Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{174547,
        author = {Lagudu Kalyani and O.Hema Latha and Abhishek Sahu and D.Rama Sujith and G.Bobby Yadav and S.A.Bhavani},
        title = {Detection of Fetal Cardiac Structural Abnormalities},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {11},
        number = {11},
        pages = {3277-3286},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=174547},
        abstract = {The diagnosis of CHD requires precise localization and detection methods for heart abnormalities through ultrasound image analysis because this condition stands as the main congenital heart defect. Active CHD fetal heart identification before birth protects fetal survival while creating opportunities for necessary medical care. The diagnosis method based on traditional ultrasound depends heavily on expert manual reading of images which produces inconsistent results because of varying professional expertise levels. The proposed framework unites U-Net architecture together with YOLOv7 detection for a new system. The U-Net system obtains spatial details to perform segmentation tasks along with YOLOv7 functionality for real-time object detection. The integration between U-Net architecture and YOLOv7 detection produces better CHD diagnostic results through simultaneous precise visualization of image anomalies. Adaptive fuzzy attention mechanisms serve to enhance localization by enabling the model to focus on significant parts of the images. The new detection method demonstrates higher accuracy results than established approaches as experimentally validated results show. This framework presents the capability to enhance prenatal diagnosis while simultaneously decreasing the professional workload and enabling faster intervention. The model requires improvements for real-time clinical use and the existing dataset must grow for better generalization effectiveness.},
        keywords = {Congenital heart disease, Fetal ultrasound, U-Net, YOLOv7, Deep learning, Image segmentation, Real-time detection.},
        month = {April},
        }