Prostate Image Segmentation using Video Internet of Things (VIoT) applications in Biomedical Engineering depending on Deep Learning algorithms pre and during COVID-19 Pandemic

Abstract

Abstract This study highlights the role of imaging modalities through prostate image segmentation, using various algorithms depending on segmentation accuracy, and (VIoT) impact on improving imaging, via analyzing relevant articles to prostate segmentation during 2018–2021, from Springer, Science Direct, MDPI, IEEE, Nature Portfolio, Hindawi, with Taylor and Francis pre and during COVID-19. This study deals with 20 articles. Findings illustrated MRI was involved in (90%) of the articles in pre-COVID-19, while during COVID-19 declined to (60%). Furthermore, CNN algorithm was the most dependent method for prostate segmentation which was (50%) of the articles rather than other models. Whereas (80%) of the articles were depended on (DSC). In conclusion, the (VIoT) shows a significant role in all imaging modalities specifically MRI due to the real-time imaging. COVID-19 had impact on prostate segmentation research with the publishers was considerable in pre and during the pandemic. In addition, the best-utilized imaging modality was MRI due to its high image quality and ease applicable for (VIoT). Nevertheless, there is no study considered transabdominal ultrasound database as imaging modality for prostate segmentation. On the other hand, the segmentation performance referred to (DSC) that has a significant influence on prostate image segmentation Quality and performance.