Sovanlal Mukherjee, Anurima Patra, Hala Khasawneh, Panagiotis Korfiatis, Naveen Rajamohan, Garima Suman, Shounak Majumder, Ananya Panda, Matthew P Johnson, Nicholas B Larson, Darryl E Wright, Timothy L Kline, Joel G Fletcher, Suresh T Chari, Ajit H Goenka
Gastroenterology . 2022 Jul 1;S0016-5085(22)00728-4. doi: 10.1053/j.gastro.2022.06.066. Online ahead of print.
Background & aims: Our purpose was to detect pancreatic ductal adenocarcinoma (PDAC) at the prediagnostic stage (3-36 months prior to clinical diagnosis) using radiomics-based machine learning (ML) models, and to compare performance against radiologists in a case-control study.
Methods: Volumetric pancreas segmentation was performed on prediagnostic CTs (median interval between CT and PDAC diagnosis: 398 days) of 155 patients and an age-matched cohort of 265 subjects with normal pancreas. Total 88 first order and gray level radiomic features were extracted and 34 features were selected through LASSO-based feature selection method. Dataset was randomly divided into training (292 CTs: 110 prediagnostic and 182 controls) and test subsets (128 CTs: 45 prediagnostic and 83 controls). Four ML classifiers - K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Random Forest (RF) and XGBoost (XGB) - were evaluated. Specificity of model with highest accuracy was further validated on an independent internal dataset (n=176) and the public NIH dataset (n=80). Two radiologists (R4 and R5) independently evaluated the pancreas on a five-point diagnostic scale.
Results: Median (range) time between prediagnostic CTs of the test subset and PDAC diagnosis was 386 (97-1092) days. SVM had the highest sensitivity (mean; 95% CI) (95.5; 85.5-100.0), specificity (90.3; 84.3-91.5), F1-score (89.5; 82.3-91.7), AUC (0.98; 0.94-0.98) and accuracy (92.2%; 86.7-93.7) for classification of CTs into prediagnostic versus normal. All three other ML models KNN, RF, and XGB had comparable AUCs (0.95, 0.95, and 0.96, respectively). The high specificity of SVM was generalizable to both the independent internal (92.6%) and the NIH dataset (96.2%). In contrast, inter-reader radiologist agreement was only fair (Cohen's kappa 0.3) and their mean AUC (0.66; 0.46-0.86) was lower than each of the four ML models (AUCs: 0.95-0.98) (p < 0.001). Radiologists also recorded false positive indirect findings of PDAC in control subjects (n=83) (7% R4, 18% R5).
Conclusions: Radiomics-based ML models can detect PDAC from normal pancreas when it is beyond human interrogation capability at a substantial lead time prior to clinical diagnosis. Prospective validation and integration of such models with complementary fluid-based biomarkers has the potential for PDAC detection at a stage when surgical cure is a possibility.
Read Full Article Here: https://doi.org/10.1053/j.gastro.2022.06.066