The effect of CT texture-based analysis using machine learning approaches on radiologists' performance in differentiating focal-type autoimmune pancreatitis and pancreatic duct carcinoma

• Published in: Japanese journal of radiology Vol. 40; no. 11; pp. 1156 - 1165
• Main Authors: Anai, Kenta, Hayashida, Yoshiko, Ueda, Issei, Hozuki, Eri, Yoshimatsu, Yuuta, Tsukamoto, Jun, Hamamura, Toshihiko, Onari, Norihiro, Aoki, Takatoshi, Korogi, Yukunori
• Format: Journal Article
• Language: English
• Published: Singapore Springer Nature Singapore 01.11.2022; Springer Nature B.V
• Subjects: Autoimmune Pancreatitis - diagnostic imaging; Carcinoma, Pancreatic Ductal - diagnostic imaging; Classifiers; Computed tomography; Data compression; Diagnosis, Differential; Diagnostic systems; Humans; Image compression; Image contrast; Image enhancement; Imaging; Machine Learning; Medicine; Medicine & Public Health; Nuclear Medicine; Original; Original Article; Pancreas; Pancreatic carcinoma; Pancreatic Ducts; Pancreatic Neoplasms; Pancreatic Neoplasms - diagnostic imaging; Pancreatitis; Principal components analysis; Radiologists; Radiology; Radiotherapy; Retrospective Studies; Support vector machines; Texture; Tomography, X-Ray Computed - methods; CT; Pancreatic duct carcinoma; Texture analysis; Autoimmune pancreatitis; Machine learning

Purpose To develop a support vector machine (SVM) classifier using CT texture-based analysis in differentiating focal-type autoimmune pancreatitis (AIP) and pancreatic duct carcinoma (PD), and to assess the radiologists’ diagnostic performance with or without SVM. Materials and methods This retrospe...