Medical AI Lab
Medical AI Lab
Kyung Hee University · Computer Science and Engineering
Professor
Prof. 김성태
Professor's page · publicationsAbout
김성태 교수 연구실(Medical AI Lab)은 경희대 컴퓨터공학과 소속으로, 의료 인공지능과 설명가능(explainable)·신뢰 가능한 인공지능을 연구한다. 의료 영상 분석 등에 적용 가능한 신뢰성 있는 딥러닝 모델 개발에 초점을 둔다.
Research keywords
Selected publications
- Learning to Abstain: Reliable Medical Image Segmentation With Rejection OptionIEEE Access · 2026
- Unsupervised domain adaptation for medical image segmentation using adaptogen-perturbationMedical Image Analysis · 2026
- Comparative validation of surgical phase recognition, instrument keypoint estimation, and instrument instance segmentation in endoscopy: Results of the PhaKIR 2024 challengeMedical Image Analysis · 2026
- LLaVA Needs More Knowledge: Retrieval Augmented Natural Language Generation with Knowledge Graph for Explaining Thoracic PathologiesAAAI 2025 · 2025
- Effects of mixed sample data augmentation on interpretability of neural networksNeural Networks · 2025
- Leveraging Textual Compositional Reasoning for Robust Change CaptioningAAAI 2026 · 2026
Other labs in this department
- KDST (KIST Data Science Team)
Kim Suhyun
Deep learning / AI: model compression & pruning, knowledge distillation, generative & diffusion models, AI safety
- Visual AI Lab
Kim Jung Uk
Computer Vision / Visual AI (audio-visual learning, sound source localization, monocular 3D object detection, incremental learning)
- Video Media Lab
Kim Hui Yong
Video/media coding and compression (learned image/video compression, video coding for machines, neural holographic video)
- Media Lab
Park Gwang Hoon
Video/Media Coding (Scalable Video Coding, HEVC, VVC, AV1, MPEG standards, point cloud compression, image/video stitching)
- Natural Language Processing Lab
Park Seong-Bae
Natural Language Processing (dialogue response generation, machine translation, question answering/generation, text summarization)
- System Security Lab (KHU-SysSec)
Park Cheol Jun
System security; 4G/5G cellular network and baseband security (testing, wireless signal analysis, attack/defense)