Events | MacsLAB

KSC 2025: Participation and Paper Presentation

Tue, 16 Dec 2025 00:00:00 +0000

MACS Lab attended KSC 2025 (Korea Software Congress) to present our work and discuss recent trends in AI and software systems with researchers across academia and industry.

KSC 2025 on-site photo (Dongjun Kang)

KSC 2025 on-site photo (Kyungsu Lee)

Presented Work

Dongjun Kang (first author), Kyungsu Lee (corresponding author)
Anatomy-Aware Distillation with Memory-Augmented SAM2 for Fracture Detection

The proposed approach combines a SAM2-based teacher-student structure with memory-augmented distillation to improve fracture detection accuracy and robustness under limited-data conditions.

Method overview and key results

Related Publication: /publication/0038-anatomy-aware-distillation-with-memory-augmented-sam2-for-fracture-detection-ksc2025/
Conference Website: https://www.kiise.or.kr/conference/KSC/2025/

RTSS 2025 (Boston): Conference Participation

Tue, 16 Dec 2025 00:00:00 +0000

MACS Lab attended RTSS 2025 (IEEE Real-Time Systems Symposium) in Boston and explored recent advances in real-time systems and AI system design through keynote talks, technical sessions, and poster presentations.

Main conference session at RTSS 2025

During the conference, we reviewed active topics in real-time intelligence and system-level reliability, and discussed potential directions for future cross-disciplinary research.

At the RTSS 2025 venue (Seoyeon Choi)

At the RTSS 2025 venue (Dongjun Kang)

At the RTSS 2025 venue (Kyungsu Lee)

The trip also provided meaningful exposure to the broader international research community and collaboration landscape.

Boston during RTSS 2025

Conference Website: https://2025.rtss.org/

KOSOMBE 2025 Fall: Participation and Presentations

Fri, 07 Nov 2025 00:00:00 +0000

MACS Lab participated in the Korean Society of Medical and Biological Engineering (KOSOMBE) 2025 Fall Conference and presented our recent work in medical AI and medical image analysis.

MACS Lab members at KOSOMBE 2025 Fall

Presentation 1

Dongjun Kang (first author), Kyungsu Lee (corresponding author)
Anatomy-Aware Distillation with Memory-Augmented SAM2 for Fracture Detection
KOSOMBE 2025 Fall (Poster 244)

This work proposes a memory-augmented knowledge distillation framework to transfer anatomy-aware knowledge from a large SAM2 model to a lightweight fracture detector.

Poster 244 presentation

Related Publication: /publication/0033-anatomy-aware-distillation-with-memory-augmented-sam2-for-fracture-detection/

Presentation 2

Sakang Hong (first author), Jun-Yung Kim, Kyungsu Lee (corresponding author)
SAM2-based Bayesian Prompt Adaptation for Cross-Modality Medical Segmentation
KOSOMBE 2025 Fall (Poster 247)

This work introduces a SAM2-based Bayesian prompt adaptation method (BayesPrompt) for cross-modality medical image segmentation under few-shot settings.

Poster 247 presentation

Related Publication: /publication/0034-sam2-based-bayesian-prompt-adaptation-for-cross-modality-medical-segmentation/

The conference offered productive exchanges with researchers in biomedical engineering, especially on practical deployment paths for medical AI.

Program: https://www.kosombe.or.kr/register/2025_fall/program/sub07.html

AIxMHC 2025 (Taiwan): Conference Participation and Paper Presentations

Mon, 13 Oct 2025 00:00:00 +0000

MACS Lab participated in AIxMHC 2025 (The 2nd International Conference on Artificial Intelligence for Medicine, Health, and Care) in Taichung, Taiwan. We presented two papers and discussed recent research trends in medical AI with international researchers.

AIxMHC 2025, Taichung, Taiwan

Presented Papers

Statistical Multi-Modal Fusion for Patient-Centric Medical Diagnosis Using DICOM (Poster)
Statistical Latent Manifold-Guided Framework for Generative Super-Resolution (Regular)

Q&A and technical discussion after the presentation

Poster presentation session

The conference provided a valuable opportunity to exchange ideas on practical medical AI applications and to expand possibilities for international collaboration.

Conference Website: https://chaoneng.github.io/aixmhc2025.github.io/

Participation and Presentations at KCC 2025

Wed, 02 Jul 2025 00:00:00 +0000

In July 2025, MACS Lab attended the Korea Computer Congress 2025 (KCC 2025) held at ICC Jeju. The team joined technical sessions, poster presentations, and workshops to share research outcomes and gain broader academic perspectives.

MACS Lab at KCC 2025

Professor Kyungsu Lee presented in the early-career researcher session under the topic “Applications of Trustworthy AI”, followed by active discussions with domestic and international participants.

Students in our integrated and graduate tracks also participated in poster and session programs, receiving valuable feedback and building new research connections.

Conference Website: https://www.kiise.or.kr/conference/kcc/2025/
Early-Career Session Program: https://kcc2025.kiise.or.kr/Proceedings/chart.asp