SIGNATURE WORK
CONFERENCE & EXHIBITION 2022

Colorectal cancer (CRC) is a commonly seen fatal disease, which is still the third leading cause of cancer death among American women and the second leading cause of cancer death among American men. In recent years, there has been a surge in the deployment of machine learning techniques for CRC diagnosis. However, postoperative complications of CRC gain little attention. Studies have confirmed that anastomotic leak, one of the most frequent postoperative complications, is a risky factor affecting postoperative local recurrence rate and tumor-related survival rate in patients with CRC. This research aims to apply machine learning and deep learning methods to predict possible anastomotic leaks of postoperative CRC patients based on both clinical features (CFs) and CT scans (CTs). The logistic regression model is used for CFs and transfer learning based on classical VGG architecture is implemented on CTs.