Double-Balanced Loss for Imbalanced Colorectal Lesion Classification

Abstract

Colorectal cancer has a high incidence rate in all countries around the world, and the survival rate of patients is improved by early detection. With the development of object detection technology based on deep learning, computer-aided diagnosis of colonoscopy medical images becomes a reality, which can effectively reduce the occurrence of missed diagnosis and misdiagnosis. In medical image recognition, the assumption that training samples follow independent identical distribution (IID) is the key to the high accuracy of deep learning. However, the classification of medical images is unbalanced in most cases. This paper proposes a new loss function named the double-balanced loss function for the deep learning model, to improve the impact of datasets on classification accuracy. It introduces the effects of sample size and sample difficulty to the loss calculation and deals with both sample size imbalance and sample difficulty imbalance. And it combines with deep learning to build the medical diagnosis model for colorectal cancer. Experimentally verified by three colorectal white-light endoscopic image datasets, the double-balanced loss function proposed in this paper has better performance on the imbalance classification problem of colorectal medical images.

Three-category colorectal lesion image automatic detection based on G-YOLOv8

“…Ahmet Karaman et al [2] introduced the Artificial Bee Colony (ABC) algorithm into the YOLO detection framework to optimize the original model algorithm framework, resulting in a significant improvement in detection rate. Chang Yu et al [3] proposed a dual-balanced loss function to address the issue of dataset imbalance in the staging of colorectal lesions and utilized the Faster R-CNN model [4] for detection, achieving corresponding improvements in average precision (AP) values for colorectal lesion classification. With the continuous development of research, the YOLO series of one-stage object detection models have demonstrated superiority in both precision and speed over other network models.…”