Philip Chikontwe

Daegu Gyeongbuk Institute of Science and Technology

Papers

Total Citations

H-Index

About

Philip Chikontwe is a researcher advancing the frontier of robot-assisted cardiac interventions through deep learning and medical image analysis. His primary research areas include real-time guidewire and catheter tracking, segmentation, and motion-constrained vision systems for micro-robot catheters used in minimally invasive cardiac surgery. Chikontwe’s major contributions center on developing deep convolutional neural networks—such as U-Net architectures with shape and motion constraints—to accurately track thin, flexible guidewire tips in fluoroscopic images, a task critical for precise robotic control. His most cited work, “Real-Time Tracking of Guidewire Robot Tips Using Deep Convolutional Neural Networks on Successive Localized Frames” (2019), has garnered 19 citations, demonstrating its impact on improving surgical stability. He has also pioneered adversarial augmentation techniques to enhance catheter segmentation when annotated data is scarce, as seen in his 2021 study. By addressing challenges of small target delineation and data scarcity, Chikontwe’s research directly supports safer, more accurate robot-assisted interventions, making him a notable contributor to the intersection of computer vision and surgical robotics.