Trivedi, Kamaxi H.

Background Failure of conventional imaging to detect pancreatic ductal adenocarcinoma (PDA) at its visually occult pre-diagnostic stage is a primary barrier to improving its otherwise poor rate of survival. Objective To develop and validate the Radiomics-based Early Detection MODel (REDMOD), an AI framework to identify subvisual radiomic signatures of pre-diagnostic PDA on standard-of-care CT. Designs REDMOD was trained on a multi-institutional cohort (n=969; 156 pre-diagnostic, 813 control) and tested on an independent set (n=493; 63 pre-diagnostic, 430 control), simulating a low prevalence (~1:6) early detection paradigm. The fully automated framework couples AI-driven segmentation with a heterogeneous ensemble architecture trained on a 40-feature radiomic signature derived from Synthetic Minority Over-sampling Technique (SMOTE)-balanced data. A tunable Youden Index-optimised classification threshold enables performance calibration without retraining. Validation included direct comparison with radiologists, longitudinal test–retest analysis and external specificity validation across two independent cohorts (n=539 and n=80). Results On an independent test set (n=493), REDMOD identified occult PDA (AUC 0.82; 73.0% sensitivity) at a median 475-day lead time. This represented nearly twofold higher sensitivity than radiologists (38.9%; p<0.001), which grew to nearly threefold (68.0% vs 23.0%) at >24 months lead time. REDMOD showed strong longitudinal stability (90–92% concordance) and generalisable specificity across multi-institutional (81.3%; n=539) and public (87.5%; n=80) datasets. Mechanistic analyses confirmed predictive power derived principally from multi-scale wavelet-filtered textural features (90% of selected signature), which outperformed unfiltered features (AUC 0.82 vs 0.74; p=0.007) in capturing subvisual architectural disruptions. Conclusions REDMOD is an automated, mechanistically grounded, longitudinally stable, externally validated AI that surpasses radiologists for PDA detection at its visually occult pre-diagnostic stage. These attributes position it for prospective validation in high-risk cohorts, a necessary step towards shifting the paradigm from late-stage symptomatic diagnosis to proactive pre-clinical interception.

BACKGROUND: Pancreatic ductal adenocarcinoma (PDAC) is an aggressive and lethal malignancy. Surgical resection is the only curative modality combined with neoadjuvant chemotherapy to improve survival. Given the limitations of traditional responses such as cross-sectional imaging (CT/MRI) or tumor markers, carbohydrate antigen 19-9 (CA19-9), the 2023 National Comprehensive Cancer Network guidelines included 18F-fluorodeoxyglucose (FDG)-PET as an adjunct to assess response to neoadjuvant chemotherapy. There are common misconceptions on the metabolic activity (tumor avidity) in PDAC so we aimed to describe the baseline characteristics and use of FDG-PET in a cohort of treatment-naive patients with PDAC. STUDY DESIGN: A single-center retrospective study was conducted capturing all biopsy-proven, treatment-naive patients with PDAC who underwent either baseline FDG-PET/CT or FDG-PET/MRI imaging between 2008 and 2023. Baseline FDG-PET characteristics were collected, including primary tumors’ maximum standardized uptake value defined as metabolic activity (FDG uptake) of tumor compared with surrounding pancreatic parenchymal background, and the identification of extrapancreatic metastatic disease. RESULTS: We identified 1,095 treatment-naive patients with PDAC who underwent baseline FDG-PET imaging at diagnosis. CA19-9 was elevated in 76% of patients. Overall, 96.3% (1,054) of patients had FDG-avid tumors with a median maximum standardized uptake value of 6.4. FDG-PET also identified suspicious extrapancreatic metastatic lesions in 50% of patients, with a higher proportion (p < 0.001) in PET/MRI (59.9%) vs PET/CT (44.3%). After controlling for CA19-9 elevation, PET/MRI was superior in detection of extrapancreatic lesions compared with PET/CT. CONCLUSIONS: FDG-PET has significant use in PDAC as a baseline imaging modality earlier neoadjuvant therapy given the majority of tumors are FDG-avid. FDG-PET can identify additional extrapancreatic suspicious lesions allowing for optimal initial staging, with PET/MRI having increased sensitivity over PET/CT.

Lung cancer is the most diagnosed cancer worldwide. Many non-malignant pulmonary lesions, such as tuberculosis, fungal infection, organizing pneumonia, inflammatory myofibroblastic tumor, and IgG4 disease, can mimic lung cancer due to their overlapping morphological appearance on imaging. These benign entities with minor differentiating imaging clues may go unnoticed in a high-volume cancer institution, leading to over-investigation that may result in repeated biopsies, pointless wedge resections, and related morbidities. However, with a thorough medical history, laboratory diagnostic work-up, and careful analysis of imaging findings, one can occasionally restrict the range of possible diagnoses or arrive at a definitive conclusion. When imaging features overlap, image-guided lung sampling is crucial since histopathological analysis is the gold standard.