Skip to main content

Bae-Sun Seo

Yonsei University · Medicine

About the Lab

Professor Bae-Sun Seo's research lab specializes in medical artificial intelligence, focusing on the integration of deep learning and large language models (LLMs) into radiology and neurology for improved diagnostic accuracy and clinical decision support. The lab investigates multimodal AI applications—particularly using medical images and clinical text—to enhance the detection and quantification of neurological and reproductive health conditions, such as temporal lobe epilepsy, Parkinson’s disease, and early menopause. Key research directions include automated intracranial volume segmentation, AI-driven assessment of brain abnormalities, and the optimization of AI performance across varying input modalities and clinical contexts. The lab also emphasizes translational validation of AI models in real-world clinical settings to support personalized medicine and precision diagnostics.

medical AIneuroimaginglarge language modelsdeep learningclinical decision support

Research Overview

Papers
35
Total Citations
229
Papers (5y)
34
Primary Field
Medicine

Research Output Trend

Figures are computed from collected data and may differ slightly.

Publications per year (5y)
34total
2018
2023
2024
2025
2026
Citations per year (5y)
199total
20182023202420252026

Selected Papers

15
1
Article|84 citations·2024
Comparing Diagnostic Accuracy of Radiologists versus GPT-4V and Gemini Pro Vision Using Image Inputs from Diagnosis Please Cases
Pae Sun Suh, Woo Hyun Shim, Chong Hyun Suh, Hwon Heo, Chae Ri Park, Hye Joung Eom, Kye Jin Park, Jooae Choe, Pyeong Hwa Kim, Hyo Jung Park, Yura Ahn, Ho Young Park
SJR Q1Radiology

= .02). Radiologists (range, 45%-88%) outperformed the LLMs at T1 (range, 24%-75%) in most subspecialties. Conclusion Using direct radiologic image inputs, GPT-4V and Gemini Pro Vision showed improved diagnostic accuracy with increasing temperature settings. Although GPT-4V slightly underperformed compared with radiologists, it nonetheless demonstrated promising potential as a supportive tool in diagnostic decision-making. © RSNA, 2024 See also the editorial by Nishino and Ballard in this issue.

Health InformaticsMedicine
2
Article|41 citations·2024
Comparing Large Language Model and Human Reader Accuracy with New England Journal of Medicine Image Challenge Case Image Inputs
Pae Sun Suh, Woo Hyun Shim, Chong Hyun Suh, Hwon Heo, Kye Jin Park, Pyeong Hwa Kim, Se Jin Choi, Yura Ahn, Sohee Park, Ho Young Park, Na Eun Oh, Min Woo Han
SJR Q1Radiology

< .001). Text input length affected LLM accuracy (odds ratio range, 3.2 [95% CI: 1.9, 5.5] to 6.6 [95% CI: 3.7, 12.0]). Conclusion LLMs demonstrated substantial accuracy with text and image inputs, outperforming a medical student. However, their accuracy decreased with shorter text lengths, regardless of image input. © RSNA, 2024

Health InformaticsMedicine
3
Article|30 citations·2015
Effects of Smoking on Menopausal Age: Results From the Korea National Health and Nutrition Examination Survey, 2007 to 2012
Hee Jung Yang, Pae Sun Suh, Soo Jeong Kim, Soon Young Lee
SJR Q2Journal of Preventive Medicine and Public HealthOA

OBJECTIVES: Decreased fertility and impaired health owing to early menopause are significant health issues. Smoking is a modifiable health-related behavior that influences menopausal age. We investigated the effects of smoking-associated characteristics on menopausal age in Korean women. METHODS: This study used data from the Korea National Health and Nutrition Examination Survey from 2007 to 2012. Menopausal age in relation to smoking was analyzed as a Kaplan-Meier survival curve for 11 510 wom

Endocrinology, Diabetes and MetabolismMedicine
4
Article|14 citations·2024
Improving Diagnostic Performance of MRI for Temporal Lobe Epilepsy With Deep Learning-Based Image Reconstruction in Patients With Suspected Focal Epilepsy
Pae Sun Suh, Ji Eun Park, Yun Hwa Roh, Seon‐Ok Kim, Mina Jung, Yong Seo Koo, Sang‐Ahm Lee, Yangsean Choi, Ho Sung Kim
SJR Q1Korean Journal of RadiologyOA

The use of 1.5-mm MRI + DLR enhanced the performance of MRI in diagnosing TLE, particularly in hippocampal evaluation, because of improved depiction of hippocampal abnormalities and enhanced image quality.

Psychiatry and Mental healthMedicine
5
Article|13 citations·2023
Development and validation of a deep learning-based automatic segmentation model for assessing intracranial volume: comparison with NeuroQuant, FreeSurfer, and SynthSeg
Pae Sun Suh, Wooseok Jung, Chong Hyun Suh, Jinyoung Kim, Jio Oh, Hwon Heo, Woo Hyun Shim, Jae‐Sung Lim, Jae‐Hong Lee, Ho Sung Kim, Sang Joon Kim
SJR Q2Frontiers in NeurologyOA

Background and purpose: To develop and validate a deep learning-based automatic segmentation model for assessing intracranial volume (ICV) and to compare the accuracy determined by NeuroQuant (NQ), FreeSurfer (FS), and SynthSeg. Materials and methods: This retrospective study included 60 subjects [30 Alzheimer's disease (AD), 21 mild cognitive impairment (MCI), 9 cognitively normal (CN)] from a single tertiary hospital for the training and validation group (50:10). The test group included 40 sub

Psychiatry and Mental healthMedicine
6
Review|8 citations·2024
Substituting with alternative iodinated contrast medium to prevent recurrent adverse drug reactions associated with its use: a meta-analysis
Su Jin Lim, Pae Sun Suh, Chong Hyun Suh, Pyeong Hwa Kim, Kye Jin Park, Hyo Jung Park, Choong Wook Lee
SJR Q1European Radiology
NephrologyMedicine
7
Article|7 citations·2024
Effectiveness of microvascular flow imaging for radiofrequency ablation in recurrent thyroid cancer: comparison with power Doppler imaging
Pae Sun Suh, Jung Hwan Baek, Jae Ho Lee, Sae Rom Chung, Young Jun Choi, Ki‐Wook Chung, Tae Yong Kim, Jeong Hyun Lee
SJR Q1European Radiology
Endocrinology, Diabetes and MetabolismMedicine
8
Article|5 citations·2023
Reduction of Radiation Dose to Eye Lens in Cerebral 3D Rotational Angiography Using Head Off-Centering by Table Height Adjustment: A Prospective Study
Jae‐Chan Ryu, Jong-Tae Yoon, Byung Jun Kim, Mi Hyeon Kim, Eun Ji Moon, Pae Sun Suh, Yun Hwa Roh, Hye Hyeon Moon, Boseong Kwon, Deok Hee Lee, Yunsun Song
SJR Q1Korean Journal of RadiologyOA

The lens radiation dose was significantly affected by table height adjustment during 3D-RA. Intentional head off-centering by elevation of the table is a simple and effective way to reduce the lens dose in clinical practice.

Radiology, Nuclear Medicine and ImagingMedicine
9
Article|5 citations·2024
Deep Learning–Based Algorithm for Automatic Quantification of Nigrosome-1 and Parkinsonism Classification Using Susceptibility Map–Weighted MRI
Pae Sun Suh, Hwan Heo, Chong Hyun Suh, MyeongOh Lee, Soohwa Song, Dong Hoon Shin, Sungyang Jo, Sun Ju Chung, Hwon Heo, Woo Hyun Shim, Ho Sung Kim, Sang Joon Kim
SJR Q1American Journal of NeuroradiologyOA

Our deep learning-based model proves rapid, accurate automatic quantification of nigral hyperintensity, facilitating IPD diagnosis, symptom severity prediction, and patient stratification for personalized therapy. Further study is warranted to validate the findings across various clinical settings.

NeurologyMedicine
10
Article|4 citations·2025
Evaluating diagnostic accuracy of large language models in neuroradiology cases using image inputs from JAMA neurology and JAMA clinical challenges
Ahmed Albaqshi, Ji Su Ko, Chong Hyun Suh, Pae Sun Suh, Woo Hyun Shim, Hwon Heo, Chang-Yun Woo, Hyung Park
SJR Q1Scientific ReportsOA

This study assesses the diagnostic performance of six LLMs -GPT-4v, GPT-4o, Gemini 1.5 Pro, Gemini 1.5 Flash, Claude 3.0, and Claude 3.5-on complex neurology cases from JAMA Neurology and JAMA, focusing on their image interpretation abilities. We selected 56 radiology cases from JAMA Neurology and JAMA (from May 2015 to April 2024), rephrasing the text and reshuffling multiple-choice answer. Each LLM processed four input types: original quiz with images, rephrased text with images, rephrased tex

Health InformaticsMedicine
11
Article|4 citations·2024
Diagnosis of Unruptured Intracranial Aneurysms Using Proton-Density Magnetic Resonance Angiography: A Comparison With High-Resolution Time-of-Flight Magnetic Resonance Angiography
Pae Sun Suh, Seung Chai Jung, Hye Hyeon Moon, Yun Hwa Roh, Yunsun Song, Minjae Kim, Jung-Bok Lee, Keum Mi Choi
SJR Q1Korean Journal of RadiologyOA

PD-MRA outperformed HR-MRA in diagnostic accuracy and demonstrated almost perfect inter-reader consistency in identifying intracranial aneurysms among patients with lesions initially indeterminate on CTA or MRA.

NeurologyMedicine
12
Article|3 citations·2026
Insufficient reporting quality in large language model studies in the field of radiology
Pae Sun Suh, So Yeong Jeong, Daiju Ueda, Woo Hyun Shim, Hwon Heo, Chang-Yun Woo, Hyungjun Park, Chong Hyun Suh
SJR Q1Insights into ImagingOA

OBJECTIVES: Our systematic review aimed to evaluate the quality of reporting in research articles involving LLMs in the radiology field. MATERIALS AND METHODS: After searching the PubMed-MEDLINE and EMBASE databases, a total of 246 eligible studies published between November 30, 2022, and December 31, 2024, were included. The analysis assessed the percentage of studies adhering to key elements required for LLM research, based on the MInimum reporting items for CLear Evaluation of Accuracy Report

Health InformaticsMedicine
13
Article|2 citations·2018
Three cases of pancreatic pseudocysts associated with dorsal pancreatic agenesis
Pae Sun Suh, Jei Hee Lee, Jeong‐Sik Yu, Joo Hee Kim, Bohyun Kim, Hye Jin Kim, Jimi Huh, Bohyun Kim, Dakeun Lee
Radiology Case ReportsOA

Agenesis of the dorsal pancreas (ADP) is an extremely rare congenital anomaly. Human pancreas is formed by ventral and dorsal endodermal buds of the foregut endoderm. The dorsal bud forms the upper part of the head, neck, body, and tail of the pancreas and the ventral bud generates most of the head and uncinate process. ADP is derived from the embryologic failure of the dorsal pancreatic bud to form the pancreatic body and tail. ADP can be related to some diseases and conditions such as pancreat

SurgeryMedicine
14
Article|1 citations·2018
Assessment of Venographic Abnormalities during Replacement of Dysfunctional Tunneled Hemodialysis Catheters and Outcome of Endovascular Salvage Techniques
Pae Sun Suh, Seon Young Park, Jinoo Kim, Chang Kwon Oh, Su Hyung Lee, Je Hwan Won
SJR Q4Journal of the Korean Society of RadiologyOA

Purpose: To assess the venographic findings of central venous abnormalities before exchanging dysfunctional tunneled hemodialysis catheters and the outcome of endovascular salvage techniques. Materials and Methods: A total of 110 episodes of tunneled hemodialysis catheter dysfunction in 78 patients undergoing catheter-directed hemodialysis treatment from January 2011 to December 2015 were retrospectively evaluated. Venography was performed before catheter exchange, and the following procedures w

Emergency Medical ServicesHealth Professions
15
Article|1 citations·2024
Ethanol Ablation of Ranulas and Risk Factor Analysis for Recurrence
Pae Sun Suh, Jeong Hyun Lee, Yun Hwa Roh, Hye Hyun Moon, Sae Rom Chung, Min Su Kwon, Young Jun Choi, Yoon Se Lee, Jung Hwan Baek, Seung‐Ho Choi
SJR Q1JAMA Otolaryngology–Head & Neck SurgeryOA

Importance: Ethanol ablation (EA) was shown to be safe and effective for treating ranula, but few studies have assessed long-term outcomes and recurrence of ranula after EA. Objective: To evaluate the long-term outcomes and the risk factors for recurrence and receipt of subsequent surgery in patients who underwent treatment with EA for ranula. Design, Setting, and Participants: This case-series study was conducted at a single tertiary hospital and assessed patients who were treated with EA betwe

SurgeryMedicine

Research Areas

Health InformaticsRadiology, Nuclear Medicine and ImagingEndocrinology, Diabetes and MetabolismNeurologyOtorhinolaryngologyPsychiatry and Mental health

Dive deeper into Bae-Sun Seo's research on Nubint

Open this lab's papers in the app to read with AI, summarize, and cite in your writing.