RESUMO
Objectives: CT scans are commonly performed in patients with chronic pancreatitis (CP). Osteopathy and fractures are recognized in CP but no osteoporosis screening guidelines are recommended. "Opportunistic" CT scan-derived bone density thresholds are assessed for identifying osteoporosis in CP. Methods: Retrospective pilot cohort study. CP subjects who had CT scans and dual-energy x-ray absorptiometry (DXA) within 1 year were included. CT-derived bone density was measured at the L1 level. Pearson's correlation was performed between age and CT-derived bone density in Hounsfield unit (HU). Univariate analysis using HU to identify osteoporosis was performed at various thresholds of bone density. The discriminatory ability of the model was evaluated with the area under the receiver operating characteristic (ROC) curve (AUC). Several HU thresholds were tested. Results: Twenty-seven CP subjects were included, of whom 11 had normal bone density, 12 osteopenia, and four osteoporosis on DXA. The mean age was 59.9 years (SD 13.0). There was a negative correlation of age with HU (r = -0.519, p = 0.006). CT-derived bone density predicted DXA-based osteoporosis in the univariable analysis (Odds Ratio (OR) = 0.97 95% Confidence Interval (CI) 0.94-1.00, p = 0.03). HU thresholds were tested. A threshold of 106 HU maximized the accuracy (AUC of 0.870). Conclusions: CT scan may be repurposed for "opportunistic" screening to rule out osteoporosis in CP. A larger study is warranted to confirm these results.
RESUMO
Research in computer-aided diagnosis (CAD) and the application of artificial intelligence (AI) in the endoscopic evaluation of the gastrointestinal tract is novel. Since colonoscopy and detection of polyps can decrease the risk of colon cancer, it is recommended by multiple national and international societies. However, the procedure of colonoscopy is performed by humans where there are significant interoperator and interpatient variations, and hence, the risk of missing detection of adenomatous polyps. Early studies involving CAD and AI for the detection and differentiation of polyps show great promise. In this appraisal, we review existing scientific aspects of AI in CAD of colon polyps and discuss the pitfalls and future directions for advancing the science. This review addresses the technical intricacies in a manner that physicians can comprehend to promote a better understanding of this novel application.