1.Machine Learning Prediction of Attachment Type From Bio-Psychological Factors in Patients With Depression
Yoon Jae CHO ; Jin Sun RYU ; Jeong-Ho SEOK ; Eunjoo KIM ; Jooyoung OH ; Byung-Hoon KIM
Psychiatry Investigation 2025;22(4):412-423
		                        		
		                        			 Objective:
		                        			Adult attachment style is linked to how an individual responds to threats or stress and is known to be related to the onset of psychiatric symptoms such as depression. However, as the current assessment of attachment type mainly relies on self-report questionnaires and can be prone to bias, there is a need to incorporate physiological factors along with psychological symptoms and history in this process. We aimed to predict the measurement of two important types of adult attachment with heart rate variability (HRV), early life stress experience, and subjective psychiatric symptoms. 
		                        		
		                        			Methods:
		                        			Five hundred eighty-two subjects with depressive disorder were recruited retrospectively from January 2015 to June 2021. The experience of early life stress and psychiatric symptoms were collected, and HRV measures were obtained as input for an ensembled Voting Regressor model of machine learning-based regression models, including linear regression, ElasticNet, Support Vector Machine (SVM), Random Forest, and Extreme Gradient Boosting (XGBoost). 
		                        		
		                        			Results:
		                        			Model performances evaluated with R-squared score averaged across 30 seeds were 0.377 and 0.188 for anxious- and avoidant-attachment, respectively. Mean absolute error averaged to 13.251 and 12.083, respectively. Shapley value importance analysis indicated that for both attachment types, the most important feature was the trait-anxiety, followed by emotional abuse, state-anxiety or self-reported depressive symptoms, and fear or helplessness felt in the moment of an early life stressor. 
		                        		
		                        			Conclusion
		                        			Our results provide the evidence base that may be utilized in clinical settings to predict the degree of attachment type using bio-psychological factors. 
		                        		
		                        		
		                        		
		                        	
2.Low-Density Lipoprotein Cholesterol Level, the Lower the Better? Analysis of Korean Patients in the Treat Stroke to Target Trial
Hanim KWON ; Jae-Chan RYU ; Jae-Kwan CHA ; Sang Min SUNG ; Tae-Jin SONG ; Kyung Bok LEE ; Eung-Gyu KIM ; Yong-Won KIM ; Ji Hoe HEO ; Man Seok PARK ; Kyusik KANG ; Byung-Chul LEE ; Keun-Sik HONG ; Oh Young BANG ; Jei KIM ; Jong S. KIM
Journal of Stroke 2025;27(2):228-236
		                        		
		                        			 Background:
		                        			and Purpose The Treat Stroke to Target (TST) was a randomized clinical trial involving French and Korean patients demonstrating that a lower low-density lipoprotein cholesterol (LDL-C, <70 mg/dL) target group (LT) experienced fewer cerebro-cardiovascular events than a higher target (90–110 mg/dL) group (HT). However, whether these results can be applied to Asian patients with different ischemic stroke subtypes remains unclear. 
		                        		
		                        			Methods:
		                        			Patients from 14 South Korean centers were analyzed separately. Patients with ischemic stroke or transient ischemic attack with evidence of atherosclerosis were randomized into LT and HT groups. The primary endpoint was a composite of ischemic stroke, myocardial infarction, coronary or cerebral revascularization, and cardiovascular death. 
		                        		
		                        			Results:
		                        			Among 712 enrolled patients, the mean LDL-C level was 71.0 mg/dL in 357 LT patients and 86.1 mg/dL in 355 HT patients. The primary endpoint occurred in 24 (6.7%) of LT and in 31 (8.7%) of HT group patients (adjusted hazard ratio [HR]=0.78; 95% confidence interval [CI]=0.45–1.33, P=0.353). Cardiovascular events alone occurred significantly less frequently in the LT than in the HT group (HR 0.26, 95% CI 0.09–0.80, P=0.019), whereas there were no significant differences in ischemic stroke events (HR 1.12, 95% CI 0.60–2.10, P=0.712). The benefit of LT was less apparent in patients with small vessel disease and intracranial atherosclerosis than in those with extracranial atherosclerosis. 
		                        		
		                        			Conclusion
		                        			In contrast to the French TST, the outcomes in Korean patients were neutral. Although LT was more effective in preventing cardiovascular diseases, it was not so in stroke prevention, probably attributed to the differences in stroke subtypes. Further studies are needed to elucidate the efficacy of statins and appropriate LDL-C targets in Asian patients with stroke. 
		                        		
		                        		
		                        		
		                        	
3.Machine Learning Prediction of Attachment Type From Bio-Psychological Factors in Patients With Depression
Yoon Jae CHO ; Jin Sun RYU ; Jeong-Ho SEOK ; Eunjoo KIM ; Jooyoung OH ; Byung-Hoon KIM
Psychiatry Investigation 2025;22(4):412-423
		                        		
		                        			 Objective:
		                        			Adult attachment style is linked to how an individual responds to threats or stress and is known to be related to the onset of psychiatric symptoms such as depression. However, as the current assessment of attachment type mainly relies on self-report questionnaires and can be prone to bias, there is a need to incorporate physiological factors along with psychological symptoms and history in this process. We aimed to predict the measurement of two important types of adult attachment with heart rate variability (HRV), early life stress experience, and subjective psychiatric symptoms. 
		                        		
		                        			Methods:
		                        			Five hundred eighty-two subjects with depressive disorder were recruited retrospectively from January 2015 to June 2021. The experience of early life stress and psychiatric symptoms were collected, and HRV measures were obtained as input for an ensembled Voting Regressor model of machine learning-based regression models, including linear regression, ElasticNet, Support Vector Machine (SVM), Random Forest, and Extreme Gradient Boosting (XGBoost). 
		                        		
		                        			Results:
		                        			Model performances evaluated with R-squared score averaged across 30 seeds were 0.377 and 0.188 for anxious- and avoidant-attachment, respectively. Mean absolute error averaged to 13.251 and 12.083, respectively. Shapley value importance analysis indicated that for both attachment types, the most important feature was the trait-anxiety, followed by emotional abuse, state-anxiety or self-reported depressive symptoms, and fear or helplessness felt in the moment of an early life stressor. 
		                        		
		                        			Conclusion
		                        			Our results provide the evidence base that may be utilized in clinical settings to predict the degree of attachment type using bio-psychological factors. 
		                        		
		                        		
		                        		
		                        	
4.Machine Learning Prediction of Attachment Type From Bio-Psychological Factors in Patients With Depression
Yoon Jae CHO ; Jin Sun RYU ; Jeong-Ho SEOK ; Eunjoo KIM ; Jooyoung OH ; Byung-Hoon KIM
Psychiatry Investigation 2025;22(4):412-423
		                        		
		                        			 Objective:
		                        			Adult attachment style is linked to how an individual responds to threats or stress and is known to be related to the onset of psychiatric symptoms such as depression. However, as the current assessment of attachment type mainly relies on self-report questionnaires and can be prone to bias, there is a need to incorporate physiological factors along with psychological symptoms and history in this process. We aimed to predict the measurement of two important types of adult attachment with heart rate variability (HRV), early life stress experience, and subjective psychiatric symptoms. 
		                        		
		                        			Methods:
		                        			Five hundred eighty-two subjects with depressive disorder were recruited retrospectively from January 2015 to June 2021. The experience of early life stress and psychiatric symptoms were collected, and HRV measures were obtained as input for an ensembled Voting Regressor model of machine learning-based regression models, including linear regression, ElasticNet, Support Vector Machine (SVM), Random Forest, and Extreme Gradient Boosting (XGBoost). 
		                        		
		                        			Results:
		                        			Model performances evaluated with R-squared score averaged across 30 seeds were 0.377 and 0.188 for anxious- and avoidant-attachment, respectively. Mean absolute error averaged to 13.251 and 12.083, respectively. Shapley value importance analysis indicated that for both attachment types, the most important feature was the trait-anxiety, followed by emotional abuse, state-anxiety or self-reported depressive symptoms, and fear or helplessness felt in the moment of an early life stressor. 
		                        		
		                        			Conclusion
		                        			Our results provide the evidence base that may be utilized in clinical settings to predict the degree of attachment type using bio-psychological factors. 
		                        		
		                        		
		                        		
		                        	
5.Comparison of Natriuretic Peptide Levels in Sinus Rhythm and Atrial Fibrillation in Acute Heart Failure
Minjae YOON ; Jin Joo PARK ; Jong-Chan YOUN ; Sang Eun LEE ; Hae-Young LEE ; Jin Oh CHOI ; Kye Hun KIM ; Dong Heon YANG ; Myeong-Chan CHO ; Seok-Min KANG ; Byung-Su YOO
International Journal of Heart Failure 2025;7(2):85-95
		                        		
		                        			 Background and Objectives:
		                        			In chronic heart failure (HF), natriuretic peptide (NP) levels are higher in atrial fibrillation (AF) compared to sinus rhythm (SR). However, due to the loss of atrial contraction, AF patients are prone to hemodynamic decompensation at earlier stages.Since NP levels reflect disease severity, acutely decompensated AF patients may exhibit lower NP levels compared to SR patients, who retain greater hemodynamic reserve. 
		                        		
		                        			Methods:
		                        			We analyzed 5,048 patients with acute HF from the Korea Acute Heart Failure registry with available NP data. NP levels and echocardiographic parameters were compared between AF and SR patients. The association of NP levels with in-hospital and one-year mortality was also assessed according to cardiac rhythm. 
		                        		
		                        			Results:
		                        			Brain natriuretic peptide (BNP) and N-terminal pro-B-type natriuretic peptide (NT-proBNP) were measured in 2,027 and 3,021 patients, respectively. NP levels were lower in AF than in SR (median BNP, 740 vs. 1,044 pg/mL; median NT-proBNP, 4,420 vs. 5,198 pg/mL), particularly in HF with reduced or mildly reduced ejection fraction. A similar trend was observed regardless of HF onset or etiology. AF patients had smaller left ventricular (LV) end-diastolic diameter and larger left atrial size compared to SR patients. Higher NP tertiles were associated with increased in-hospital and one-year mortality in both groups. 
		                        		
		                        			Conclusions
		                        			In acute HF, NP levels are lower in AF than in SR. AF patients also exhibited smaller LV chamber sizes. Nevertheless, NP levels remain strong predictors of outcomes in both AF and SR patients. 
		                        		
		                        		
		                        		
		                        	
6.Low-Density Lipoprotein Cholesterol Level, the Lower the Better? Analysis of Korean Patients in the Treat Stroke to Target Trial
Hanim KWON ; Jae-Chan RYU ; Jae-Kwan CHA ; Sang Min SUNG ; Tae-Jin SONG ; Kyung Bok LEE ; Eung-Gyu KIM ; Yong-Won KIM ; Ji Hoe HEO ; Man Seok PARK ; Kyusik KANG ; Byung-Chul LEE ; Keun-Sik HONG ; Oh Young BANG ; Jei KIM ; Jong S. KIM
Journal of Stroke 2025;27(2):228-236
		                        		
		                        			 Background:
		                        			and Purpose The Treat Stroke to Target (TST) was a randomized clinical trial involving French and Korean patients demonstrating that a lower low-density lipoprotein cholesterol (LDL-C, <70 mg/dL) target group (LT) experienced fewer cerebro-cardiovascular events than a higher target (90–110 mg/dL) group (HT). However, whether these results can be applied to Asian patients with different ischemic stroke subtypes remains unclear. 
		                        		
		                        			Methods:
		                        			Patients from 14 South Korean centers were analyzed separately. Patients with ischemic stroke or transient ischemic attack with evidence of atherosclerosis were randomized into LT and HT groups. The primary endpoint was a composite of ischemic stroke, myocardial infarction, coronary or cerebral revascularization, and cardiovascular death. 
		                        		
		                        			Results:
		                        			Among 712 enrolled patients, the mean LDL-C level was 71.0 mg/dL in 357 LT patients and 86.1 mg/dL in 355 HT patients. The primary endpoint occurred in 24 (6.7%) of LT and in 31 (8.7%) of HT group patients (adjusted hazard ratio [HR]=0.78; 95% confidence interval [CI]=0.45–1.33, P=0.353). Cardiovascular events alone occurred significantly less frequently in the LT than in the HT group (HR 0.26, 95% CI 0.09–0.80, P=0.019), whereas there were no significant differences in ischemic stroke events (HR 1.12, 95% CI 0.60–2.10, P=0.712). The benefit of LT was less apparent in patients with small vessel disease and intracranial atherosclerosis than in those with extracranial atherosclerosis. 
		                        		
		                        			Conclusion
		                        			In contrast to the French TST, the outcomes in Korean patients were neutral. Although LT was more effective in preventing cardiovascular diseases, it was not so in stroke prevention, probably attributed to the differences in stroke subtypes. Further studies are needed to elucidate the efficacy of statins and appropriate LDL-C targets in Asian patients with stroke. 
		                        		
		                        		
		                        		
		                        	
7.Machine Learning Prediction of Attachment Type From Bio-Psychological Factors in Patients With Depression
Yoon Jae CHO ; Jin Sun RYU ; Jeong-Ho SEOK ; Eunjoo KIM ; Jooyoung OH ; Byung-Hoon KIM
Psychiatry Investigation 2025;22(4):412-423
		                        		
		                        			 Objective:
		                        			Adult attachment style is linked to how an individual responds to threats or stress and is known to be related to the onset of psychiatric symptoms such as depression. However, as the current assessment of attachment type mainly relies on self-report questionnaires and can be prone to bias, there is a need to incorporate physiological factors along with psychological symptoms and history in this process. We aimed to predict the measurement of two important types of adult attachment with heart rate variability (HRV), early life stress experience, and subjective psychiatric symptoms. 
		                        		
		                        			Methods:
		                        			Five hundred eighty-two subjects with depressive disorder were recruited retrospectively from January 2015 to June 2021. The experience of early life stress and psychiatric symptoms were collected, and HRV measures were obtained as input for an ensembled Voting Regressor model of machine learning-based regression models, including linear regression, ElasticNet, Support Vector Machine (SVM), Random Forest, and Extreme Gradient Boosting (XGBoost). 
		                        		
		                        			Results:
		                        			Model performances evaluated with R-squared score averaged across 30 seeds were 0.377 and 0.188 for anxious- and avoidant-attachment, respectively. Mean absolute error averaged to 13.251 and 12.083, respectively. Shapley value importance analysis indicated that for both attachment types, the most important feature was the trait-anxiety, followed by emotional abuse, state-anxiety or self-reported depressive symptoms, and fear or helplessness felt in the moment of an early life stressor. 
		                        		
		                        			Conclusion
		                        			Our results provide the evidence base that may be utilized in clinical settings to predict the degree of attachment type using bio-psychological factors. 
		                        		
		                        		
		                        		
		                        	
8.Low-Density Lipoprotein Cholesterol Level, the Lower the Better? Analysis of Korean Patients in the Treat Stroke to Target Trial
Hanim KWON ; Jae-Chan RYU ; Jae-Kwan CHA ; Sang Min SUNG ; Tae-Jin SONG ; Kyung Bok LEE ; Eung-Gyu KIM ; Yong-Won KIM ; Ji Hoe HEO ; Man Seok PARK ; Kyusik KANG ; Byung-Chul LEE ; Keun-Sik HONG ; Oh Young BANG ; Jei KIM ; Jong S. KIM
Journal of Stroke 2025;27(2):228-236
		                        		
		                        			 Background:
		                        			and Purpose The Treat Stroke to Target (TST) was a randomized clinical trial involving French and Korean patients demonstrating that a lower low-density lipoprotein cholesterol (LDL-C, <70 mg/dL) target group (LT) experienced fewer cerebro-cardiovascular events than a higher target (90–110 mg/dL) group (HT). However, whether these results can be applied to Asian patients with different ischemic stroke subtypes remains unclear. 
		                        		
		                        			Methods:
		                        			Patients from 14 South Korean centers were analyzed separately. Patients with ischemic stroke or transient ischemic attack with evidence of atherosclerosis were randomized into LT and HT groups. The primary endpoint was a composite of ischemic stroke, myocardial infarction, coronary or cerebral revascularization, and cardiovascular death. 
		                        		
		                        			Results:
		                        			Among 712 enrolled patients, the mean LDL-C level was 71.0 mg/dL in 357 LT patients and 86.1 mg/dL in 355 HT patients. The primary endpoint occurred in 24 (6.7%) of LT and in 31 (8.7%) of HT group patients (adjusted hazard ratio [HR]=0.78; 95% confidence interval [CI]=0.45–1.33, P=0.353). Cardiovascular events alone occurred significantly less frequently in the LT than in the HT group (HR 0.26, 95% CI 0.09–0.80, P=0.019), whereas there were no significant differences in ischemic stroke events (HR 1.12, 95% CI 0.60–2.10, P=0.712). The benefit of LT was less apparent in patients with small vessel disease and intracranial atherosclerosis than in those with extracranial atherosclerosis. 
		                        		
		                        			Conclusion
		                        			In contrast to the French TST, the outcomes in Korean patients were neutral. Although LT was more effective in preventing cardiovascular diseases, it was not so in stroke prevention, probably attributed to the differences in stroke subtypes. Further studies are needed to elucidate the efficacy of statins and appropriate LDL-C targets in Asian patients with stroke. 
		                        		
		                        		
		                        		
		                        	
9.Machine Learning Prediction of Attachment Type From Bio-Psychological Factors in Patients With Depression
Yoon Jae CHO ; Jin Sun RYU ; Jeong-Ho SEOK ; Eunjoo KIM ; Jooyoung OH ; Byung-Hoon KIM
Psychiatry Investigation 2025;22(4):412-423
		                        		
		                        			 Objective:
		                        			Adult attachment style is linked to how an individual responds to threats or stress and is known to be related to the onset of psychiatric symptoms such as depression. However, as the current assessment of attachment type mainly relies on self-report questionnaires and can be prone to bias, there is a need to incorporate physiological factors along with psychological symptoms and history in this process. We aimed to predict the measurement of two important types of adult attachment with heart rate variability (HRV), early life stress experience, and subjective psychiatric symptoms. 
		                        		
		                        			Methods:
		                        			Five hundred eighty-two subjects with depressive disorder were recruited retrospectively from January 2015 to June 2021. The experience of early life stress and psychiatric symptoms were collected, and HRV measures were obtained as input for an ensembled Voting Regressor model of machine learning-based regression models, including linear regression, ElasticNet, Support Vector Machine (SVM), Random Forest, and Extreme Gradient Boosting (XGBoost). 
		                        		
		                        			Results:
		                        			Model performances evaluated with R-squared score averaged across 30 seeds were 0.377 and 0.188 for anxious- and avoidant-attachment, respectively. Mean absolute error averaged to 13.251 and 12.083, respectively. Shapley value importance analysis indicated that for both attachment types, the most important feature was the trait-anxiety, followed by emotional abuse, state-anxiety or self-reported depressive symptoms, and fear or helplessness felt in the moment of an early life stressor. 
		                        		
		                        			Conclusion
		                        			Our results provide the evidence base that may be utilized in clinical settings to predict the degree of attachment type using bio-psychological factors. 
		                        		
		                        		
		                        		
		                        	
10.Clinical Usefulness of a Cell-based Assay for Detecting Myelin Oligodendrocyte Glycoprotein Antibodies in Central Nervous System Inflammatory Disorders
Jin Myoung SEOK ; Patrick WATERS ; Mi Young JEON ; Hye Lim LEE ; Seol-Hee BAEK ; Jin-Sung PARK ; Sa-Yoon KANG ; Ohyun KWON ; Jeeyoung OH ; Byung-Jo KIM ; Kyung-Ah PARK ; Sei Yeul OH ; Byoung Joon KIM ; Ju-Hong MIN
Annals of Laboratory Medicine 2024;44(1):56-63
		                        		
		                        			 Background:
		                        			The clinical implications of myelin oligodendrocyte glycoprotein autoantibodies (MOG-Abs) are increasing. Establishing MOG-Ab assays is essential for effectively treating patients with MOG-Abs. We established an in-house cell-based assay (CBA) to detect MOG-Abs to identify correlations with patients’ clinical characteristics. 
		                        		
		                        			Methods:
		                        			We established the CBA using HEK 293 cells transiently overexpressing fulllength human MOG, tested it against 166 samples from a multicenter registry of central nervous system (CNS) inflammatory disorders, and compared the results with those of the Oxford MOG-Ab-based CBA and a commercial MOG-Ab CBA kit. We recruited additional patients with MOG-Abs and compared the clinical characteristics of MOG-Ab-associated disease (MOGAD) with those of neuromyelitis optica spectrum disorder (NMOSD). 
		                        		
		                        			Results:
		                        			Of 166 samples tested, 10 tested positive for MOG-Abs, with optic neuritis (ON) being the most common manifestation (4/15, 26.7%). The in-house and Oxford MOG-Ab CBAs agreed for 164/166 (98.8%) samples (κ = 0.883, P < 0.001); two patients (2/166, 1.2%) were only positive in our in-house CBA, and the CBA scores of the two laboratories correlated well (r = 0.663, P < 0.001). The commercial MOG-Ab CBA kit showed one falsenegative and three false-positive results. The clinical presentation at disease onset differed between MOGAD and NMOSD; ON was the most frequent manifestation in MOGAD, and transverse myelitis was most frequent in NMOSD. 
		                        		
		                        			Conclusions
		                        			The in-house CBA for MOG-Abs demonstrated reliable results and can potentially be used to evaluate CNS inflammatory disorders. A comprehensive, long-term study with a large patient population would clarify the clinical significance of MOG-Abs. 
		                        		
		                        		
		                        		
		                        	
            
Result Analysis
Print
Save
E-mail