Professor Kim Youngwoo's Team at Kumoh National Institute of Technology Selected for 'Early Accept' at Prestigious Medical AI Conference MICCAI 2026

AI Model Combining MRI and Clinical Data Enables Non-Invasive Identification of High-Risk Polycystic Kidney Disease Patients
Recognized Among Top 9% of Research Papers Worldwide

Kumoh National Institute of Technology announced on June 10 that the research team led by Professor Kim Youngwoo of the Department of Computer Engineering had their research paper selected for 'Early Accept' at MICCAI 2026, the most prestigious international conference in the field of medical artificial intelligence (AI).

'Early Accept' is the highest level of acceptance, granted to papers recognized for their originality and completeness, as it is confirmed for publication immediately at the initial review stage without a rebuttal process.

This year, a total of 4,601 papers were submitted to MICCAI, and only about 9% were selected as Early Accept.

From the left) Professor Kim Youngwoo (Advisor & Corresponding Author) (Professor, Department of Computer Engineering), Researcher Oybek Valiyev (First Author) (Master's Program, Department of Software Engineering), Researcher Cho Junbeom (Co-author) (Master's Program, Department of Software Engineering)
Photo by Kumoh National Institute of Technology

The selected paper, titled "Radiogenomics-Driven Hierarchical Multimodal Stacking for Non-Invasive Triage of High-Risk PKD1-Truncating Genotypes in ADPKD," proposed an artificial intelligence framework that enables the identification of high-risk genotypes of autosomal dominant polycystic kidney disease (ADPKD) using only MRI images and basic clinical information, without the need for genetic testing.

The research team developed a hierarchical multimodal AI model that combines features extracted from three-dimensional MRI images with clinical information. In validation involving a cohort of 414 patients from eight hospitals in the United States, the model demonstrated superior performance compared to approaches using imaging or clinical data alone.

In particular, the study was recognized for suggesting the potential of this AI framework as a non-invasive decision support tool for early identification of high-risk patients, even in environments where genetic testing is costly or less accessible.

Professor Kim Youngwoo stated, "This research demonstrates that combining imaging and clinical data can provide early risk stratification opportunities to patients with limited access to genetic testing. We will continue to expand our research in medical imaging AI with our students."

This study was supported by the National Research Foundation of Korea (NRF) and the Korea Health Industry Development Institute (KHIDI). MICCAI 2026 will be held in Strasbourg, France, in September.

The fact that a research team from a regional national university has been recognized on a world-class medical AI stage is a meaningful achievement, highlighting new possibilities for the research competitiveness of regional universities.

This content was produced with the assistance of AI translation services.