Publications by authors named "Yen-Ming Kuan"
Bioengineering (Basel)
April 2024
Article Synopsis
- Local advanced rectal cancer (LARC) is challenging to treat due to its location and high rates of recurrence, making early detection essential; this study examines the potential of AI to analyze CT scans as a more efficient diagnostic tool compared to MRI.
- The researchers analyzed CT images of 1070 rectal cancer patients using AI models trained on 739 cases, employing advanced techniques for image filtering and classification to identify positive circumferential resection margins (CRM), facilitating the diagnosis of LARC.
- The AI models showed high accuracy with an AUC of 0.93 when using a soft voting system, indicating a better ability to predict poor clinical outcomes, including higher local recurrence rates and mortality trends among identified LARC cases.
