A NOVEL BREAST CANCER DETECTION IN ULTRASOUND IMAGE BASED ON YOLO v4
Year: 2025
Type: article
Abstract: : Breast cancer is one of the leading causes of death among women worldwide, and early detection significantly improves survival rates. Ultrasound imaging is a widely used, noninvasive and cost-effective method for breast cancer diagnosis. However, manual interpretation of ultrasound images is time-consuming and prone to human error. This paper presents a computer aided diagnosis(CAD)-based YOLOv4 to segment pectoral muscle region and classify tumor cells. The model is optimized to achieve better accuracy in real-time tumor detection by fine-tuning parameters. We compare our method with state-of art criteria like U-Net interms of accuracy and we prove that our method yields better results. The proposed approach was tested on kaggle dataset and achieved high performance with an accuracy (Acc) of 99.8%.
Authors: K.Suseela, T.Indu, V.Sumasri, S.Lavanya, K.Anitha
Language: English
Topic: AI in cancer detection
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W6968222906
FWCI:
Cites:
Cited by:
Related to: