Artificial intelligence for assessment of vascular involvement and tumor resectability on CT in patients with pancreatic cancer.

OBJECTIVE: This study aimed to develop and evaluate an automatic model using artificial intelligence (AI) for quantifying vascular involvement and classifying tumor resectability stage in patients with pancreatic ductal adenocarcinoma (PDAC), primarily to support radiologists in referral centers. Resectability of PDAC is determined by the degree of vascular involvement on computed tomography scans (CTs), which is associated with considerable inter-observer variability. METHODS: We developed a semisupervised machine learning segmentation model to segment the PDAC and surrounding vasculature using 613 CTs of 467 patients with pancreatic tumors and 50 control patients. After segmenting the relevant structures, our model quantifies vascular involvement by measuring the degree of the vessel wall that is in contact with the tumor using AI-segmented CTs. Based on these measurements, the model classifies the resectability stage using the Dutch Pancreatic Cancer Group criteria as either resectable, borderline resectable, or locally advanced (LA). RESULTS: We evaluated the performance of the model using a test set containing 60 CTs from 60 patients, consisting of 20 resectable, 20 borderline resectable, and 20 locally advanced cases, by comparing the automated analysis obtained from the model to expert visual vascular involvement assessments. The model concurred with the radiologists on 227/300 (76%) vessels for determining vascular involvement. The model's resectability classification agreed with the radiologists on 17/20 (85%) resectable, 16/20 (80%) for borderline resectable, and 15/20 (75%) for locally advanced cases. CONCLUSIONS: This study demonstrates that an AI model may allow automatic quantification of vascular involvement and classification of resectability for PDAC. RELEVANCE STATEMENT: This AI model enables automated vascular involvement quantification and resectability classification for pancreatic cancer, aiding radiologists in treatment decisions, and potentially improving patient outcomes. KEY POINTS: • High inter-observer variability exists in determining vascular involvement and resectability for PDAC. • Artificial intelligence accurately quantifies vascular involvement and classifies resectability for PDAC. • Artificial intelligence can aid radiologists by automating vascular involvement and resectability assessments.

MR Elastography of the Pancreas: Bowel Preparation and Repeatability Assessment in Pancreatic Cancer Patients and Healthy Controls.

BACKGROUND: Pancreatic ductal adenocarcinoma (PDAC) stromal viscoelasticity can be measured using MR elastography (MRE). Bowel preparation regimens could affect MRE quality and knowledge on repeatability is crucial for clinical implementation. PURPOSE: To assess effects of four bowel preparation regimens on MRE quality and to evaluate repeatability and differentiate patients from healthy controls. STUDY TYPE: Prospective. POPULATION: 15 controls (41 ± 16 years; 47% female), 16 PDAC patients (one excluded, 66 ± 12 years; 40% female) with 15 age-/sex-matched controls (65 ± 11 years; 40% female). Final sample size was 25 controls and 15 PDAC. FIELD STRENGTH/SEQUENCE: 3-T, spin-echo echo-planar-imaging, turbo spin-echo, and fast field echo gradient-echo. ASSESSMENT: Four different regimens were used: fasting; scopolaminebutyl; drinking 0.5 L water; combination of 0.5 L water and scopolaminebutyl. MRE signal-to-noise ratio (SNR) was compared between all regimens. MRE repeatability (test-retest) and differences in shear wave speed (SWS) and phase angle (ϕ) were assessed in PDAC and controls. Regions-of-interest were defined for tumor, nontumorous (n = 8) tissue in PDAC, and whole pancreas in controls. Two radiologists delineated tumors twice for evaluation of intraobserver and interobserver variability. STATISTICAL TESTS: Repeated measures analysis of variance, coefficients of variation (CoVs), Bland-Altman analysis, (un)paired t-test, Mann-Whitney U-test, and Wilcoxon signed-rank test. P-value<0.05 was considered statistically significant. RESULTS: Preparation regimens did not significantly influence MRE-SNR. Therefore, the least burdensome preparation (fasting only) was continued. CoVs for tumor SWS were: intrasession (12.8%) and intersession (21.7%), and intraobserver (7.9%) and interobserver (10.3%) comparisons. For controls, CoVs were intrasession (4.6%) and intersession (6.4%). Average SWS for tumor, nontumor, and healthy tissue were: 1.74 ± 0.58, 1.38 ± 0.27, and 1.18 ± 0.16 m/sec (ϕ: 1.02 ± 0.17, 0.91 ± 0.07, and 0.85 ± 0.08 rad), respectively. Significant differences were found between all groups, except for ϕ between healthy-nontumor (P = 0.094). DATA CONCLUSION: The proposed bowel preparation regimens may not influence MRE quality. MRE may be able to differentiate between healthy tissue-tumor and tumor-nontumor. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY STAGE: 2.