Identification of Uterine Fibroids in Medical Pictures Employing Deep Neural Networks
2024 International Conference on Emerging Systems and Intelligent Computing (ESIC)(2024)
摘要
Uterine fibroids, commonly known as leiomyomas, are prevalent non-cancerous pelvic tumors in fertile women. These growths, comprising muscle and fibrous tissue, vary in size. Often asymptomatic, fibroids can be challenging to detect through CT scans and ultrasounds if small in size. Due to its high specificity and sensitivity, as well as its affordability and greater accessibility compared to CT and MRI exams, ultrasonography is currently the initial form of imaging used for the clinical diagnosis of uterine fibroids. The primary issue is misinterpretation of big and subplasmic fibroids, as well as pelvic and adnexal tumours. In order to do this, an automated method that utilises deep learning classification models— namely, VGG 16—for identifying the presence of fibroids in the uterus has been proposed in this study. Our approach has demonstrated a 97.5% accuracy rate which is the best model in predicting the uterine fibrosis class based on imaging data as compared with earlier studies.
更多查看译文
关键词
Deep learning,fibroids,ultrasonography,VGG 16,dicom
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要