Wei-Chih Liao

Wei-Chih Liao, Dr.

Professor of Medicine / Attending physician
Organization: College of Medicine, National Taiwan University
Nationality: Taiwan
Email: david.ntuh@gmail.com

Brief Introduction

Dr. Wei-Chih Liao, M.D., Ph.D., FASGE, is a Professor of Medicine at the National Taiwan University, College of Medicine and an Attending Gastroenterologist in the Department of Internal Medicine at National Taiwan University Hospital. He earned his M.D. from National Taiwan University College of Medicine in 2005 and subsequently completed both his M.Sc. and Ph.D. in Epidemiology at National Taiwan University.

Specialty

  • Clinical and translational research on pancreatic cancer and pancreatitis
  • Biliary and pancreatic endoscopy
  • Medical application of artificial intelligence

Education

  • National Taiwan University Institute of Epidemiology and Preventive Medicine, PhD
  • National Taiwan University Institute of Epidemiology and Preventive Medicine, Master
  • National Taiwan University College of Medicine, MD

Experience

  • 2020 – Present: Professor, National Taiwan University College of Medicine
  • 2005 – Present: Attending physician, Department of internal medicine, National Taiwan University Hospital
  • 2001 - 2003: Fellowship, National Taiwan University Hospital
  • 1998 - 2001: Residency, National Taiwan University Hospital
  • 1995 - 1996: Internship, National Taiwan University Hospital

Selected Publications

  1. Yang CH, Lung YY, Chen CC, Lo IH, Lien CJ, Wu WK, Wu MS, Liao WC (co-corresponding author), Lin JD. Peripheral immune landscape in pancreatic ductal adenocarcinoma reveals expansion of effector states with disease progression. iScience 2026 Feb 14;29(3):115034. doi: 10.1016/j.isci.2026.115034.
  2. Chen PT, Chang D, Chen Y, Wang P, Yeh AY, Liu KL, Wu MS, Liao WC (co-corresponding author), Weichung Wang. Pancreatic Cancer Diagnosis on Unenhanced CT with Deep Learning for Opportunistic Diagnosis. Radiology Advances 2026, umag017, https://doi.org/10.1093/radadv/umag017.
  3. AY Yeh, Chang D, Wang P, Chen Y, Liu KL, Roth H, Yen HH, Chen DY, Chen PT, Liao WC (senior author), Wang W. Artificial intelligence for early detection of pancreatic cancer in prediagnostic and diagnostic computed tomography examinations: A multicenter retrospective case-control study. Diagnostic and Interventional Imaging 2026 Jan 31:S2211-5684(26)00007-0. doi: 10.1016/j.diii.2026.01.007.
  4. Han ML, Wu JW, Tu CH, Chen CC, Liao WC. Usefulness of a novel wired magnetic-assisted capsule endoscopy in stable patients with acute upper gastrointestinal bleeding: A prospective cohort study. Journal of the Formosan Medical Association. 2025 Sep 29:S0929-6646(25)00519-4.
  5. TY Lin, CC Chen, YT Kuo, WC Liao (corresponding author). Endoscopic and novel approaches for evaluation of indeterminate biliary strictures. Journal of the Formosan Medical Association. 2025 May 7:S0929-6646(25)00266-9.
  6. Chang WY, Liao WC (equal contribution with Chang LC), Chang LC, Lin HH, Wei PY, Wu HC, Chiu HM, Wu MS. Comparison of Adenoma Detection Rate Between Three-dimensional and Standard Colonoscopy: A Multicenter Randomized Controlled Trial. Endoscopy 2025 Jan 7. doi: 10.1055/a-2510-8759.
Unveiling Pancreatic Cancer with Artificial Intelligence
Wei-Chih Liao, MD, PhD, FASGE
College of Medicine, National Taiwan University
Department of Internal Medicine, National Taiwan University Hospital

Pancreatic cancer (PC) is the most lethal cancer and ranks as the seventh and third leading cause of cancer deaths in Taiwan and the US, respectively. Computed tomography (CT) is the major modality for detecting PC, but small PCs are often obscure on CT with approximately 40% of tumors less than 2 cm being missed. Furthermore, the diagnostic performance of CT is interpreter-dependent and influenced by radiologist workload and expertise. Novel tools that can facilitate detection of PCs on CT are urgently needed given that patient survival sharply shortens with tumor growth.

We have developed a novel artificial intelligence (AI)-empowered computer-aided detection/diagnosis (CAD) tool to supplement radiologists in detecting PCs. This fully automatic end-to-end CAD tool can determine whether CT images harbor PC and indicate the location of the PC without requiring manual image preprocessing. When tested with images obtained in real clinical practice at institutions throughout Taiwan from the National Health Insurance, PANCREASaver achieved 89.7% sensitivity and 91.4% accuracy in differentiating between PC patients and controls, supporting its accuracy and generalizability. Because of the unique ability to detect PCs that are difficult to detect with the naked eye, this CAD tool may supplement radiologists in interpreting abdominal CT images to reduce the miss rate and enhance early detection of PC.