Transfer Learning-Based Detection of Endometrial Cancer Lesion Regions on MRI Images
- Resource Type
- Conference
- Authors
- Mao, Wei; Xiong, Liu; Li, Zhifang; Lin, Yongping
- Source
- 2022 IEEE 2nd International Conference on Software Engineering and Artificial Intelligence (SEAI) Software Engineering and Artificial Intelligence (SEAI), 2022 IEEE 2nd International Conference on. :46-49 Jun, 2022
- Subject
- Computing and Processing
Location awareness
Magnetic resonance imaging
Computational modeling
Transfer learning
Object detection
Learning (artificial intelligence)
Lesions
endometrial cancer
MRI
object detection
transfer learning
- Language
Accurate localization of the lesion region on mag-netic resonance imaging (MRI) images of patients with endome-trial cancer (EC) facilitates subsequent diagnosis by computer-aided diagnostic systems or clinicians. MRI images of 294 patients with early-stage EC were retrospectively studied. Four classical object detection models (SSD, Faster R-CNN, CenterNet, YOLOv4) were trained by using the transfer learning approach. The detection performances of each model for the lesion region were compared and analyzed. The results show that the SSD model has the best detection performance in the test dataset with the AP of 99.52 % and F1 score of 0.99 for detecting uterus and the AP of 96.12% and F1 score of 0.91 for detecting tumor at an IoU threshold of 0.5. The SSD model could be a useful tool for the implementation of a computer-aided diagnosis system for EC patients on MRI images.