Deep Learning-Based Interpretable AI for Prostate T2W MRI Quality Evaluation

RATIONALE AND OBJECTIVES

Prostate MRI quality is essential in guiding prostate biopsies. However, assessment of MRI quality is subjective with variation. Quality degradation sources exert varying impacts based on the sequence under consideration, such as T2W versus DWI. As a result, employing sequence-specific techniques for quality assessment could yield more advantageous outcomes. This study aims to develop an AI tool that offers a more consistent evaluation of T2W prostate MRI quality, efficiently identifying suboptimal scans while minimizing user bias.

MATERIALS AND METHODS

This retrospective study included 1046 patients from three cohorts (ProstateX [n = 347], All-comer in-house [n = 602], enriched bad-quality MRI in-house [n = 97]) scanned between January 2011 and May 2022. An expert reader assigned T2W MRIs a quality score. A train-validation-test split of 70:15:15 was applied, ensuring equal distribution of MRI scanners and protocols across all partitions. T2W quality AI classification model was based on 3D DenseNet121 architecture using MONAI framework. In addition to multiclassification, binary classification was utilized (Classes 0/1 vs. 2). A score of 0 was given to scans considered non-diagnostic or unusable, a score of 1 was given to those with acceptable diagnostic quality with some usability but with some quality distortions present, and a score of 2 was given to those considered optimal diagnostic quality and usability. Partial occlusion sensitivity maps were generated for anatomical correlation. Three body radiologists assessed reproducibility within a subgroup of 60 test cases using weighted Cohen Kappa.

RESULTS

The best validation multiclass accuracy of 77.1% (121/157) was achieved during training. In the test dataset, multiclassification accuracy was 73.9% (116/157), whereas binary accuracy was 84.7% (133/157). Sub-class sensitivity for binary quality distortion classification for class 0 was 100% (18/18), and sub-class specificity for T2W classification of absence/minimal quality distortions for class 2 was 90.5% (95/105). All three readers showed moderate to substantial agreement with ground truth (R1-R3 κ = 0.588, κ = 0.649, κ = 0.487, respectively), moderate to substantial agreement with each other (R1-R2 κ = 0.599, R1-R3 κ = 0.612, R2-R3 κ = 0.685), fair to moderate agreement with AI (R1-R3 κ = 0.445, κ = 0.410, κ = 0.292, respectively). AI showed substantial agreement with ground truth (κ = 0.704). 3D quality heatmap evaluation revealed that the most critical non-diagnostic quality imaging features from an AI perspective related to obscuration of the rectoprostatic space (94.4%, 17/18).

CONCLUSION

The 3D AI model can assess T2W prostate MRI quality with moderate accuracy and translate whole sequence-level classification labels into 3D voxel-level quality heatmaps for interpretation. Image quality has a significant downstream impact on ruling out clinically significant cancers. AI may be able to help with reproducible identification of MRI sequences requiring re-acquisition with explainability.

The kidney imaging surveillance scoring system (KISSS): using qualitative MRI features to predict growth rate of renal tumors in patients with von-Hippel Lindau (VHL) syndrome

OBJECTIVE

To determine the reliability of an MRI-based qualitative kidney imaging surveillance scoring system (KISSS) and assess which imaging features predict growth rate (GR) of renal tumors in patients with VHL.

MATERIALS AND METHODS

We identified 55 patients with VHL with 128 renal tumors who underwent intervention from 2015 to 2020 at the National Cancer Institute. All patients had 2 preoperative MRIs at least 3 months apart. Two fellowship-trained radiologists scored each tumor on location and MR-sequence-specific imaging parameters from the earlier MRI. Weighted kappa was used to determine the degree of agreement between radiologists for each parameter. GR was calculated as the difference in maximum tumor dimension over time (cm/year). Differences in mean growth rate (MGR) within categories of each imaging variable were assessed by ANOVA.

RESULTS

Apart from tumor margin and renal sinus, reliability was at least moderate (K > 0.40) for imaging parameters. Median initial tumor size was 2.1 cm, with average follow-up of 1.2 years. Tumor MGR was 0.42 cm/year. T2 hypointense, mixed/predominantly solid, and high restricted diffusion tumors grew faster. When comparing different combinations of these variables, the model with the lowest mean error among both radiologists utilized only solid/cystic and restricted diffusion features.

CONCLUSIONS

We demonstrate a novel MR-based scoring system (KISSS) that has good precision with minimal training and can be applied to other qualitative radiology studies. A subset of imaging variables (T2 intensity; restricted diffusion; and solid/cystic) were independently associated with growth rate in VHL renal tumors, with the combination of the latter two most optimal. Additional validation, including in sporadic RCC population, is warranted.

Evaluating the performance of Generative Pre-trained Transformer-4 (GPT-4) in standardizing radiology reports

OBJECTIVE

Radiology reporting is an essential component of clinical diagnosis and decision-making. With the advent of advanced artificial intelligence (AI) models like GPT-4 (Generative Pre-trained Transformer 4), there is growing interest in evaluating their potential for optimizing or generating radiology reports. This study aimed to compare the quality and content of radiologist-generated and GPT-4 AI-generated radiology reports.

METHODS

A comparative study design was employed in the study, where a total of 100 anonymized radiology reports were randomly selected and analyzed. Each report was processed by GPT-4, resulting in the generation of a corresponding AI-generated report. Quantitative and qualitative analysis techniques were utilized to assess similarities and differences between the two sets of reports.

RESULTS

The AI-generated reports showed comparable quality to radiologist-generated reports in most categories. Significant differences were observed in clarity (p = 0.027), ease of understanding (p = 0.023), and structure (p = 0.050), favoring the AI-generated reports. AI-generated reports were more concise, with 34.53 fewer words and 174.22 fewer characters on average, but had greater variability in sentence length. Content similarity was high, with an average Cosine Similarity of 0.85, Sequence Matcher Similarity of 0.52, BLEU Score of 0.5008, and BERTScore F1 of 0.8775.

CONCLUSION

The results of this proof-of-concept study suggest that GPT-4 can be a reliable tool for generating standardized radiology reports, offering potential benefits such as improved efficiency, better communication, and simplified data extraction and analysis. However, limitations and ethical implications must be addressed to ensure the safe and effective implementation of this technology in clinical practice.

CLINICAL RELEVANCE STATEMENT

The findings of this study suggest that GPT-4 (Generative Pre-trained Transformer 4), an advanced AI model, has the potential to significantly contribute to the standardization and optimization of radiology reporting, offering improved efficiency and communication in clinical practice.

KEY POINTS

• Large language model-generated radiology reports exhibited high content similarity and moderate structural resemblance to radiologist-generated reports. • Performance metrics highlighted the strong matching of word selection and order, as well as high semantic similarity between AI and radiologist-generated reports. • Large language model demonstrated potential for generating standardized radiology reports, improving efficiency and communication in clinical settings.

Deep-Learning-Based Whole-Lung and Lung-Lesion Quantification Despite Inconsistent Ground Truth: Application to Computerized Tomography in SARS-CoV-2 Nonhuman Primate Models

RATIONALE AND OBJECTIVES

Animal modeling of infectious diseases such as coronavirus disease 2019 (COVID-19) is important for exploration of natural history, understanding of pathogenesis, and evaluation of countermeasures. Preclinical studies enable rigorous control of experimental conditions as well as pre-exposure baseline and longitudinal measurements, including medical imaging, that are often unavailable in the clinical research setting. Computerized tomography (CT) imaging provides important diagnostic, prognostic, and disease characterization to clinicians and clinical researchers. In that context, automated deep-learning systems for the analysis of CT imaging have been broadly proposed, but their practical utility has been limited. Manual outlining of the ground truth (i.e., lung-lesions) requires accurate distinctions between abnormal and normal tissues that often have vague boundaries and is subject to reader heterogeneity in interpretation. Indeed, this subjectivity is demonstrated as wide inconsistency in manual outlines among experts and from the same expert. The application of deep-learning data-science tools has been less well-evaluated in the preclinical setting, including in nonhuman primate (NHP) models of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection/COVID-19, in which the translation of human-derived deep-learning tools is challenging. The automated segmentation of the whole lung and lung lesions provides a potentially standardized and automated method to detect and quantify disease.

MATERIALS AND METHODS

We used deep-learning-based quantification of the whole lung and lung lesions on CT scans of NHPs exposed to SARS-CoV-2. We proposed a novel multi-model ensemble technique to address the inconsistency in the ground truths for deep-learning-based automated segmentation of the whole lung and lung lesions. Multiple models were obtained by training the convolutional neural network (CNN) on different subsets of the training data instead of having a single model using the entire training dataset. Moreover, we employed a feature pyramid network (FPN), a CNN that provides predictions at different resolution levels, enabling the network to predict objects with wide size variations.

RESULTS

We achieved an average of 99.4 and 60.2% Dice coefficients for whole-lung and lung-lesion segmentation, respectively. The proposed multi-model FPN outperformed well-accepted methods U-Net (50.5%), V-Net (54.5%), and Inception (53.4%) for the challenging lesion-segmentation task. We show the application of segmentation outputs for longitudinal quantification of lung disease in SARS-CoV-2-exposed and mock-exposed NHPs.

CONCLUSION

Deep-learning methods should be optimally characterized for and targeted specifically to preclinical research needs in terms of impact, automation, and dynamic quantification independently from purely clinical applications.

Treatment Strategies for Hereditary Kidney Cancer: Current Recommendations and Updates

A subset of renal tumors (5-8%) are associated with syndromes such as von Hippel-Lindau (VHL) syndrome, Birt-Hogg-Dubé syndrome (BHD), tuberous sclerosis complex (TSC), hereditary papillary renal carcinoma (HPRC), hereditary leiomyomatosis and renal cell cancer syndrome (HLRCC), and BRCA1 associated protein (BAP1) tumor predisposition syndrome, succinate dehydrogenase RCC (SDHB/C/D). These syndromes have their specific defined genetic alterations and associated extrarenal manifestations. Due to varying histopathology and aggressiveness of the tumors amongst these syndromes, the management strategies can range from active surveillance to upfront surgical resection. This review delineates specific characteristics of the most common familial renal cancer syndromes and discusses current management strategies.

Self-gated 3D stack-of-spirals UTE pulmonary imaging at 0.55T

PURPOSE

To develop an isotropic high-resolution stack-of-spirals UTE sequence for pulmonary imaging at 0.55 Tesla by leveraging a combination of robust respiratory-binning, trajectory correction, and concomitant-field corrections.

METHODS

A stack-of-spirals golden-angle UTE sequence was used to continuously acquire data for 15.5 minutes. The data was binned to a stable respiratory phase based on superoinferior readout self-navigator signals. Corrections for trajectory errors and concomitant field artifacts, along with image reconstruction with conjugate gradient SENSE, were performed inline within the Gadgetron framework. Finally, data were retrospectively reconstructed to simulate scan times of 5, 8.5, and 12 minutes. Image quality was assessed using signal-to-noise, image sharpness, and qualitative reader scores. The technique was evaluated in healthy volunteers, patients with coronavirus disease 2019 infection, and patients with lung nodules.

RESULTS

The technique provided diagnostic quality images with parenchymal lung SNR of 3.18 ± 0.0.60, 4.57 ± 0.87, 5.45 ± 1.02, and 5.89 ± 1.28 for scan times of 5, 8.5, 12, and 15.5 minutes, respectively. The respiratory binning technique resulted in significantly sharper images (p < 0.001) as measured with relative maximum derivative at the diaphragm. Concomitant field corrections visibly improved sharpness of anatomical structures away from iso-center. The image quality was maintained with a slight loss in SNR for simulated scan times down to 8.5 minutes. Inline image reconstruction and artifact correction were achieved in <5 minutes.

CONCLUSION

The proposed pulmonary imaging technique combined efficient stack-of-spirals imaging with robust respiratory binning, concomitant field correction, and trajectory correction to generate diagnostic quality images with 1.75 mm isotropic resolution in 8.5 minutes on a high-performance 0.55 Tesla system.

POS0802 18F-FLUORODEOXYGLUCOSE POSITRON EMISSION TOMOGRAPHY AS A PREDICTOR OF ANGIOGRAPHIC PROGRESSION OF DISEASE IN LARGE-VESSEL VASCULITIS

Background:

Giant cell arteritis (GCA) and Takayasu’s arteritis (TAK) are the two main forms of large-vessel vasculitis (LVV). Although angiography is essential to detect vascular disease in patients with LVV, there is limited prospective data characterizing change in arterial lesions over time, and factors that predict angiographic change remain unknown.

Objectives:

The objectives of this study were to: 1) describe longitudinal change in angiographic studies in patients with GCA and TAK and 2) determine whether FDG-PET activity predicts angiographic progression of disease.

Methods:

Patients with GCA or TAK were recruited into a prospective, observational cohort. All patients underwent baseline magnetic resonance (MR) or computed tomography (CT) angiography and a follow-up study (same modality) ≥6 months after baseline per a standardized imaging protocol. For patients who had multiple angiograms, the baseline and most recent images were compared. Arterial lesions, defined as stenosis, occlusion, or aneurysm, were evaluated by visual inspection in 4 segments of the aorta and 13 branch arteries by a single reader blinded to clinical status. On follow up angiography, the development of new lesions in these same territories was recorded, and existing lesions were characterized as improved, worsened, or unchanged by visual inspection, with confirmation by an independent reader.

All patients underwent FDG-PET on the same date as angiography. Qualitative assessment of FDG uptake was performed in each corresponding arterial territory evaluated by angiography. Active vasculitis was defined as greater FDG uptake in the arterial wall compared to the liver by visual inspection.

Results:

At the baseline visit, there were 248 arterial lesions (21%) out of 1162 arterial territories evaluated from 70 patients with LVV (TAK=38; GCA=32). Baseline characteristics were as follows: Age [TAK=29.5 years (18.4-39.5), GCA=69.6 years (60.7-75.5)], Female gender [TAK=30 patients (79%), GCA=23 patients (72%)], Disease duration [TAK=2.2 years (0.6-5.5), GCA=0.7 years (0.1-2.6)], Active clinical disease [TAK=17 patients (45%), GCA=20 patients (63%)].

Over 1.6 years (1.0-2.7) of median follow-up, no angiographic change was observed in 1,132 (97%) arterial territories. New lesions developed in 8 arterial territories, exclusively in 5 patients with TAK. Arterial lesions improved in 16 territories (GCA = 7, TAK = 9) and worsened in 6 territories (GCA = 1, TAK = 5). Patients with angiographic improvement were initially imaged earlier in the disease course compared to patients with new/worsening lesions (median 1.1 vs 16.4 months, p=0.09). Patients with angiographic improvement had significantly lower acute phase reactants at follow-up compared to patients with new/worsening arterial lesions [median ESR 3.0 (2.0-15.0) vs. 27.0 (7.3-39) mm/h, p<0.01; median CRP 0.7 (0.3-1.4) vs. 6.1 (3.1-19.6) mg/L, p<0.01]. Seventy-nine percent of patients with new/worsening arterial lesions had received increased treatment over the follow-up interval compared to 100% patients with improved arterial lesions, p=0.09.

FDG-PET activity was evaluated in 1091/1162 (94%) of corresponding arterial territories. PET activity in an arterial territory at baseline was significantly associated with change in that arterial territory (either new/worsening or improvement) on follow-up angiography (p<0.01) (FIGURE 1). PET activity had a sensitivity of 80% and specificity of 74% for predicting change in arterial lesions. Most arterial territories without PET activity at baseline remained unchanged over time by angiography, yielding a negative predictive value of 99%. (FIGURE 1).

Conclusion:

Development of new arterial lesions is infrequent in LVV. Change in arterial lesions is dynamic, and improvement can occur. FDG-PET activity predicts change in angiographic lesions, and lack of PET activity is strongly associated with stable angiographic disease. These data may inform guideline recommendations for imaging monitoring in LVV.

Figure 1.

Disclosure of Interests:

None declared

Artificial intelligence for the detection of COVID-19 pneumonia on chest CT using multinational datasets

Chest CT is emerging as a valuable diagnostic tool for clinical management of COVID-19 associated lung disease. Artificial intelligence (AI) has the potential to aid in rapid evaluation of CT scans for differentiation of COVID-19 findings from other clinical entities. Here we show that a series of deep learning algorithms, trained in a diverse multinational cohort of 1280 patients to localize parietal pleura/lung parenchyma followed by classification of COVID-19 pneumonia, can achieve up to 90.8% accuracy, with 84% sensitivity and 93% specificity, as evaluated in an independent test set (not included in training and validation) of 1337 patients. Normal controls included chest CTs from oncology, emergency, and pneumonia-related indications. The false positive rate in 140 patients with laboratory confirmed other (non COVID-19) pneumonias was 10%. AI-based algorithms can readily identify CT scans with COVID-19 associated pneumonia, as well as distinguish non-COVID related pneumonias with high specificity in diverse patient populations.

Update on Hereditary Renal Cancer and Imaging Implications

Up to 8% of renal cancers are thought to have a hereditary component. Several hereditary renal cancer syndromes have been identified over the last few decades. It is important for the radiologist to be aware of findings associated with hereditary renal cancer syndromes to detect tumors early, enroll patients in appropriate surveillance programs, and improve outcomes for the patient and affected family members. This review discusses from a radiologist's perspective well-known hereditary renal cancer syndromes and emerging genetic mutations associated with renal cancer that are less well characterized, focusing on imaging features and known associations.

CT analysis of anatomical distribution of melorheostosis challenges the sclerotome hypothesis

Melorheostosis (MEL) is a rare disease of high bone mass with patchy skeletal distribution affecting the long bones. We recently reported somatic mosaic mutations in MAP2K1 in 8 of 15 patients with the disease. The unique anatomic distribution of melorheostosis is of great interest. The disease remains limited to medial or lateral side of the extremity with proximo-distal progression. This pattern of distribution has historically been attributed to sclerotomes (area of bone which is innervated by a single spinal nerve level). In a further analysis of our study on MEL, 30 recruited patients underwent whole body CT scans to characterize the anatomic distribution of the disease. Two radiologists independently reviewed these scans and compared it to the proposed map of sclerotomes. We found that the disease distribution conformed to the distribution of a single sclerotome in only 5 patients (17%). In another 12 patients, the lesions spanned parts of contiguous sclerotomes but did not involve the entire extent of the sclerotomes. Our findings raise concerns about the sclerotomal hypothesis being the definitive explanation for the pattern of anatomic distribution in MEL. We believe that the disease distribution can be explained by clonal proliferation of a mutated skeletal progenitor cell along the limb axis. Studies in mice models on clonal proliferation in limb buds mimic the patterns seen in melorheostosis. We also support this hypothesis by the dorso-ventral confinement of melorheostotic lesion in a patient with low allele frequency of MAP2K1-positive osteoblasts and low skeletal burden of the disease. This suggests that the mutation occurred after the formation of dorso-ventral plane. Further studies on limb development are needed to better understand the etiology, pathophysiology and pattern of disease distribution in all patients with MEL.

Cumulative Radiation Exposures from CT Screening and Surveillance Strategies for von Hippel-Lindau-associated Solid Pancreatic Tumors

Purpose To assess the potential ionizing radiation exposure from CT scans for both screening and surveillance of patients with von Hippel-Lindau (VHL) syndrome. Materials and Methods For this retrospective study, abdomen-pelvic (AP) and chest-abdomen-pelvic (CAP) CT scans were performed with either a three-phase (n = 1242) or a dual-energy virtual noncontrast protocol (VNC; n = 149) in 747 patients with VHL syndrome in the National Institutes of Health Clinical Center between 2009 and 2015 (mean age, 47.6 years ± 14.6 [standard deviation]; age range, 12-83 years; 320 women [42.8%]). CT scanning parameters for patients with pancreatic neuroendocrine tumors (PNETs; 124 patients and 381 scans) were compared between a tumor diameter-based surveillance protocol and a VHL genotype and tumor diameter-based algorithm (a tailored algorithm) developed by three VHL clinicians. Organ and lifetime radiation doses were estimated by two radiologists and five radiation scientists. Cumulative radiation doses were compared between the PNET surveillance algorithms by analyses of variance, and a two-tailed P value less than .05 indicated statistical significance. Results Median cumulative colon doses for annual CAP and AP CT scans from age 15 to 40 years ranged from 0.34 Gy (5th-95th percentiles, 0.18-0.75; dual-energy VNC CT) to 0.89 Gy (5th-95th percentiles, 0.42-1.0; three-phase CT). For the current PNET surveillance protocol, the cumulative effective radiation dose from age 40 to 65 years was 682 mSv (tumors < 1.2 cm) and 2125 mSv (tumors > 3 cm). The tailored algorithm could halve these doses for patients with initial tumor diameter less than 1.2 cm (P < .001). Conclusion CT screening of patients with von Hippel-Lindau syndrome can lead to substantial radiation exposures, even with dual-energy virtual noncontrast CT. A genome and tumor diameter-based algorithm for pancreatic neuroendocrine tumor surveillance may potentially reduce lifetime radiation exposure. © RSNA, 2018 Online supplemental material is available for this article.

Oxidative Stress and Cardiovascular Risk in Type 1 Diabetes Mellitus: Insights From the DCCT/EDIC Study

Background

Hyperglycemia leading to increased oxidative stress is implicated in the increased risk for the development of macrovascular and microvascular complications in patients with type 1 diabetes mellitus.

Methods and Results

A random subcohort of 349 participants was selected from the DCCT / EDIC (Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications) cohort. This included 320 controls and 29 cardiovascular disease cases that were augmented with 98 additional known cases to yield a case cohort of 447 participants (320 controls, 127 cases). Biosamples from DCCT baseline, year 1, and closeout of DCCT , and 1 to 2 years post‐ DCCT ( EDIC years 1 and 2) were measured for markers of oxidative stress, including plasma myeloperoxidase, paraoxonase activity, urinary F 2α isoprostanes, and its metabolite, 2,3 dinor‐8 iso prostaglandin F 2α . Following adjustment for glycated hemoblobin and weighting the observations inversely proportional to the sampling selection probabilities, higher paraoxonase activity, reflective of antioxidant activity, and 2,3 dinor‐8 iso prostaglandin F 2α , an oxidative marker, were significantly associated with lower risk of cardiovascular disease (−4.5% risk for 10% higher paraoxonase, P <0.003; −5.3% risk for 10% higher 2,3 dinor‐8 iso prostaglandin F 2α , P =0.0092). In contrast, the oxidative markers myeloperoxidase and F 2α isoprostanes were not significantly associated with cardiovascular disease after adjustment for glycated hemoblobin. There were no significant differences between DCCT intensive and conventional treatment groups in the change in all biomarkers across time segments.

Conclusions

Heightened antioxidant activity (rather than diminished oxidative stress markers) is associated with lower cardiovascular disease risk in type 1 diabetes mellitus, but these biomarkers did not change over time with intensification of glycemic control.

Clinical Trial Registration

URL : https://www.clinicaltrials.gov . Unique identifiers: NCT 00360815 and NCT 00360893.

Evaluation of optimized breath-hold and free-breathing 3D ultrashort echo time contrast agent-free MRI of the human lung

PURPOSE

To evaluate an optimized stack of radials ultrashort echo time (UTE) 3D magnetic resonance imaging (MRI) sequence for breath-hold and free-breathing imaging of the human lung.

MATERIALS AND METHODS

A 3D stack of ultrashort echo time radials trajectory was optimized for coronal and axial lower-resolution breath-hold and higher-resolution free-breathing scans using Bloch simulations. The sequence was evaluated in 10 volunteers, without the use of contrast agents. Signal-to-noise ratio (SNR) mean and 95% confidence interval (CI) were determined from separate signal and noise images in a semiautomated fashion. The four scanning schemes were evaluated for significant differences in image quality using Student's t-test. Ten clinical patients were scanned with the sequence and findings were compared with concomitant computed tomography (CT) in nine patients. Breath-hold 3D spokes images were compared with 3D stack of radials in five volunteers. A Mann-Whitney U-test was performed to test significance in both cases.

RESULTS

Breath-hold imaging of the entire lung in volunteers was performed with SNR (mean = 42.5 [CI]: 35.5-49.5; mean = 34.3 [CI]: 28.6-40) in lung parenchyma for coronal and axial scans, respectively, which can be used as a quick scout scan. Longer respiratory triggered free-breathing scan enabled high-resolution UTE scanning with mean SNR of 14.2 ([CI]: 12.9-15.5) and 9.2 ([CI]: 8.2-10.2) for coronal and axial scans, respectively. Axial free-breathing scans showed significantly higher image quality (P = 0.008) than the three other scanning schemes. The mean score for comparison with CT was 1.67 (score 0: n = 0; 1: n = 3; 2: n = 6). There was no significant difference between CT and MRI (P = 0.25). 3D stack of radials images were significantly better than 3D spokes images (P < 0.001).

CONCLUSION

The optimized 3D stack of radials trajectory was shown to provide high-quality MR images of the lung parenchyma without the use of MRI contrast agents. The sequence may offer the possibility of breath-hold imaging and provides greater flexibility in trading off slice thickness and parallel imaging for scan time.

Analysis of fiducials implanted during endoscopic ultrasound (EUS) for locally advanced rectal cancer patients receiving high-dose rate endorectal brachytherapy (Endo-HDR)

655

Background: Rectal cancer affects over 40,000 patients in the US per year. The current standard of care for patients with localized rectal cancer is neoadjuvant radiation therapy with concurrent chemotherapy (NCRT) followed by surgery; however, it has shown no proven survival benefit for locally advanced rectal cancer patients. Preliminary results show that a short course of radiation therapy, using high-dose rate endorectal brachytherapy (Endo-HDR), may be as effective with less toxicity and delay to time of surgery. This requires the placement of fiducial markers, using an endoscopic ultrasound guided method (EUS), into the tumor for accurate source placement and treatment. Our aim is to compare three different types of fiducials in terms of visibility and migration. Methods: 12 patients with locally advanced rectal cancer that received Endo-HDR and EUS guided fiducial placement were retrospectively evaluated at JHH. Results: 12 patients underwent EUS guided placement of 42 fiducials. For 11 of our 12 patients, the mean number of fiducials placed per patient was 3.63 (SD 1.03) using a 19-gauge needle. One patient received 2 fiducials using a 22- gauge needle. Of the 12 patients that received fiducials, 3 received traditional fiducials (TF), 8 received segmented fiducials (SF) and 1 received foldable fiducials. All fiducials were clearly visible. The mean number of fiducials that detached from implanted site before surgery for patients with TFs was 0.667, and for patients with SFs was 0.875 (p=0.744). The median migration distance, as measured by interfiduciary distance, for segmented fiducials was significantly larger when compared to traditional fiducials (0.45 cm for SF compared to 0.1 cm for TF; p=0.049) Conclusions: SFs appear to be less stable, with regards to migration, in the rectum when compared to traditional fiducials in our patient population. These differences could be due to placement difficulty or operator dependent differences. Improvement in fiducial structure is required in order to help decrease migration and detachment and maximize visualization, which will lead to more accurate administration of Endo-HDR.