT images has a significant potential towards development of effective prognostic biomarkers in PD.

Comparison of image quality between step-and-shoot and continuous bed motion techniques in whole-body 18F-fluorodeoxyglucose positron emission tomography with the same acquisition duration

OBJECTIVE

This study aimed to compare the qualities of whole-body positron emission tomography (PET) images acquired by the step-and-shoot (SS) and continuous bed motion (CBM) techniques with approximately the same acquisition duration, through phantom and clinical studies.

METHODS

A body phantom with 10-37 mm spheres was filled with ¹⁸F-fluorodeoxyglucose (FDG) solution at a sphere-to-background radioactivity ratio of 4:1 and acquired by both techniques. Reconstructed images were evaluated by visual assessment, percentages of contrast (%Q _H) and background variability (%N) in accordance with the Japanese guideline for oncology FDG-PET/computed tomography (CT). To evaluate the variability of the standardized uptake value (SUV), the coefficient of variation (CV) for both maximum SUV and peak SUV was examined. Both the SUV values were additionally compared with those of standard images acquired for 30 min, and their accuracy was evaluated by the %difference (%Diff). In the clinical study, whole-body ¹⁸F-FDG PET/CT images of 60 patients acquired by both techniques were compared for liver signal-to-noise ratio (SNRliver), CV at end planes, and both SUV values.

RESULTS

In the phantom study, the visual assessment and %Q _H values of the two techniques did not differ from each other. However, the %N values of the CBM technique were significantly higher than those of the SS technique. Additionally, the CV and %Diff for both SUV values in the CBM images tended to be slightly higher than those in SS images. In the clinical study, the SNRliver values of CBM images were significantly lower than those of SS images, although the CV at the end planes in CBM images was significantly lower than those in SS images. In the Bland-Altman analysis for both SUV values, the mean differences were close to 0, and most lesions exhibited SUVs within the limits of agreement.

CONCLUSIONS

The CBM technique exhibited slightly lesser uniformity in the center plane than the SS technique. Additionally, in the phantom study, the CV and %Diff of SUV values in CBM images tended to be slightly higher than those of SS images. However, since these differences were subtle, they might be negligible in clinical settings.

Improving the prediction of overall survival for head and neck cancer patients using image biomarkers in combination with clinical parameters

PURPOSE

To develop and validate prediction models of overall survival (OS) for head and neck cancer (HNC) patients based on image biomarkers (IBMs) of the primary tumor and positive lymph nodes (Ln) in combination with clinical parameters.

MATERIAL AND METHODS

The study cohort was composed of 289 nasopharyngeal cancer (NPC) patients from China and 298 HNC patients from the Netherlands. Multivariable Cox-regression analysis was performed to select clinical parameters from the NPC and HNC datasets, and IBMs from the NPC dataset. Final prediction models were based on both IBMs and clinical parameters.

RESULTS

Multivariable Cox-regression analysis identified three independent IBMs (tumor Volume-density, Run Length Non-uniformity and Ln Major-axis-length). This IBM model showed a concordance(c)-index of 0.72 (95%CI: 0.65-0.79) for the NPC dataset, which performed reasonably with a c-index of 0.67 (95%CI: 0.62-0.72) in the external validation HNC dataset. When IBMs were added in clinical models, the c-index of the NPC and HNC datasets improved to 0.75 (95%CI: 0.68-0.82; p=0.019) and 0.75 (95%CI: 0.70-0.81; p<0.001), respectively.

CONCLUSION

The addition of IBMs from the primary tumor and Ln improved the prognostic performance of the models containing clinical factors only. These combined models may improve pre-treatment individualized prediction of OS for HNC patients.