Meeting Banner
Abstract #1399

Early Detection of Bladder Cancer in MRI Using Deep Learning Segmentation and Rule-Based Classification: a Pilot study

Ki Choon Sim1,2, Min Ju Kim1,2, Deuk Jae Sung1,2, Beom Jin Park1,2, Na Yeon Han1,2, Yeo Eun Han1,2, Keewon Shin2,3, and Taehun Kim2,3
1Radiology, Korea University Anam Hospital, Seoul, Korea, Republic of, 2Advanced Medical Imaging Institute, Korea University Anam Hospital, Korea University College of Medicine, Seoul, Korea, Republic of, 3Artificial Intelligence, Korea University Anam Hospital, Seoul, Korea, Republic of

Synopsis

Keywords: Diagnosis/Prediction, Bladder

Motivation: Bladder cancer is often diagnosed at advanced stages due to limited early detection methods, leading to poor patient outcomes.

Goal(s): To evaluate the efficacy of deep learning segmentation combined with rule-based classification in the early detection of bladder cancer using MRI.

Approach: A study using T2WI of MRI performed for bladder cancer, analyzed using nnU-Net for segmentation and a rule-based classification system based on segmented voxel count. A total of 260 MRI datasets, including 120 bladder cancer cases, were used.

Results: The author's approach demonstrated a sensitivity of 86.4% and specificity of 95% in detecting early-stage bladder cancer (AUROC, up to 0.88).

Impact: Improved early detection of bladder cancer will enable early treatment, which will have a positive impact on treatment outcomes.

How to access this content:

For one year after publication, abstracts and videos are only open to registrants of this annual meeting. Registrants should use their existing login information. Non-registrant access can be purchased via the ISMRM E-Library.

After one year, current ISMRM & ISMRT members get free access to both the abstracts and videos. Non-members and non-registrants must purchase access via the ISMRM E-Library.

After two years, the meeting proceedings (abstracts) are opened to the public and require no login information. Videos remain behind password for access by members, registrants and E-Library customers.

Click here for more information on becoming a member.

Keywords