188宝金博页面版

Articles © The authors | Journal compilation © Gastroenterol Res and Elmer Press Inc™ | https://gr.elmerpub.comThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, including commercial use, provided the original work is properly cited136Original Article Gastroenterol Res. 2026;19(3):136-146Computed Tomography-Based Assessment of Sarcopenia and Disease Progression in Pancreatic Ductal Adenocarcinoma: A Radiomics and Machine Learning ApproachJean-Philippe R. Krieg a , Luisa Gallee a, b , Daniel Wolf a, b, c , Konstantin Muller a , Michael Goetz a, b, d , Thomas J. Ettrich d, e, f , Thomas Seufferlein d, e, f , Christopher Kloth a , Meinrad Beer a, d , Daniel Vogele a, gAbstractBackground: Sarcopenia is a known negative prognostic factor in oncology and is frequently observed in patients with pancreatic ductal adenocarcinoma (PDAC). Computed tomography (CT) enables lon-gitudinal muscle assessment and may provide additional prognostic information. This study aims to assess their association with prog-nosis in pancreatic cancer and explore the diagnostic and prognostic value of CT radiomics.Methods: A retrospective single-center study included 62 patients with primary PDAC who underwent at least three abdominal CT scans: baseline (t0), 3 months (t1), 6 months (t2), and, in 35 patients, 12 months (t3). CT-based sarcopenia was assessed using the psoas muscle index (PMI) based on reference cutoffs and cohort-specific sex-specific quartiles. Skeletal muscles at the L3 level were semi-au-tomatically segmented. Radiomic features of the psoas were extracted and analyzed using k-nearest neighbor, decision tree, and random for-est models. Prognostic relevance was evaluated using logistic regres-sion and feature selection via least absolute shrinkage and selection operator (LASSO) regression. Tumor progression was assessed radio-logically according to RECIST 1.1 criteria.Results: CT-based sarcopenia prevalence was 45.3% using reference-based PMI thresholds. PMI declined significantly from baseline to t1 and remained stable thereafter, with women exhibiting consistently lower values. Outcome analysis showed a higher proportion of dis-ease progression at t1 in sarcopenic patients using reference cutoffs, whereas cohort-specific quartiles demonstrated no consistent differ-ences. Random forest models predicted sarcopenia with up to 0.73 accuracy and receiver operating characteristics area under the curve (ROC-AUC) of 0.81. LASSO regression identified the psoas short axis and cross-sectional area as the most informative features. Logis-tic regression using baseline radiomic features predicted disease pro-gression status at 12 months with 0.85 accuracy, weighted F 1 0.841, and AUC 0.823. Interobserver agreement for psoas measurements was high (r = 0.86).Conclusion: Longitudinal CT-based assessment of PMI demon-strates progressive sarcopenia within the studied PDAC cohort, with sex-specific declines. Radiomic analysis of skeletal muscle provides complementary information and predictive insights, highlighting their potential to enhance the characterization of muscle status and its association with disease course in patients able to undergo repeated imaging.Keywords: Sarcopenia; Pancreatic cancer; CT radiomics; Machine learning; Tumor progressionIntroductionSarcopenia and clinical relevance in oncologySarcopenia is a progressive, generalized loss of muscle mass and strength. According to the European Working Group on Sarcopenia in Older People (EWGSOP), it can be classified as Manuscript submitted February 3, 2026, accepted April 16, 2026Published online June 16, 2026a Department of Diagnostic and Interventional Radiology, University Hospital Ulm, 89081 Ulm, Germanyb XAIRAD—Artificial Intelligence in Experimental Radiology, University Hospital Ulm, 89081 Ulm, Germanyc Institute for Media Informatics, Visual Computing Group, University of Ulm, Germanyd i2SouI—Innovative Imaging in Surgical Oncology Ulm, University Hospital Ulm, 89081 Ulm, Germanye Comprehensive Cancer Center Ulm (CCCU), University Hospital Ulm, 89081 Ulm, Germanyf Department of Internal Medicine I, University Hospital Ulm, 89081 Ulm, Germanyg Corresponding Author: Daniel Vogele, Department of Diagnostic and Inter-ventional Radiology, University Hospital Ulm, 89081 Ulm, Germany. Email: daniel.vogele@uniklinik-ulm.dedoi: https://doi.org/10.14740/gr2132 Gastroenterology Research 1918-2805 (print), 1918-2813 (online)