Yonsei University · 医学
Professor Seok-Jae Heo's research lab specializes in applying advanced statistical and machine learning methods to public health and biomedical data. The lab focuses on developing and refining signal detection techniques in pharmacovigilance using zero-inflated and Bayesian models, particularly for adverse drug reaction monitoring. It also conducts cutting-edge research in medical image analysis, leveraging deep learning architectures like CNNs to improve early detection of diseases such as tuberculosis from chest X-rays. Additionally, the lab investigates dietary patterns and their impact on childhood obesity, aiming to tailor nutritional interventions based on population subgroups.
Figures are computed from collected data and may differ slightly.
We aimed to use deep learning to detect tuberculosis in chest radiographs in annual workers' health examination data and compare the performances of convolutional neural networks (CNNs) based on images only (I-CNN) and CNNs including demographic variables (D-CNN). The I-CNN and D-CNN models were trained on 1000 chest X-ray images, both positive and negative, for tuberculosis. Feature extraction was conducted using VGG19, InceptionV3, ResNet50, DenseNet121, and InceptionResNetV2. Age, weight, hei
There are several different proposed data mining methods for the postmarketing surveillance of drug safety. Adverse events are often classified into a hierarchical structure. Our objective was to compare the performance of several of these different data mining methods for adverse drug events data with a hierarchical structure. We generated datasets based on the World Health Organization's Adverse Reaction Terminology (WHO-ART) hierarchical structure. We evaluated different data mining methods f
Among biological agents for the treatment of atopic dermatitis (AD), dupilumab is a front-runner. Although many studies have been conducted on the real-world use of dupilumab, the sample size is often small and data is primarily on Western people. Therefore, we investigated the efficacy and safety of dupilumab in patients with moderate-to-severe AD in Korea. All patients with moderate-to-severe AD treated with dupilumab from September 2018 to June 2019 in this institution were included and analy
CRD42021258498.
BPSD are clinically heterogeneous, and their occurrence can be predicted by different contributing factors. Our results for various BPSD suggest a critical window for timely intervention and care planning. Findings from this study will help devise symptom-targeted and individualized interventions to prevent and manage BPSD and facilitate personalized dementia care.
Background Triglyceride and glucose (TyG) index and TyG-related indices combined with obesity-related markers are considered important markers of insulin resistance. We aimed to examine the association between the TyG index and modified TyG indices with new-onset hypertension and their predictive ability stratified by sex. Methods and Results We analyzed data from 5414 Korean Genome and Epidemiology Study participants aged 40 to 69 years. Multiple Cox proportional hazard regression analyses were
Dialysis adequacy is an important survival indicator in patients with chronic hemodialysis. However, there are inconveniences and disadvantages to measuring dialysis adequacy by blood samples. This study used machine learning models to predict dialysis adequacy in chronic hemodialysis patients using repeatedly measured data during hemodialysis. This study included 1333 hemodialysis sessions corresponding to the monthly examination dates of 61 patients. Patient demographics and clinical parameter
The behavioral and psychological symptoms of dementia (BPSD) are challenging aspects of dementia care. This study used machine learning models to predict the occurrence of BPSD among community-dwelling older adults with dementia. We included 187 older adults with dementia for model training and 35 older adults with dementia for external validation. Demographic and health data and premorbid personality traits were examined at the baseline, and actigraphy was utilized to monitor sleep and activity
Previously, we have reported short term effectiveness and safety of dupilumab in Korea. In this study, we are trying to report the long-term effectiveness and safety of dupilumab in Korea. Ninety-nine patients with moderate to severe AD were analyzed. They were evaluated using Eczema Area and Severity Index (EASI), Numerical Rating Scale (NRS), Patient Oriented Eczema Measure (POEM), and Dermatology Quality of Life Index (DLQI) at baseline, week 16, 32 and 52. Efficacy outcomes showed higher imp
Bayesian signal detection methods, including the multiitem gamma Poisson shrinker (MGPS), assume a Poisson distribution for the number of reports. However, the database of the adverse event reporting system often has a large number of zero-count cells. A zero-inflated Poisson (ZIP) distribution can be more appropriate in this situation than a Poisson distribution. Few studies have considered ZIP-based models for Bayesian signal detection. In addition, most studies on Bayesian signal detection me
METS-IR surpassed HOMA-IR in predicting CKD incidence and was as effective as HOMA-IR in predicting CKD prevalence. This implies that METS-IR could be a valuable indicator for early detection and prevention of CKD among Korean adults.
The deep learning model performed well only using monitoring measurement of hemodialysis machine in predicting IDH without any personal information that could risk privacy infringement.
Maintaining optimal dietary quality and patterns are crucial to prevent childhood obesity. Further research is warranted to explore specific dietary interventions tailored to different clusters to effectively address childhood obesity.
Open papers in the app to read, cite, and organize with AI.
