Deep learning analysis of magnetic resonance imaging accurately detects early-stage perihilar cholangiocarcinoma in patients with primary sclerosing cholangitis.

• Tác giả: Andreas Abildgaard, Ida Bjoerk, Ty S Diwan, John E Eaton, Bradley J Erickson, Shahriar Faghani, Trine Folseraas, Gregory J Gores, Aliya F Gulamhusein, Julie K Heimbach, Sumera I Ilyas, Kartik Jhaveri, Kristin K Jorgensen, Tom H Karlsen, Nicholas F LaRusso, Konstantinos N Lazaridis, Anne Negard, Kosta Petrovic, Yashbir Singh, Byron Smith, Timucin Taner, Sudhakar K Venkatesh, Mette Vesterhus, Christopher L Welle
• Ngôn ngữ: eng
• Thông tin xuất bản: United States : Hepatology (Baltimore, Md.) , 2025
• Bộ sưu tập: NCBI
• ID: 726378

Tóm tắt

 BACKGROUND AND AIMS: Among those with primary sclerosing cholangitis (PSC), perihilar CCA (pCCA) is often diagnosed at a late-stage and is a leading source of mortality. Detection of pCCA in PSC when curative action can be taken is challenging. Our aim was to create a deep learning model that analyzed magnetic resonance imaging (MRI) to detect early-stage pCCA and compare its diagnostic performance with expert radiologists. APPROACH AND RESULTS: We conducted a multicenter, international, retrospective cohort study involving adults with large duct PSC who underwent contrast-enhanced MRI. Senior abdominal radiologists reviewed the images. All patients with pCCA had early-stage cancer and were registered for liver transplantation. We trained a 3D DenseNet-121 model, a form of deep learning, using MRI images and assessed its performance in a separate test cohort. The study included 398 patients (training cohort n=150
  test cohort n=248). pCCA was present in 230 individuals (training cohort n=64
  test cohort n=166). In the test cohort, the respective performances of the model compared to the radiologists were: sensitivity 87.9% versus 50.0%, p<
 0.001
  specificity 84.1% versus 100.0%, p<
 0.001
  area under receiving operating curve 86.0% versus 75.0%, p<
 0.001. Even when a mass was absent, the model had a higher sensitivity for pCCA than radiologists (91.6% vs. 50.6%, p<
 0.001) and maintained good specificity (84.1%). CONCLUSION: The 3D DenseNet-121 MRI model effectively detects early-stage pCCA in PSC patients. Compared to expert radiologists, the model missed fewer cases of cancer.