Klug, Jason R.

Objectives: The aims of the study are to develop a prostate cancer risk prediction model that combines clinical and magnetic resonance imaging (MRI)–related findings and to assess the impact of adding Prostate Imaging-Reporting and Data System (PI-RADS) ≥3 lesions-level findings on its diagnostic performance. Methods: This 3-center retrospective study included prostate MRI examinations performed with clinical suspicion of clinically significant prostate cancer (csPCa) between 2018 and 2022. Pathological diagnosis within 1 year after the MRI was used to diagnose csPCa. Seven clinical, 3 patient-level MRI-related, and 4 lesion-level MRI-related findings were extracted. After feature selection, 2 logistic regression models with and without lesions-level findings were created using data from facility I and II (development cohort). The area under the receiver operating characteristic curve (AUC) between the 2 models was compared in the PI-RADS ≥3 population in the development cohort and Facility III (validation cohort) using the Delong test. Interfacility differences of the selected predictive variables were evaluated using the Kruskal-Wallis test or chi-squared test. Results: Selected lesion-level features included the peripheral zone involvement and apparent diffusion coefficient (ADC) values. The model with lesions-level findings had significantly higher AUC than the model without in 655 examinations in the development cohort (0.81 vs 0.79, respectively, P = 0.005), but not in 553 examinations in the validation cohort (0.77 vs 0.76, respectively). Large interfacility differences were seen in the ADC distribution (P < 0.001) and csPCa proportion in PI-RADS 3–5 (P < 0.001). Conclusions: Adding lesions-level findings improved the csPCa discrimination in the development but not the validation cohort. Interfacility differences impeded model generalization, including the distribution of reported ADC values and PI-RADS score-level csPCa proportion.

Purpose This study aimed to compare accuracy and efficiency of a convolutional neural network (CNN)–enhanced workflow for pancreas segmentation versus radiologists in the context of interreader reliability. Methods Volumetric pancreas segmentations on a data set of 294 portal venous computed tomographies were performed by 3 radiologists (R1, R2, and R3) and by a CNN. Convolutional neural network segmentations were reviewed and, if needed, corrected (“corrected CNN [c-CNN]” segmentations) by radiologists. Ground truth was obtained from radiologists' manual segmentations using simultaneous truth and performance level estimation algorithm. Interreader reliability and model's accuracy were evaluated with Dice-Sorenson coefficient (DSC) and Jaccard coefficient (JC). Equivalence was determined using a two 1-sided test. Convolutional neural network segmentations below the 25th percentile DSC were reviewed to evaluate segmentation errors. Time for manual segmentation and c-CNN was compared. Results Pancreas volumes from 3 sets of segmentations (manual, CNN, and c-CNN) were noninferior to simultaneous truth and performance level estimation–derived volumes [76.6 cm3 (20.2 cm3), P < 0.05]. Interreader reliability was high (mean [SD] DSC between R2-R1, 0.87 [0.04]; R3-R1, 0.90 [0.05]; R2-R3, 0.87 [0.04]). Convolutional neural network segmentations were highly accurate (DSC, 0.88 [0.05]; JC, 0.79 [0.07]) and required minimal-to-no corrections (c-CNN: DSC, 0.89 [0.04]; JC, 0.81 [0.06]; equivalence, P < 0.05). Undersegmentation (n = 47 [64%]) was common in the 73 CNN segmentations below 25th percentile DSC, but there were no major errors. Total inference time (minutes) for CNN was 1.2 (0.3). Average time (minutes) taken by radiologists for c-CNN (0.6 [0.97]) was substantially lower compared with manual segmentation (3.37 [1.47]; savings of 77.9%–87% [P < 0.0001]). Conclusions Convolutional neural network–enhanced workflow provides high accuracy and efficiency for volumetric pancreas segmentation on computed tomography.

110 Background: Sarcopenia or a loss of muscle mass increases with aging and is associated with increased overall mortality in patients with cancer. Recent advances in machine learning–based CNN algorithms have allowed for the rapid processing of digital images to produce image classifications of body composition. Since incidence of MM is highly associated with aging, we sought to determine if the presence of sarcopenia, as determined by utilizing this machine learning–based CNN algorithm on CT images, had prognostic value in patients with NDMM. Methods: We identified all patients with NDMM from January 2003 to July 2019 who had a standard-dose CT scan that included the L3 vertebral level performed within 6 months of diagnosis. Using a machine learning–based CNN-algorithm, abdominal CT images were analyzed to measure muscle area. These measurements were normalized by dividing the area values by the height of the patient squared (m2) to obtain skeletal muscle index (SMI) values. Patients were categorized as sarcopenic according to international gender-specific consensus cutoffs for SMI (male: < 55 cm2/m2 and female: < 39 cm2/m2). Patients with the following FISH cytogenetics were considered high risk (HR): t(4;14), t(14;16), t(14;20), and deletion 17p/monosomy 17 whereas the remainder were standard risk (SR). Survival analysis was performed using the Kaplan-Meier method and compared via the log-rank method. Results: The study cohort consisted of 344 patients. 68 (20%) were categorized as HR based on FISH cytogenetics.187 (54%) patients were sarcopenic based on their peri-diagnosis standard-dose CT scan. Sarcopenic patients were more likely to have ISS-3 disease (45% vs. 30%; p =.023), be male (73% vs. 48%; p <.001), and be ≥ age 75 (27% vs. 14%; p =.002) compared to non-sarcopenic patients. The median OS for patients with HR FISH and ISS 2 / 3 disease was 40 months and 57 months respectively compared to 90 months and 119 months for those with SR FISH and ISS-1 disease respectively (FISH: p <.004; ISS: p <.001). The median OS for sarcopenic patients was 44 months compared to 90 months for those not sarcopenic (p <.001). The time to next therapy (TTNT) for sarcopenic patients was 39 months compared to 45 months for those not sarcopenic (p =.05). In a multivariable model, the presence of sarcopenia (HR 1.64, 95% CI, 1.05–2.56; p =.03) retained significance in the presence of HR FISH, ISS 2 / 3 disease, and age ≥ 75. Conclusions: Gender-specific sarcopenia identified by a machine learning–based CNN algorithm significantly affects OS in patients with NDMM and is independent of age, ISS stage, and cytogenetic status. Future studies utilizing this machine learning–based methodology of assessing sarcopenia in larger prospective clinical trials are required to validate these findings.