JaeSeong Hong
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
JaeSeong Hong has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Yu Rang Park
Department of Biomedical Systems Informatics, Yonsei University College of Medicine, Seoul, Republic of Korea
7 shared publications - Bo Kyu Choi
Department of Biomedical Systems Informatics Yonsei University College of Medicine Seoul 03722 Korea
2 shared publications - Chanyoung Ko
Department of Biomedical Systems Informatics, Yonsei University College of Medicine, Seoul, Republic of Korea
2 shared publications - Ho Heon Yang
Department of Biomedical Systems Informatics Yonsei University College of Medicine Seoul 03722 Korea
2 shared publications - Jin Yong Kim
Department of Biomedical Systems Informatics, Yonsei University College of Medicine, Seoul, Republic of Korea
2 shared publications - Jong Hyun Kim
Department of Biomedical Systems Informatics Yonsei University College of Medicine Seoul 03722 Korea
2 shared publications
Latest Publications
- Open Access
- 22 August 2026
- Published by Elsevier in Information Fusion
- Vol. 138
Geometry-guided unimodal-to-multimodal composition for incomplete and asynchronous clinical information fusion
- 22 August 2026
- Published by Elsevier in Information Fusion
- Vol. 138
Open AccessGeometry-guided unimodal-to-multimodal composition for incomplete and asynchronous clinical information fusion
- Open Access
- 24 June 2026
- Published by Elsevier in SSRN Electronic Journal
Geometry-guided unimodal-to-multimodal composition for incomplete and asynchronous clinical information fusion
Clinical prediction often requires fusing heterogeneous data sources such as coded patient context, lab or vital trajectories, and free-text documentation, but the set of observable modalities varies across care workflows and observation timepoints. We study this setting as an incomplete clinical information fusion problem rather than as explicit longitudinal sequence modeling. ETF-UML addresses the challenge by training modality-specific encoders independently while aligning them to a shared Equiangular Tight Frame (ETF) decision geometry, then composing only the observed modalities through metadata-conditioned fusion. This design targets robust unimodal-to-multimodal composition when modality acquisition is asynchronous, incomplete, and time-varying. We clarify the training and evaluation protocol and add representation analysis to illustrate how the shared geometry structures unimodal and fused spaces. On MIMIC-IV 30/90-day mortality and eICU 10-day in-hospital mortality, the reported aggregate results indicate consistent gains in AUROC and average performance over end-to-end fusion and missing-modality baselines.
- 24 June 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessGeometry-guided unimodal-to-multimodal composition for incomplete and asynchronous clinical information fusion
Clinical prediction often requires fusing heterogeneous data sources such as coded patient context, lab or vital trajectories, and free-text documentation, but the set of observable modalities varies across care workflows and observation timepoints. We study this setting as an incomplete clinical information fusion problem rather than as explicit longitudinal sequence modeling. ETF-UML addresses the challenge by training modality-specific encoders independently while aligning them to a shared Equiangular Tight Frame (ETF) decision geometry, then composing only the observed modalities through metadata-conditioned fusion. This design targets robust unimodal-to-multimodal composition when modality acquisition is asynchronous, incomplete, and time-varying. We clarify the training and evaluation protocol and add representation analysis to illustrate how the shared geometry structures unimodal and fused spaces. On MIMIC-IV 30/90-day mortality and eICU 10-day in-hospital mortality, the reported aggregate results indicate consistent gains in AUROC and average performance over end-to-end fusion and missing-modality baselines.
- Open Access
- 26 March 2025
- Published by Wiley in Advanced Intelligent Systems
- Vol. 7 (6)
Deep‐Learning Model for Central Nervous System Infection Diagnosis and Prognosis Using Label‐Free 3D Immune‐Cell Morphology in the Cerebrospinal Fluid
Early diagnosis and prognostication of a central nervous system (CNS) infection is essential. This study aims to use immune‐cell morphology to develop a deep‐learning model for this purpose. Overall, 1427 3D images of cerebrospinal fluid (CSF) immune cells from 14 patients with CNS infections are obtained using holotomography. The images are categorized into infection etiology groups (viral and non‐viral) and prognosis groups (based on the modified Rankin Scale score at discharge). A deep‐learning model is constructed to predict the etiology and prognosis of CNS infections using the immune‐cell morphology. Cell morphological features and spatial distribution of CSF immune cells differ significantly between patients in the viral and nonviral groups and between prognosis groups. The model yields areas under the receiver operating characteristic curve of 0.89 and 0.79 for the diagnosis and prognosis, respectively. As more cell images are used, the prediction and model robustness improve. With <10 cells, both tasks exhibit a nearly 100% predictive performance. After dividing the cells into eight shells, significant refractive index variations are observed. This is the first study to use CSF cell morphology for the diagnosis and prognostication of CSF infections. These findings can help improve patient outcomes.
- 26 March 2025
- Published by Wiley in Advanced Intelligent Systems
- Vol. 7 (6)
Open AccessDeep‐Learning Model for Central Nervous System Infection Diagnosis and Prognosis Using Label‐Free 3D Immune‐Cell Morphology in the Cerebrospinal Fluid
Early diagnosis and prognostication of a central nervous system (CNS) infection is essential. This study aims to use immune‐cell morphology to develop a deep‐learning model for this purpose. Overall, 1427 3D images of cerebrospinal fluid (CSF) immune cells from 14 patients with CNS infections are obtained using holotomography. The images are categorized into infection etiology groups (viral and non‐viral) and prognosis groups (based on the modified Rankin Scale score at discharge). A deep‐learning model is constructed to predict the etiology and prognosis of CNS infections using the immune‐cell morphology. Cell morphological features and spatial distribution of CSF immune cells differ significantly between patients in the viral and nonviral groups and between prognosis groups. The model yields areas under the receiver operating characteristic curve of 0.89 and 0.79 for the diagnosis and prognosis, respectively. As more cell images are used, the prediction and model robustness improve. With <10 cells, both tasks exhibit a nearly 100% predictive performance. After dividing the cells into eight shells, significant refractive index variations are observed. This is the first study to use CSF cell morphology for the diagnosis and prognostication of CSF infections. These findings can help improve patient outcomes.
- Open Access
- 1 March 2024
- Published by Wolters Kluwer Health in Stroke
- Vol. 55 (3) , 715-724
Screening of Moyamoya Disease From Retinal Photographs: Development and Validation of Deep Learning Algorithms
BACKGROUND: Moyamoya disease (MMD) is a rare and complex pathological condition characterized by an abnormal collateral circulation network in the basal brain. The diagnosis of MMD and its progression is unpredictable and influenced by many factors. MMD can affect the blood vessels supplying the eyes, resulting in a range of ocular symptoms. In this study, we developed a deep learning model using real-world data to assist a diagnosis and determine the stage of the disease using retinal photographs. METHODS: This retrospective observational study conducted from August 2006 to March 2022 included 498 retinal photographs from 78 patients with MMD and 3835 photographs from 1649 healthy participants. Photographs were preprocessed, and an ResNeXt50 model was developed. Model performance was measured using receiver operating curves and their area under the receiver operating characteristic curve, accuracy, sensitivity, and F1-score. Heatmaps and progressive erasing plus progressive restoration were performed to validate the faithfulness. RESULTS: Overall, 322 retinal photographs from 67 patients with MMD and 3752 retinal photographs from 1616 healthy participants were used to develop a screening and stage prediction model for MMD. The average age of the patients with MMD was 44.1 years, and the average follow-up time was 115 months. Stage 3 photographs were the most prevalent, followed by stages 4, 5, 2, 1, and 6 and healthy. The MMD screening model had an average area under the receiver operating characteristic curve of 94.6%, with 89.8% sensitivity and 90.4% specificity at the best cutoff point. MMD stage prediction models had an area under the receiver operating characteristic curve of 78% or higher, with stage 3 performing the best at 93.6%. Heatmap identified the vascular region of the fundus as important for prediction, and progressive erasing plus progressive restoration result shows an area under the receiver operating characteristic curve of 70% only with 50% of the important regions. CONCLUSIONS: This study demonstrated that retinal photographs could be used as potential biomarkers for screening and staging of MMD and the disease stage could be classified by a deep learning algorithm.
- 1 March 2024
- Published by Wolters Kluwer Health in Stroke
- Vol. 55 (3) , 715-724
Open AccessScreening of Moyamoya Disease From Retinal Photographs: Development and Validation of Deep Learning Algorithms
BACKGROUND: Moyamoya disease (MMD) is a rare and complex pathological condition characterized by an abnormal collateral circulation network in the basal brain. The diagnosis of MMD and its progression is unpredictable and influenced by many factors. MMD can affect the blood vessels supplying the eyes, resulting in a range of ocular symptoms. In this study, we developed a deep learning model using real-world data to assist a diagnosis and determine the stage of the disease using retinal photographs. METHODS: This retrospective observational study conducted from August 2006 to March 2022 included 498 retinal photographs from 78 patients with MMD and 3835 photographs from 1649 healthy participants. Photographs were preprocessed, and an ResNeXt50 model was developed. Model performance was measured using receiver operating curves and their area under the receiver operating characteristic curve, accuracy, sensitivity, and F1-score. Heatmaps and progressive erasing plus progressive restoration were performed to validate the faithfulness. RESULTS: Overall, 322 retinal photographs from 67 patients with MMD and 3752 retinal photographs from 1616 healthy participants were used to develop a screening and stage prediction model for MMD. The average age of the patients with MMD was 44.1 years, and the average follow-up time was 115 months. Stage 3 photographs were the most prevalent, followed by stages 4, 5, 2, 1, and 6 and healthy. The MMD screening model had an average area under the receiver operating characteristic curve of 94.6%, with 89.8% sensitivity and 90.4% specificity at the best cutoff point. MMD stage prediction models had an area under the receiver operating characteristic curve of 78% or higher, with stage 3 performing the best at 93.6%. Heatmap identified the vascular region of the fundus as important for prediction, and progressive erasing plus progressive restoration result shows an area under the receiver operating characteristic curve of 70% only with 50% of the important regions. CONCLUSIONS: This study demonstrated that retinal photographs could be used as potential biomarkers for screening and staging of MMD and the disease stage could be classified by a deep learning algorithm.
- Open Access
- 21 February 2024
- Published by JMIR Publications Inc. in JMIR Formative Research
- Vol. 8 (1) , e51996
Comprehensive Assessment and Early Prediction of Gross Motor Performance in Toddlers With Graph Convolutional Networks–Based Deep Learning: Development and Validation Study
Background: Accurate and timely assessment of children’s developmental status is crucial for early diagnosis and intervention. More accurate and automated developmental assessments are essential due to the lack of trained health care providers and imprecise parental reporting. In various areas of development, gross motor development in toddlers is known to be predictive of subsequent childhood developments. Objective: The purpose of this study was to develop a model to assess gross motor behavior and integrate the results to determine the overall gross motor status of toddlers. This study also aimed to identify behaviors that are important in the assessment of overall gross motor skills and detect critical moments and important body parts for the assessment of each behavior. Methods: We used behavioral videos of toddlers aged 18-35 months. To assess gross motor development, we selected 4 behaviors (climb up the stairs, go down the stairs, throw the ball, and stand on 1 foot) that have been validated with the Korean Developmental Screening Test for Infants and Children. In the child behavior videos, we estimated each child’s position as a bounding box and extracted human keypoints within the box. In the first stage, the videos with the extracted human keypoints of each behavior were evaluated separately using a graph convolutional networks (GCN)–based algorithm. The probability values obtained for each label in the first-stage model were used as input for the second-stage model, the extreme gradient boosting (XGBoost) algorithm, to predict the overall gross motor status. For interpretability, we used gradient-weighted class activation mapping (Grad-CAM) to identify important moments and relevant body parts during the movements. The Shapley additive explanations method was used for the assessment of variable importance, to determine the movements that contributed the most to the overall developmental assessment. Results: Behavioral videos of 4 gross motor skills were collected from 147 children, resulting in a total of 2395 videos. The stage-1 GCN model to evaluate each behavior had an area under the receiver operating characteristic curve (AUROC) of 0.79 to 0.90. Keypoint-mapping Grad-CAM visualization identified important moments in each behavior and differences in important body parts. The stage-2 XGBoost model to assess the overall gross motor status had an AUROC of 0.90. Among the 4 behaviors, “go down the stairs” contributed the most to the overall developmental assessment. Conclusions: Using movement videos of toddlers aged 18-35 months, we developed objective and automated models to evaluate each behavior and assess each child’s overall gross motor performance. We identified the important behaviors for assessing gross motor performance and developed methods to recognize important moments and body parts while evaluating gross motor performance.
- 21 February 2024
- Published by JMIR Publications Inc. in JMIR Formative Research
- Vol. 8 (1) , e51996
Open AccessComprehensive Assessment and Early Prediction of Gross Motor Performance in Toddlers With Graph Convolutional Networks–Based Deep Learning: Development and Validation Study
Background: Accurate and timely assessment of children’s developmental status is crucial for early diagnosis and intervention. More accurate and automated developmental assessments are essential due to the lack of trained health care providers and imprecise parental reporting. In various areas of development, gross motor development in toddlers is known to be predictive of subsequent childhood developments. Objective: The purpose of this study was to develop a model to assess gross motor behavior and integrate the results to determine the overall gross motor status of toddlers. This study also aimed to identify behaviors that are important in the assessment of overall gross motor skills and detect critical moments and important body parts for the assessment of each behavior. Methods: We used behavioral videos of toddlers aged 18-35 months. To assess gross motor development, we selected 4 behaviors (climb up the stairs, go down the stairs, throw the ball, and stand on 1 foot) that have been validated with the Korean Developmental Screening Test for Infants and Children. In the child behavior videos, we estimated each child’s position as a bounding box and extracted human keypoints within the box. In the first stage, the videos with the extracted human keypoints of each behavior were evaluated separately using a graph convolutional networks (GCN)–based algorithm. The probability values obtained for each label in the first-stage model were used as input for the second-stage model, the extreme gradient boosting (XGBoost) algorithm, to predict the overall gross motor status. For interpretability, we used gradient-weighted class activation mapping (Grad-CAM) to identify important moments and relevant body parts during the movements. The Shapley additive explanations method was used for the assessment of variable importance, to determine the movements that contributed the most to the overall developmental assessment. Results: Behavioral videos of 4 gross motor skills were collected from 147 children, resulting in a total of 2395 videos. The stage-1 GCN model to evaluate each behavior had an area under the receiver operating characteristic curve (AUROC) of 0.79 to 0.90. Keypoint-mapping Grad-CAM visualization identified important moments in each behavior and differences in important body parts. The stage-2 XGBoost model to assess the overall gross motor status had an AUROC of 0.90. Among the 4 behaviors, “go down the stairs” contributed the most to the overall developmental assessment. Conclusions: Using movement videos of toddlers aged 18-35 months, we developed objective and automated models to evaluate each behavior and assess each child’s overall gross motor performance. We identified the important behaviors for assessing gross motor performance and developed methods to recognize important moments and body parts while evaluating gross motor performance.
- Open Access
- 19 August 2023
- Published by JMIR Publications Inc. in JMIR Preprints
Comprehensive Assessment and Early Prediction of Gross Motor Performance in Toddlers With Graph Convolutional Networks–Based Deep Learning: Development and Validation Study (Preprint)
BACKGROUND Accurate and timely assessment of children’s developmental status is crucial for early diagnosis and intervention. More accurate and automated developmental assessments are essential due to the lack of trained health care providers and imprecise parental reporting. In various areas of development, gross motor development in toddlers is known to be predictive of subsequent childhood developments. OBJECTIVE The purpose of this study was to develop a model to assess gross motor behavior and integrate the results to determine the overall gross motor status of toddlers. This study also aimed to identify behaviors that are important in the assessment of overall gross motor skills and detect critical moments and important body parts for the assessment of each behavior. METHODS We used behavioral videos of toddlers aged 18-35 months. To assess gross motor development, we selected 4 behaviors (climb up the stairs, go down the stairs, throw the ball, and stand on 1 foot) that have been validated with the Korean Developmental Screening Test for Infants and Children. In the child behavior videos, we estimated each child’s position as a bounding box and extracted human keypoints within the box. In the first stage, the videos with the extracted human keypoints of each behavior were evaluated separately using a graph convolutional networks (GCN)–based algorithm. The probability values obtained for each label in the first-stage model were used as input for the second-stage model, the extreme gradient boosting (XGBoost) algorithm, to predict the overall gross motor status. For interpretability, we used gradient-weighted class activation mapping (Grad-CAM) to identify important moments and relevant body parts during the movements. The Shapley additive explanations method was used for the assessment of variable importance, to determine the movements that contributed the most to the overall developmental assessment. RESULTS Behavioral videos of 4 gross motor skills were collected from 147 children, resulting in a total of 2395 videos. The stage-1 GCN model to evaluate each behavior had an area under the receiver operating characteristic curve (AUROC) of 0.79 to 0.90. Keypoint-mapping Grad-CAM visualization identified important moments in each behavior and differences in important body parts. The stage-2 XGBoost model to assess the overall gross motor status had an AUROC of 0.90. Among the 4 behaviors, “go down the stairs” contributed the most to the overall developmental assessment. CONCLUSIONS Using movement videos of toddlers aged 18-35 months, we developed objective and automated models to evaluate each behavior and assess each child’s overall gross motor performance. We identified the important behaviors for assessing gross motor performance and developed methods to recognize important moments and body parts while evaluating gross motor performance.
- 19 August 2023
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessComprehensive Assessment and Early Prediction of Gross Motor Performance in Toddlers With Graph Convolutional Networks–Based Deep Learning: Development and Validation Study (Preprint)
BACKGROUND Accurate and timely assessment of children’s developmental status is crucial for early diagnosis and intervention. More accurate and automated developmental assessments are essential due to the lack of trained health care providers and imprecise parental reporting. In various areas of development, gross motor development in toddlers is known to be predictive of subsequent childhood developments. OBJECTIVE The purpose of this study was to develop a model to assess gross motor behavior and integrate the results to determine the overall gross motor status of toddlers. This study also aimed to identify behaviors that are important in the assessment of overall gross motor skills and detect critical moments and important body parts for the assessment of each behavior. METHODS We used behavioral videos of toddlers aged 18-35 months. To assess gross motor development, we selected 4 behaviors (climb up the stairs, go down the stairs, throw the ball, and stand on 1 foot) that have been validated with the Korean Developmental Screening Test for Infants and Children. In the child behavior videos, we estimated each child’s position as a bounding box and extracted human keypoints within the box. In the first stage, the videos with the extracted human keypoints of each behavior were evaluated separately using a graph convolutional networks (GCN)–based algorithm. The probability values obtained for each label in the first-stage model were used as input for the second-stage model, the extreme gradient boosting (XGBoost) algorithm, to predict the overall gross motor status. For interpretability, we used gradient-weighted class activation mapping (Grad-CAM) to identify important moments and relevant body parts during the movements. The Shapley additive explanations method was used for the assessment of variable importance, to determine the movements that contributed the most to the overall developmental assessment. RESULTS Behavioral videos of 4 gross motor skills were collected from 147 children, resulting in a total of 2395 videos. The stage-1 GCN model to evaluate each behavior had an area under the receiver operating characteristic curve (AUROC) of 0.79 to 0.90. Keypoint-mapping Grad-CAM visualization identified important moments in each behavior and differences in important body parts. The stage-2 XGBoost model to assess the overall gross motor status had an AUROC of 0.90. Among the 4 behaviors, “go down the stairs” contributed the most to the overall developmental assessment. CONCLUSIONS Using movement videos of toddlers aged 18-35 months, we developed objective and automated models to evaluate each behavior and assess each child’s overall gross motor performance. We identified the important behaviors for assessing gross motor performance and developed methods to recognize important moments and body parts while evaluating gross motor performance.