Research
Research Field
Medical data exists in a wide variety of forms, including proteins and molecules (graphs), genomes and transcriptomes (natural language), medical images (multidimensional images), spectrums (frequency signals), heart rate and metabolomics variations (time-series signals), patient metadata and information on interactions between substances (multidimensional matrices). For personalized precision diagnostics, a comprehensive analysis of various modalities for individual patients is essential, and, above all, the basis for diagnosis must be presented clearly.Our laboratory is dedicated to the precise processing of such complex multimodal medical data and developing advanced foundational generative AI technologies that produce results intuitively understandable by the general public. We welcome the participation and collaboration of all researchers interested in developing cutting-edge multimodal AI technologies that are grounded in theoretical evidence and capable of validating their effectiveness in real clinical settings.
Multimodal AI (medical images/graphs/natural language/frequency/time-series/matrix data, etc.)
Generative explainable AI
Unsupervised anomaly detection AI
Causal/biomarker detection AI
Reliability assessment AI
Bias removal AI
3D/4D/time-series/natural language signal restoration AI
Keyword
Bioinformatics, AI, Medical ImagingIntensive Major
Publication
- Sanghyun Jo, Soohyun Ryu, Sungyub Kim, Eunho Yang, Kyungsu Kim#. "TTD: Text-Tag Distillation For Open-Vocabulary Classification and Segmentation." European Conference on Computer Vision (2024).
- Sanghyun Jo, Fei Pan, In-Jae Yu, and Kyungsu Kim#. "DHR: Enhancing Weakly Supervised Semantic Segmentation with Hierarchical Spatial-Class Rebalancing." European Conference on Computer Vision (2024).
- SungYub Kim, Kyungsu Kim#, and Eunho Yang#. "GEX: A flexible method for approximating influence via Geometric Ensemble." Advances in Neural Information Processing Systems 36 (2024).
- JungEun Kim*, Hangyul Yoon*, Geondo Park, Kyungsu Kim#, and Eunho Yang. "Data-Efficient Unsupervised Interpolation Without Any Intermediate Frame for 4D Medical Images." In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 11353-11364. 2024.
- Harim Kim*, Kyungsu Kim*, Seong Je Oh, Sungjoo Lee, Jung Han Woo, Jong Hee Kim, Yoon Ki Cha, Kyunga Kim, and Myung Jin Chung#. "AI-assisted Analysis to Facilitate Detection of Humeral Lesions on Chest Radiographs." Radiology: Artificial Intelligence 6, no. 3 (2024): e230094.