Detection of Hysteroscopic Hysteromyoma in Real-Time Based on Deep Learning

Hysteromyoma is the most common benign tumor in women. By the age of 50, 70% of women have one or more uterine fibroids, and about 30% of them have symptoms and need treatment [1]. In hysteroscopic surgery, doctors' inexperience and fatigue will reduce the accuracy of hysteromyoma diagnosis. In this paper, a hybrid model based on YOLOv3(YOLO) Network and DCGAN network(DCGAN) is proposed to detect hysteromyoma in real time to assist doctors in diagnosis and reduce subjective randomness. The real-time detection speed of the hybrid model reaches 25FPS, and the accuracy rate reaches 91.73%, which meets the requirements of clinical application and improves the diagnosis efficiency of hysteromyoma.