Risk Rating of Infantile Hemangioma using Deep Learning

Infantile hemangioma (IH) is a vascular anomaly observed in newborns, with potential severe complications if left undetected. Consequently, researchers have turned to artificial intelligence (AI) and digital imaging (DI) methods for detection, segmentation, and assessing the treatment response in IH cases. This paper conducts a systematic literature review (SLR) following the Kitchenham framework to scrutinize the utilization of AI and digital imaging techniques in IH applications. A total of 21 research articles spanning from 2014 to April 2024 were carefully selected and analyzed to address four key research questions: the issues solved in IH using AI and DI, the most-used AI and DI techniques, the best-performing technique in detecting IH, and the limitations and future directions in the various fields of IH. After an extensive review of the selected articles, it was found that 10 of the 21 articles focused on detecting IH, and 15 articles utilized AI. However, the best-performing technique in detecting IH employed DI. Additionally, the SLR offers insights and recommendations into future directions for IH applications.

show abstract

Risk Rating of Infantile Hemangioma using Deep Learning

Cited by 2 publications

References 15 publications

Infantile Hemangiomas Segmentation using U-Net and SegNet Architectures Models

Infantile Hemangiomas Segmentation using U-Net and SegNet Architectures Models

Digital Imaging and Artificial Intelligence in Infantile Hemangioma: A Systematic Literature Review

Contact Info

Product

Resources

About