1.Dementia Overdiagnosis in Younger, Higher Educated Individuals Based on MMSE Alone: Analysis Using Deep Learning Technology
Hye-Geum KIM ; Dai-Seg BAI ; Bon-Hoon KOO ; Eun-Jin CHEON ; Seokho YUN ; So Hye JO ; Byoungyoung GU
Journal of Korean Medical Science 2025;40(9):e20-
Background:
Dementia is a multifaceted disorder that affects cognitive function, necessitating accurate diagnosis for effective management and treatment. Although the Mini-Mental State Examination (MMSE) is widely used to assess cognitive impairment, its standalone efficacy is debated. This study examined the effectiveness of the MMSE alone versus in combination with other cognitive assessments in predicting dementia diagnosis, with the aim of refining the diagnostic accuracy for dementia.
Methods:
A total of 2,863 participants with subjective cognitive complaints who underwent comprehensive neuropsychological assessments were included. We developed two random forest models: one using only the MMSE and another incorporating additional cognitive tests.These models were evaluated based on their accuracy, precision, recall, F1-score, and area under the receiver operating characteristic curve (AUC) on a 70:30 training-to-testing split.
Results:
The MMSE-alone model predicted dementia with an accuracy of 86% and AUC of 0.872. The expanded model demonstrated increased accuracy (88%) and an AUC of 0.934.Notably, 17.46% of the cases were reclassified from dementia to non-dementia category upon including additional tests. Higher educational level and younger age were associated with these shifts.
Conclusion
The findings suggest that although the MMSE is a valuable screening tool, it should not be used in isolation to determine dementia severity. The addition of diverse cognitive assessments can significantly enhance diagnostic precision, particularly in younger and more educated populations. Future diagnostic protocols should integrate multifaceted cognitive evaluations to reflect the complexity of dementia accurately.
2.Dementia Overdiagnosis in Younger, Higher Educated Individuals Based on MMSE Alone: Analysis Using Deep Learning Technology
Hye-Geum KIM ; Dai-Seg BAI ; Bon-Hoon KOO ; Eun-Jin CHEON ; Seokho YUN ; So Hye JO ; Byoungyoung GU
Journal of Korean Medical Science 2025;40(9):e20-
Background:
Dementia is a multifaceted disorder that affects cognitive function, necessitating accurate diagnosis for effective management and treatment. Although the Mini-Mental State Examination (MMSE) is widely used to assess cognitive impairment, its standalone efficacy is debated. This study examined the effectiveness of the MMSE alone versus in combination with other cognitive assessments in predicting dementia diagnosis, with the aim of refining the diagnostic accuracy for dementia.
Methods:
A total of 2,863 participants with subjective cognitive complaints who underwent comprehensive neuropsychological assessments were included. We developed two random forest models: one using only the MMSE and another incorporating additional cognitive tests.These models were evaluated based on their accuracy, precision, recall, F1-score, and area under the receiver operating characteristic curve (AUC) on a 70:30 training-to-testing split.
Results:
The MMSE-alone model predicted dementia with an accuracy of 86% and AUC of 0.872. The expanded model demonstrated increased accuracy (88%) and an AUC of 0.934.Notably, 17.46% of the cases were reclassified from dementia to non-dementia category upon including additional tests. Higher educational level and younger age were associated with these shifts.
Conclusion
The findings suggest that although the MMSE is a valuable screening tool, it should not be used in isolation to determine dementia severity. The addition of diverse cognitive assessments can significantly enhance diagnostic precision, particularly in younger and more educated populations. Future diagnostic protocols should integrate multifaceted cognitive evaluations to reflect the complexity of dementia accurately.
3.Dementia Overdiagnosis in Younger, Higher Educated Individuals Based on MMSE Alone: Analysis Using Deep Learning Technology
Hye-Geum KIM ; Dai-Seg BAI ; Bon-Hoon KOO ; Eun-Jin CHEON ; Seokho YUN ; So Hye JO ; Byoungyoung GU
Journal of Korean Medical Science 2025;40(9):e20-
Background:
Dementia is a multifaceted disorder that affects cognitive function, necessitating accurate diagnosis for effective management and treatment. Although the Mini-Mental State Examination (MMSE) is widely used to assess cognitive impairment, its standalone efficacy is debated. This study examined the effectiveness of the MMSE alone versus in combination with other cognitive assessments in predicting dementia diagnosis, with the aim of refining the diagnostic accuracy for dementia.
Methods:
A total of 2,863 participants with subjective cognitive complaints who underwent comprehensive neuropsychological assessments were included. We developed two random forest models: one using only the MMSE and another incorporating additional cognitive tests.These models were evaluated based on their accuracy, precision, recall, F1-score, and area under the receiver operating characteristic curve (AUC) on a 70:30 training-to-testing split.
Results:
The MMSE-alone model predicted dementia with an accuracy of 86% and AUC of 0.872. The expanded model demonstrated increased accuracy (88%) and an AUC of 0.934.Notably, 17.46% of the cases were reclassified from dementia to non-dementia category upon including additional tests. Higher educational level and younger age were associated with these shifts.
Conclusion
The findings suggest that although the MMSE is a valuable screening tool, it should not be used in isolation to determine dementia severity. The addition of diverse cognitive assessments can significantly enhance diagnostic precision, particularly in younger and more educated populations. Future diagnostic protocols should integrate multifaceted cognitive evaluations to reflect the complexity of dementia accurately.
4.Dementia Overdiagnosis in Younger, Higher Educated Individuals Based on MMSE Alone: Analysis Using Deep Learning Technology
Hye-Geum KIM ; Dai-Seg BAI ; Bon-Hoon KOO ; Eun-Jin CHEON ; Seokho YUN ; So Hye JO ; Byoungyoung GU
Journal of Korean Medical Science 2025;40(9):e20-
Background:
Dementia is a multifaceted disorder that affects cognitive function, necessitating accurate diagnosis for effective management and treatment. Although the Mini-Mental State Examination (MMSE) is widely used to assess cognitive impairment, its standalone efficacy is debated. This study examined the effectiveness of the MMSE alone versus in combination with other cognitive assessments in predicting dementia diagnosis, with the aim of refining the diagnostic accuracy for dementia.
Methods:
A total of 2,863 participants with subjective cognitive complaints who underwent comprehensive neuropsychological assessments were included. We developed two random forest models: one using only the MMSE and another incorporating additional cognitive tests.These models were evaluated based on their accuracy, precision, recall, F1-score, and area under the receiver operating characteristic curve (AUC) on a 70:30 training-to-testing split.
Results:
The MMSE-alone model predicted dementia with an accuracy of 86% and AUC of 0.872. The expanded model demonstrated increased accuracy (88%) and an AUC of 0.934.Notably, 17.46% of the cases were reclassified from dementia to non-dementia category upon including additional tests. Higher educational level and younger age were associated with these shifts.
Conclusion
The findings suggest that although the MMSE is a valuable screening tool, it should not be used in isolation to determine dementia severity. The addition of diverse cognitive assessments can significantly enhance diagnostic precision, particularly in younger and more educated populations. Future diagnostic protocols should integrate multifaceted cognitive evaluations to reflect the complexity of dementia accurately.
5.Mendelian Randomization Studies in Atherosclerotic Cardiovascular Diseases
Journal of Lipid and Atherosclerosis 2024;13(3):280-291
This review aimed to highlight the pivotal role of Mendelian randomization (MR) in advancing atherosclerotic cardiovascular disease (ASCVD) research—a field often hindered by the complexities and limitations of traditional studies. MR, which uses genetic variants as instrumental variables, provides a robust mechanism for inferring causality, offering insights untainted by the confounding factors and biases often prevalent in observational and randomized controlled trials. We explored the significant contributions of MR for elucidating the causal relationship between low-density lipoprotein cholesterol and ASCVD, and analyzed its assumptions and methodological nuances. We discussed issues surrounding instrumental variable selection, pleiotropy, and ethical considerations, in an effort to offer a balanced and insightful analysis. We highlighted the promising integration of MR with emerging technologies and global data sharing, as well as its potential to drive personalized medicine. This review provided a concise yet comprehensive journey into MR’s transformative impact on ASCVD research, offering a blend of current insights and challenges, in addition to future prospects. We aimed to serve a valuable resource for those seeking to navigate the intricate pathways of causality and intervention in ASCVD, to aid the development of enhanced understanding and targeted treatment strategies in the future.
6.Clinical Characteristics and Followup Assessment in Patients Diagnosed With Alzheimer’s Dementia Through Regional Dementia Centers and Conventional Hospital System
Eunhwan JEONG ; Dougho PARK ; Su Yun LEE ; Haejong KIM ; Heum Dai KWON ; Mun-Chul KIM ; Kyung Won PARK
Journal of Korean Medical Science 2023;38(33):e257-
Background:
The rapidly increasing socioeconomic strain caused by dementia represents a significant public health concern. Regional dementia centers (RDCs) have been established nationwide, and they aim to provide timely screening and diagnosis of dementia. This study investigated the clinical characteristics and progression of patients diagnosed with Alzheimer’s dementia (AD), who underwent treatment in RDCs or conventional communitybased hospital systems.
Methods:
This retrospective single-center cohort study included patients who were diagnosed with AD between January 2019 and March 2022. This study compared two groups of patients: the hospital group, consisting of patients who presented directly to the hospital, and the RDC group, those who were referred to the hospital from the RDCs in Pohang city. The clinical courses of the patients were monitored for a year after AD diagnosis.
Results:
A total of 1,209 participants were assigned to the hospital (n = 579) or RDC group (n = 630). The RDC group had a mean age of 80.1 years ± 6.6 years, which was significantly higher than that of the hospital group (P < 0.001). The RDC group had a higher proportion of females (38.3% vs. 31.9%; P = 0.022), higher risk for alcohol consumption (12.4% vs. 3.3%; P < 0.001), and greater number of patients who discontinued treatment 1 year after diagnosis (48.3% vs. 39.0%; P = 0.001). In the linear regression model, the RDC group was independently associated with the clinical dementia rating sum of boxes increment (β = 22.360, R 2 = 0.048, and P < 0.001).
Conclusion
Patients in the RDC group were older, had more advanced stages of conditions, and exhibited a more rapid rate of cognitive decline than patients diagnosed through the conventional hospital system. Our results suggested that RDC contributed to the screening of AD in a local region, and further nationwide study with the RDC database of various areas of Korea is needed.
7.Relative Risks for Dementia among Individuals with Glaucoma: A Meta-Analysis of Observational Cohort Studies
Min Gu HUH ; Young Kook KIM ; Jaekyoung LEE ; Young In SHIN ; Yun Jeong LEE ; Sooyeon CHOE ; Dai Woo KIM ; Yoon JEONG ; Jin Wook JEOUNG ; Ki Ho PARK
Korean Journal of Ophthalmology 2023;37(6):490-500
Purpose:
To investigate the relative risks (RRs) for dementia among individuals with glaucoma.
Methods:
We conducted a search of PubMed, Web of Science, Scopus, and Cochrane databases for observational cohort studies examining the association between glaucoma and dementia until March 2023. Two authors independently screened all titles and abstracts according to predefined inclusion and exclusion criteria. Pooled RR and 95% confidence intervals (CIs) were generated using random-effect models.
Results:
The meta-analysis included 18 cohort studies conducted in eight countries and involving 4,975,325 individuals. The pooled RR for the association between glaucoma and all-cause dementia was 1.314 (95% CI, 1.099–1.572; I2 = 95%). The pooled RRs for the associations of open-angle glaucoma with Alzheimer dementia and Parkinson disease were 1.287 (95% CI, 1.007–1.646; I2 = 96%) and 1.233 (95% CI, 0.677–2.243; I2 = 73%), respectively. The pooled RRs for the associations of angle-closure glaucoma with all-cause dementia and Alzheimer dementia were 0.978 (95% CI, 0.750–1.277; I2 = 17%) and 0.838 (95% CI, 0.421–1.669; I2 = 16%), respectively. No evidence of publication bias was detected in the Begg-Mazumdar adjusted rank correlation test (p = 0.47).
Conclusions
Based on current observational cohort studies, there is evidence supporting that glaucoma is a risk factor for dementia in the adult population.
8.Resting heart rate control and prognosis in coronary artery disease patients with hypertension previously treated with bisoprolol: a sub-group analysis of the BISO-CAD study.
Yun-Dai CHEN ; Xin-Chun YANG ; Vinh Nguyen PHAM ; Shi-An HUANG ; Guo-Sheng FU ; Xiao-Ping CHEN ; Binh Quang TRUONG ; Yu YANG ; Shao-Wen LIU ; Tian-Rong MA ; Dong-Soo KIM ; Tae-Hoon KIM
Chinese Medical Journal 2020;133(10):1155-1165
BACKGROUND:
Resting heart rate (RHR) is considered as a strong predictor of total mortality and hospitalization due to heart failure in hypertension patients. Bisoprolol fumarate, a second-generation beta-adrenoreceptor blockers (β-blocker) is commonly prescribed drug to manage hypertension. The present study was to retrospectively evaluate changes in the average RHR and its association with cardiovascular outcomes in bisoprolol-treated coronary artery disease (CAD) patients from the CAD treated with bisoprolol (BISO-CAD) study who had comorbid hypertension.
METHODS:
We performed ad-hoc analysis for hypertension sub-group of the BISO-CAD study (n = 866), which was a phase IV, multination, multi-center, single-arm, observational study carried out from October 2011 to July 2015 across China, South Korea, and Vietnam. Multivariate regression analysis was used to identify factors associated with incidence of composite cardiac clinical outcome (CCCO), the results were presented as adjusted odds ratio (OR) along with 95% confidence interval (CI) and adjusted P value.
RESULTS:
A total of 681 patients (mean age: 64.77 ± 10.33 years) with hypertension from BISO-CAD study were included in the analysis. Bisoprolol improved CCCOs in CAD patients with comorbid hypertension, with RHR <65 and <70 beats/min compared with RHR ≥65 and ≥75 beats/min, respectively, in the efficacy analysis (EA) set. In addition, it lowered RHR in both intent-to-treat (ITT) and EA groups after 6, 12, and 18 months of treatment. Further, RHR 70 to 74 beats/min resulted in significantly higher risk of CCCOs EA set of patients (adjusted OR: 4.34; 95% CI: 1.19-15.89; P = 0.03). Also, events of hospitalization due to acute coronary syndrome were higher when RHR 69 to 74 beats/min compared to RHR <69 beats/min in ITT patients.
CONCLUSION
Bisoprolol can effectively reduce RHR in Asian CAD patients with comorbid hypertension and hence, improve CCCO without affecting their blood pressure.
9.Tissue Fluid Enzyme-Linked Immunosorbant Assay for Piglets Experimentally Infected with Toxoplasma gondii and Survey on Local and Imported Pork in Korean Retail Meat Markets
Won Gi YOO ; Sun Min KIM ; Eun Jeong WON ; Ji Yun LEE ; Fuhong DAI ; Ho Choon WOO ; Ho Woo NAM ; Tae Im KIM ; Jeong Hee HAN ; Dongmi KWAK ; Yun Sang CHO ; Seung Won KANG ; Tong Soo KIM ; Xing Quan ZHU ; Chunren WANG ; Heejeong YOUN ; Sung Jong HONG
The Korean Journal of Parasitology 2018;56(5):437-446
To investigate the prevalence of Toxoplasma gondii in pork on the market in Korea, an in-house enzyme-linked immunosorbent assay for tissue fluid (CAU-tf-ELISA) was developed using a soluble extract of T. gondii RH strain tachyzoites. As the standard positive controls, the piglets were experimentally infected with T. gondii: Group A (1,000 cysts-containing bradyzoites), Group B (500 cysts-containing bradyzoites) and Group C (1.0×103 or 1.0×104 tachyzoites). The CAU-tf-ELISA demonstrated infection intensity-dependent positivity toward tissue fluids with average cut-off value 0.15: 100% for Group A, 93.8% for Group B and 40.6% for Group C. When tissue-specific cut-off values 0.066–0.199 were applied, CAU-tf-ELISA showed 96.7% sensitivity, 100% specificity, 100% positive and 90.0% negative predictive values. When compared with the same tissue fluids, performance of CAU-tf-ELISA was better than that of a commercial ELISA kit. Of the 583 Korea domestic pork samples tested, anti-T. gondii antibodies were detected from 9.1% of whole samples and 37.9% from skirt meat highest among pork parts. In the 386 imported frozen pork samples, 1.8% (skirt meat and shoulder blade) were positive for anti-T. gondii antibodies. In Korea, prevalence of anti-T. gondii antibodies in the pork on retail markets appeared high, suggesting that regulations on pig farming and facilities are necessary to supply safe pork on the tables.
Agriculture
;
Antibodies
;
Enzyme-Linked Immunosorbent Assay
;
Korea
;
Meat
;
Prevalence
;
Red Meat
;
Sensitivity and Specificity
;
Shoulder
;
Social Control, Formal
;
Toxoplasma
10.Pre-treated Populus tomentiglandulosa extract inhibits neuronal loss and alleviates gliosis in the gerbil hippocampal CA1 area induced by transient global cerebral ischemia.
Joon Ha PARK ; Tae Kyeong LEE ; Ji Hyeon AHN ; Bich Na SHIN ; Jeong Hwi CHO ; In Hye KIM ; Jae Chul LEE ; Jong Dai KIM ; Young Joo LEE ; Il Jun KANG ; Seongkweon HONG ; Yang Hee KIM ; Yong Hwan JEON ; Yun Lyul LEE ; Moo Ho WON
Anatomy & Cell Biology 2017;50(4):284-292
The genus Populus (poplar) belonging to the Salicaceae family has been used in traditional medicine, and its several species show various pharmacological properties including antioxidant and anti-inflammatory effects. No study regarding protective effects of Populus species against cerebral ischemia has been reported. Therefore, in the present study, we examined neuroprotective effects of ethanol extract from Populus tomentiglandulosa (Korea poplar) in the hippocampal cornu ammonis (CA1) area of gerbils subjected to 5 minutes of transient global cerebral ischemia. Pretreatment with 200 mg/kg of P. tomentiglandulosa extract effectively protected CA1 pyramidal neurons from transient global cerebral ischemia. In addition, glial fibrillary acidic protein immunoreactive astrocytes and ionized calcium binding adapter molecule 1 immunoreactive microglia were significantly diminished in the ischemic CA1 area by pretreatment with 200 mg/kg of P. tomentiglandulosa extract. Briefly, our results indicate that pretreatment with P. tomentiglandulosa extract protects neurons from transient cerebral ischemic injury and diminish cerebral ischemia-induced reactive gliosis in ischemic CA1 area. Based on these results, we suggest that P. tomentiglandulosa can be used as a potential candidate for prevention of ischemic injury.
Astrocytes
;
Brain Ischemia*
;
Calcium
;
Ethanol
;
Gerbillinae*
;
Glial Fibrillary Acidic Protein
;
Gliosis*
;
Hippocampus
;
Humans
;
Medicine, Traditional
;
Microglia
;
Neurons*
;
Neuroprotective Agents
;
Populus*
;
Pyramidal Cells
;
Salicaceae

Result Analysis
Print
Save
E-mail