Ki-Hyun Jeon
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Ki-Hyun Jeon has a total of 217 co-authors. Here are the top co-authors by number of shared publications:
- Joon-Myoung Kwon
Medical AI Inc Seoul
10 shared publications - In-Ho Chae
Division of Cardiology, Department of Internal Medicine, Seoul National University Bundang Hospital, Seongnam, Korea
8 shared publications Jinsik Park
Department of Internal Medicine, Division of Cardiology Cardiovascular Center, Mediplex Sejong Hospital , 20, Gyeyangmunhwa-ro, Gyeyang-gu, Incheon, 21080, Republic of South Korea
8 shared publicationsByung-Hee Oh
Department of Internal Medicine, Division of Cardiology Cardiovascular Center, Mediplex Sejong Hospital , 20, Gyeyangmunhwa-ro, Gyeyang-gu, Incheon, 21080, Republic of South Korea
7 shared publications- Kyung-Hee Kim
Department of Applied Chemistry, School of Science and Technology, Kookmin University, Seoul 02707, Republic of Korea
7 shared publications - Chang-Hwan Yoon
Seoul National University Bundang Hospital, Seongnam, Korea
5 shared publications
Latest Publications
- Open Access
- 14 November 2026
- Published by MDPI AG in Diagnostics
- Vol. 16 (22) , 3777
Diagnostic Performance of Angiography-Derived MPFFR for Predicting Hemodynamically Significant Coronary Stenosis: A Multicenter Validation Study
Background and Objectives: Angiography-derived fractional flow reserve (FFR) methods offer noninvasive alternatives to wire-based FFR for functional assessment of coronary stenosis, but their diagnostic performance requires validation. This study evaluated the diagnostic performance of artificial intelligence (AI)-driven angiography-based FFR for detecting hemodynamically significant coronary stenosis, as compared with percentage diameter stenosis (%DS) from quantitative coronary angiography (QCA), using invasive FFR as the reference standard. Methods: In this prospective, multicenter study, 189 patients (211 lesions) undergoing clinically indicated angiography were enrolled. AI-driven angiography-based FFR and %DS were derived from angiographic images and compared with invasive FFR (cutoff ≤ 0.80). Diagnostic performance was assessed using AUROC, sensitivity, specificity, and predictive values, along with correlation analysis and analysis time. Results: AI-driven angiography-based FFR showed significantly greater diagnostic performance than %DS (per-patient AUROC: 0.87 vs. 0.63, p < 0.001; per-lesion AUROC: 0.89 vs. 0.65, p < 0.001). AI-driven angiography-based FFR demonstrated higher specificity (92.7%) and positive predictive value (86.7%) than %DS (64.2% and 51.9%, respectively). AI-driven angiography-based FFR correlated strongly with invasive FFR (r = 0.746), whereas %DS showed a weak inverse correlation (r = −0.250). Median analysis time for AI-driven angiography-based FFR was 20.9 s. A gray-zone hybrid strategy (thresholds 0.73–0.83) yielded 89.8% accuracy and could reduce invasive FFR use by 57.1%. Conclusions: AI-driven angiography-based FFR demonstrated high diagnostic accuracy, strong correlation with invasive FFR, and rapid analysis time. These findings support AI-driven angiography-based FFR as a pressure-wire-free tool for physiological assessment during coronary angiography.
- 14 November 2026
- Published by MDPI AG in Diagnostics
- Vol. 16 (22) , 3777
Open AccessDiagnostic Performance of Angiography-Derived MPFFR for Predicting Hemodynamically Significant Coronary Stenosis: A Multicenter Validation Study
Background and Objectives: Angiography-derived fractional flow reserve (FFR) methods offer noninvasive alternatives to wire-based FFR for functional assessment of coronary stenosis, but their diagnostic performance requires validation. This study evaluated the diagnostic performance of artificial intelligence (AI)-driven angiography-based FFR for detecting hemodynamically significant coronary stenosis, as compared with percentage diameter stenosis (%DS) from quantitative coronary angiography (QCA), using invasive FFR as the reference standard. Methods: In this prospective, multicenter study, 189 patients (211 lesions) undergoing clinically indicated angiography were enrolled. AI-driven angiography-based FFR and %DS were derived from angiographic images and compared with invasive FFR (cutoff ≤ 0.80). Diagnostic performance was assessed using AUROC, sensitivity, specificity, and predictive values, along with correlation analysis and analysis time. Results: AI-driven angiography-based FFR showed significantly greater diagnostic performance than %DS (per-patient AUROC: 0.87 vs. 0.63, p < 0.001; per-lesion AUROC: 0.89 vs. 0.65, p < 0.001). AI-driven angiography-based FFR demonstrated higher specificity (92.7%) and positive predictive value (86.7%) than %DS (64.2% and 51.9%, respectively). AI-driven angiography-based FFR correlated strongly with invasive FFR (r = 0.746), whereas %DS showed a weak inverse correlation (r = −0.250). Median analysis time for AI-driven angiography-based FFR was 20.9 s. A gray-zone hybrid strategy (thresholds 0.73–0.83) yielded 89.8% accuracy and could reduce invasive FFR use by 57.1%. Conclusions: AI-driven angiography-based FFR demonstrated high diagnostic accuracy, strong correlation with invasive FFR, and rapid analysis time. These findings support AI-driven angiography-based FFR as a pressure-wire-free tool for physiological assessment during coronary angiography.
- Open Access
- 15 October 2026
- Published by MDPI AG in Diagnostics
- Vol. 16 (20) , 3361
Near-Infrared Spectroscopy-Guided Selective Distal Perfusion in Femoral Venoarterial Extracorporeal Membrane Oxygenation: A Feasibility Study
Background/Objectives: We evaluated the feasibility of a near-infrared spectroscopy (NIRS)-guided selective strategy based on a single absolute regional tissue oxygen saturation (rSO2) threshold and described clinically overt limb ischemia. Methods: This single-center retrospective study included 110 adults receiving femoral venoarterial extracorporeal membrane oxygenation between April 2020 and December 2024. All underwent bilateral lower-extremity NIRS monitoring. In 83 patients, distal perfusion catheter (DPC) placement was triggered by rSO2 < 40% in the cannulated limb or clinically evident limb ischemia; 27 received routine prophylactic DPC placement at cannulation. Primary endpoints were DPC avoidance and clinically overt limb ischemia. Results: Extracorporeal cardiopulmonary resuscitation (ECPR) accounted for 64.5% of patients. Under the selective strategy, 45/83 patients (54.2%; exact 95% CI, 42.9–65.2%) avoided DPC placement, with no clinically overt limb ischemia (0/45; exact 95% CI, 0–7.9%). Limb ischemia occurred in 2/83 versus 2/27 patients (2.4% versus 7.4%; risk difference, −5.0 percentage points; 95% CI, −21.1 to +3.0; p = 0.252). Because threshold crossing triggered DPC placement, the reported threshold characteristics reflect the operation of the management protocol rather than independent diagnostic accuracy. Conclusions: A single-threshold NIRS-guided selective strategy was feasible in this ECPR-dominant population, allowing just over half of patients to avoid prophylactic DPC placement. No clinically overt limb ischemia was observed in this subgroup, but the upper 95% confidence limit of 7.9% precludes a definitive safety conclusion. Prospective studies are needed to assess limb ischemia risk and refine selection criteria.
- 15 October 2026
- Published by MDPI AG in Diagnostics
- Vol. 16 (20) , 3361
Open AccessNear-Infrared Spectroscopy-Guided Selective Distal Perfusion in Femoral Venoarterial Extracorporeal Membrane Oxygenation: A Feasibility Study
Background/Objectives: We evaluated the feasibility of a near-infrared spectroscopy (NIRS)-guided selective strategy based on a single absolute regional tissue oxygen saturation (rSO2) threshold and described clinically overt limb ischemia. Methods: This single-center retrospective study included 110 adults receiving femoral venoarterial extracorporeal membrane oxygenation between April 2020 and December 2024. All underwent bilateral lower-extremity NIRS monitoring. In 83 patients, distal perfusion catheter (DPC) placement was triggered by rSO2 < 40% in the cannulated limb or clinically evident limb ischemia; 27 received routine prophylactic DPC placement at cannulation. Primary endpoints were DPC avoidance and clinically overt limb ischemia. Results: Extracorporeal cardiopulmonary resuscitation (ECPR) accounted for 64.5% of patients. Under the selective strategy, 45/83 patients (54.2%; exact 95% CI, 42.9–65.2%) avoided DPC placement, with no clinically overt limb ischemia (0/45; exact 95% CI, 0–7.9%). Limb ischemia occurred in 2/83 versus 2/27 patients (2.4% versus 7.4%; risk difference, −5.0 percentage points; 95% CI, −21.1 to +3.0; p = 0.252). Because threshold crossing triggered DPC placement, the reported threshold characteristics reflect the operation of the management protocol rather than independent diagnostic accuracy. Conclusions: A single-threshold NIRS-guided selective strategy was feasible in this ECPR-dominant population, allowing just over half of patients to avoid prophylactic DPC placement. No clinically overt limb ischemia was observed in this subgroup, but the upper 95% confidence limit of 7.9% precludes a definitive safety conclusion. Prospective studies are needed to assess limb ischemia risk and refine selection criteria.
- Open Access
A pre-registered prospective study of large language models predicting late-breaking cardiovascular trial results at ESC Congress 2026
Aims: It is unknown whether large language models (LLMs) can predict a trial’s result at the design stage, from the information available when it is registered. We tested three LLMs prospectively on the Late-Breaking Science programme of the European Society of Cardiology (ESC) Congress 2026. Methods and results: Of 139 trials, 50 were selected by pre-specified criteria, and predictions were locked and registered on the Open Science Framework before the congress. Claude Fable 5, GPT-5.5 and Gemini 3.5 Flash, without web access, received each title and an investigator-compiled summary of the registered design. They gave the probability that the primary endpoint would be met, and a point estimate with an 80% prediction interval for a locked effect measure. Results were adjudicated from publications (19) or presenters’ slides (31). Of 48 scorable trials, 28 (58%) were positive. The area under the receiver operating characteristic curve was 0.80 (95% CI 0.66–0.92) for Claude, 0.79 (0.66–0.91) for GPT and 0.74 (0.59–0.87) for Gemini, with accuracies of 73%, 71% and 67%. Among 26 trials whose effect was reported exactly as locked, the direction was correct in 88% of predictions and the median multiplicative error of ratio estimates was about 15%. The 80% prediction intervals contained the observed effect in 73.1% (Claude), 84.6% (GPT) and 57.7% (Gemini). Conclusion: From the registered design alone, LLMs showed statistically significant discrimination between positive and negative trials and usually predicted the direction and approximate size of the effect. Such predictions might help trial design, and further studies are needed.
Open AccessA pre-registered prospective study of large language models predicting late-breaking cardiovascular trial results at ESC Congress 2026
Aims: It is unknown whether large language models (LLMs) can predict a trial’s result at the design stage, from the information available when it is registered. We tested three LLMs prospectively on the Late-Breaking Science programme of the European Society of Cardiology (ESC) Congress 2026. Methods and results: Of 139 trials, 50 were selected by pre-specified criteria, and predictions were locked and registered on the Open Science Framework before the congress. Claude Fable 5, GPT-5.5 and Gemini 3.5 Flash, without web access, received each title and an investigator-compiled summary of the registered design. They gave the probability that the primary endpoint would be met, and a point estimate with an 80% prediction interval for a locked effect measure. Results were adjudicated from publications (19) or presenters’ slides (31). Of 48 scorable trials, 28 (58%) were positive. The area under the receiver operating characteristic curve was 0.80 (95% CI 0.66–0.92) for Claude, 0.79 (0.66–0.91) for GPT and 0.74 (0.59–0.87) for Gemini, with accuracies of 73%, 71% and 67%. Among 26 trials whose effect was reported exactly as locked, the direction was correct in 88% of predictions and the median multiplicative error of ratio estimates was about 15%. The 80% prediction intervals contained the observed effect in 73.1% (Claude), 84.6% (GPT) and 57.7% (Gemini). Conclusion: From the registered design alone, LLMs showed statistically significant discrimination between positive and negative trials and usually predicted the direction and approximate size of the effect. Such predictions might help trial design, and further studies are needed.
- Open Access
Citation reliability of frontier large language models in medical writing and its automated verification
Large language models (LLMs) are increasingly used to draft medical manuscripts, yet their citations are unreliable and clinicians lack a validated way to verify them. We evaluated three frontier LLMs, Claude Opus 4.8, GPT-5.5, and Gemini 3.5 Flash, generating 270 cardiology narrative reviews with web search enabled, and verified all 8,050 references against PubMed. Problematic references accounted for 11.5% of GPT-5.5 output, 29.2% of Claude output, and 29.6% of Gemini output (P < 0.001), with no significant gradient across topics of differing publication volume (P = 0.052). Misattribution, a valid PubMed identifier that resolves to a different article, made up 77% of errors, whereas fabrication was rare (0.6%). Against an expert-adjudicated set of 270 references, an LLM-based Chain-of-Verification (CoVe) detected 60 of 62 problematic references (sensitivity 96.8%, specificity 98.6%), including every misattribution and fabrication. LLM-generated citations require identifier-level verification, and CoVe provides it at expert-level accuracy.
Open AccessCitation reliability of frontier large language models in medical writing and its automated verification
Large language models (LLMs) are increasingly used to draft medical manuscripts, yet their citations are unreliable and clinicians lack a validated way to verify them. We evaluated three frontier LLMs, Claude Opus 4.8, GPT-5.5, and Gemini 3.5 Flash, generating 270 cardiology narrative reviews with web search enabled, and verified all 8,050 references against PubMed. Problematic references accounted for 11.5% of GPT-5.5 output, 29.2% of Claude output, and 29.6% of Gemini output (P < 0.001), with no significant gradient across topics of differing publication volume (P = 0.052). Misattribution, a valid PubMed identifier that resolves to a different article, made up 77% of errors, whereas fabrication was rare (0.6%). Against an expert-adjudicated set of 270 references, an LLM-based Chain-of-Verification (CoVe) detected 60 of 62 problematic references (sensitivity 96.8%, specificity 98.6%), including every misattribution and fabrication. LLM-generated citations require identifier-level verification, and CoVe provides it at expert-level accuracy.
- 28 August 2026
- Published by Oxford University Press (OUP) in European Heart Journal
Microvascular resistance reserve and cardiovascular outcomes: the FLOW-CMD registry
Background and Aims: Microvascular resistance reserve (MRR) is a novel index for assessing coronary microvascular function that is independent of epicardial coronary artery stenosis and subtended myocardial mass. However, limited data are available regarding the clinical relevance of coronary microvascular dysfunction, defined by MRR, in patients undergoing clinically indicated invasive coronary angiography. This study sought to evaluate the incidence and prognostic impact of coronary microvascular dysfunction defined by MRR in routine practice. Methods: In the prospective, multicentre FLOW-CMD Registry, 1003 consecutive patients with suspected ischaemic heart disease who underwent clinically indicated invasive coronary angiography with comprehensive coronary physiologic assessment were prospectively enrolled. Coronary microvascular dysfunction was defined as MRR ≤2.5 in any evaluated vessel. The primary endpoint was a composite of all-cause death, myocardial infarction, clinically-driven repeat revascularization, or hospitalization for heart failure. Results: Coronary microvascular dysfunction was identified in 334 of 1003 patients (33.3%). MRR as a continuous variable was associated with the risk of the primary endpoint (hazard ratio [HR] per 1-unit decrease, 1.16; 95% confidence interval [CI] 1.03–1.32; P = .026). At a median follow-up of 1.9 years, the primary endpoint occurred in 47 of 334 patients (18.2%) with coronary microvascular dysfunction and 49 of 669 patients (8.6%) with preserved microvascular function (HR 1.94; 95% CI 1.30–2.90; P = .001). On multivariable analysis, coronary microvascular dysfunction defined by MRR ≤2.5 was independently associated with the primary endpoint (adjusted HR 1.78; 95% CI 1.06–2.99; P = .030). Conclusions: In patients with suspected ischaemic heart disease undergoing invasive coronary angiography, coronary microvascular dysfunction defined by MRR ≤2.5 was observed among one-third of patients, and was associated with an increased risk of a composite of all-cause death, myocardial infarction, clinically-driven repeat revascularization, or hospitalization for heart failure (Multicenter FLOW-CMD Registry ClinicalTrials.gov number, NCT05369182).
- 28 August 2026
- Published by Oxford University Press (OUP) in European Heart Journal
Microvascular resistance reserve and cardiovascular outcomes: the FLOW-CMD registry
Background and Aims: Microvascular resistance reserve (MRR) is a novel index for assessing coronary microvascular function that is independent of epicardial coronary artery stenosis and subtended myocardial mass. However, limited data are available regarding the clinical relevance of coronary microvascular dysfunction, defined by MRR, in patients undergoing clinically indicated invasive coronary angiography. This study sought to evaluate the incidence and prognostic impact of coronary microvascular dysfunction defined by MRR in routine practice. Methods: In the prospective, multicentre FLOW-CMD Registry, 1003 consecutive patients with suspected ischaemic heart disease who underwent clinically indicated invasive coronary angiography with comprehensive coronary physiologic assessment were prospectively enrolled. Coronary microvascular dysfunction was defined as MRR ≤2.5 in any evaluated vessel. The primary endpoint was a composite of all-cause death, myocardial infarction, clinically-driven repeat revascularization, or hospitalization for heart failure. Results: Coronary microvascular dysfunction was identified in 334 of 1003 patients (33.3%). MRR as a continuous variable was associated with the risk of the primary endpoint (hazard ratio [HR] per 1-unit decrease, 1.16; 95% confidence interval [CI] 1.03–1.32; P = .026). At a median follow-up of 1.9 years, the primary endpoint occurred in 47 of 334 patients (18.2%) with coronary microvascular dysfunction and 49 of 669 patients (8.6%) with preserved microvascular function (HR 1.94; 95% CI 1.30–2.90; P = .001). On multivariable analysis, coronary microvascular dysfunction defined by MRR ≤2.5 was independently associated with the primary endpoint (adjusted HR 1.78; 95% CI 1.06–2.99; P = .030). Conclusions: In patients with suspected ischaemic heart disease undergoing invasive coronary angiography, coronary microvascular dysfunction defined by MRR ≤2.5 was observed among one-third of patients, and was associated with an increased risk of a composite of all-cause death, myocardial infarction, clinically-driven repeat revascularization, or hospitalization for heart failure (Multicenter FLOW-CMD Registry ClinicalTrials.gov number, NCT05369182).
- Open Access
- 10 February 2026
- Published by Korean Medical Association in Journal of the Korean Medical Association
- Vol. 69 (2) , 118-125
Limitations and mitigation strategies for using generative artificial intelligence in medical writing: a narrative review
Purpose: Large language models (LLMs) improve medical writing efficiency but introduce methodological, ethical, and legal risks. This review examines current evidence on the limitations of LLM-assisted medical writing and proposes principles for its responsible integration into biomedical research.Current concepts: LLMs are commonly used for draft generation, language editing, literature summarization, reference handling, statistical code generation, and manuscript structuring. Studies consistently report improved readability and reduced writing time, particularly among non-native English-speaking authors. However, recurrent challenges include factual hallucinations; fabricated or inaccurate citations; incomplete retrieval of recent literature due to training cutoffs; prompt-sensitive statistical errors; ambiguity regarding authorship and accountability; risks of unintended plagiarism; and concerns related to patient data privacy. These limitations arise from the probabilistic nature of LLMs and their lack of intrinsic fact-verification mechanisms or ethical reasoning.Discussion and conclusion: Risks associated with LLM use vary by manuscript stage and therefore require differentiated oversight. LLMs should be confined primarily to language refinement rather than fact generation, and literature-related outputs must be verified against primary sources, preferably using retrieval-augmented tools. Statistical analyses should remain under human control, with independent validation of all outputs. Ethical governance requires transparent disclosure of LLM use, clear assignment of human responsibility, and strict safeguards for sensitive data. A dual framework combining human-in-the-loop and human-on-the-loop oversight offers a pragmatic model for balancing efficiency with scientific rigor. When positioned as augmentative tools rather than autonomous agents, LLMs can be responsibly integrated into medical research without compromising integrity or reproducibility.
- 10 February 2026
- Published by Korean Medical Association in Journal of the Korean Medical Association
- Vol. 69 (2) , 118-125
Open AccessLimitations and mitigation strategies for using generative artificial intelligence in medical writing: a narrative review
Purpose: Large language models (LLMs) improve medical writing efficiency but introduce methodological, ethical, and legal risks. This review examines current evidence on the limitations of LLM-assisted medical writing and proposes principles for its responsible integration into biomedical research.Current concepts: LLMs are commonly used for draft generation, language editing, literature summarization, reference handling, statistical code generation, and manuscript structuring. Studies consistently report improved readability and reduced writing time, particularly among non-native English-speaking authors. However, recurrent challenges include factual hallucinations; fabricated or inaccurate citations; incomplete retrieval of recent literature due to training cutoffs; prompt-sensitive statistical errors; ambiguity regarding authorship and accountability; risks of unintended plagiarism; and concerns related to patient data privacy. These limitations arise from the probabilistic nature of LLMs and their lack of intrinsic fact-verification mechanisms or ethical reasoning.Discussion and conclusion: Risks associated with LLM use vary by manuscript stage and therefore require differentiated oversight. LLMs should be confined primarily to language refinement rather than fact generation, and literature-related outputs must be verified against primary sources, preferably using retrieval-augmented tools. Statistical analyses should remain under human control, with independent validation of all outputs. Ethical governance requires transparent disclosure of LLM use, clear assignment of human responsibility, and strict safeguards for sensitive data. A dual framework combining human-in-the-loop and human-on-the-loop oversight offers a pragmatic model for balancing efficiency with scientific rigor. When positioned as augmentative tools rather than autonomous agents, LLMs can be responsibly integrated into medical research without compromising integrity or reproducibility.
- Open Access
- 1 January 2026
- Published by XMLink in Korean Circulation Journal
- Vol. 56 (3) , 199-215
Artificial Intelligence-Enabled Electrocardiography in Practice: A State-of-the-Art Review
Lee HS, et al. Korean Circ J. 2026 Jan;56:e55. https://doi.org/10.4070/kcj.2025.0486
- 1 January 2026
- Published by XMLink in Korean Circulation Journal
- Vol. 56 (3) , 199-215
Open AccessArtificial Intelligence-Enabled Electrocardiography in Practice: A State-of-the-Art Review
Lee HS, et al. Korean Circ J. 2026 Jan;56:e55. https://doi.org/10.4070/kcj.2025.0486
- Open Access
- 1 January 2025
- Published by XMLink in Journal of Cardiovascular Intervention
- Vol. 4 (4) , 231
ChatGPT and Medical Statistics: A Narrative Review on Opportunities, Pitfalls, and the Principle of “Trust, but Verify”
Jeon KH, et al. J Cardiovasc Interv. 2025 Jul;4:e29. https://doi.org/10.54912/jci.2025.0020
- 1 January 2025
- Published by XMLink in Journal of Cardiovascular Intervention
- Vol. 4 (4) , 231
Open AccessChatGPT and Medical Statistics: A Narrative Review on Opportunities, Pitfalls, and the Principle of “Trust, but Verify”
Jeon KH, et al. J Cardiovasc Interv. 2025 Jul;4:e29. https://doi.org/10.54912/jci.2025.0020
- Open Access
- 17 December 2024
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 13 (24) , e037705
Brachial‐Ankle Pulse Wave Velocity as a Predictor of Diabetes Development: Elevated Risk Within Normal Range Values in a Low‐Risk Population
Background: Recent studies have suggested that increased brachial‐ankle pulse wave velocity (baPWV) is a risk factor for diabetes. Exploring its relationship with insulin resistance is of interest, necessitating further studies across different sexes and age groups. Methods and Results: This cohort study involved 119 170 Korean adults with an average age of 39.8 years, none of whom had diabetes at baseline. As part of a health screening, baPWV measurements were taken. Over a median follow‐up period of 5.6 years, fasting blood glucose, glycated hemoglobin, insulin levels, and questionnaire responses were collected. The risk of developing diabetes was evaluated using a flexible parametric proportional hazards model with data stratified by sex and age group (P for trend <0.001). A significant sex‐based interaction was noted in this association, with women showing a higher risk of diabetes development. Furthermore, higher baPWV quartiles were associated with an increased risk of developing insulin resistance, defined as the homeostatic model assessment of insulin resistance. Conclusions: These findings highlight the importance of arterial stiffness, as measured by elevated baPWV, in the development of diabetes and insulin resistance. Notably, this study highlighted a strong association, particularly among women.
- 17 December 2024
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 13 (24) , e037705
Open AccessBrachial‐Ankle Pulse Wave Velocity as a Predictor of Diabetes Development: Elevated Risk Within Normal Range Values in a Low‐Risk Population
Background: Recent studies have suggested that increased brachial‐ankle pulse wave velocity (baPWV) is a risk factor for diabetes. Exploring its relationship with insulin resistance is of interest, necessitating further studies across different sexes and age groups. Methods and Results: This cohort study involved 119 170 Korean adults with an average age of 39.8 years, none of whom had diabetes at baseline. As part of a health screening, baPWV measurements were taken. Over a median follow‐up period of 5.6 years, fasting blood glucose, glycated hemoglobin, insulin levels, and questionnaire responses were collected. The risk of developing diabetes was evaluated using a flexible parametric proportional hazards model with data stratified by sex and age group (P for trend <0.001). A significant sex‐based interaction was noted in this association, with women showing a higher risk of diabetes development. Furthermore, higher baPWV quartiles were associated with an increased risk of developing insulin resistance, defined as the homeostatic model assessment of insulin resistance. Conclusions: These findings highlight the importance of arterial stiffness, as measured by elevated baPWV, in the development of diabetes and insulin resistance. Notably, this study highlighted a strong association, particularly among women.
- Open Access
- 17 July 2024
- Published by Springer Nature in Scientific Reports
- Vol. 14 (1) , 1-9
AI-enabled ECG index for predicting left ventricular dysfunction in patients with ST-segment elevation myocardial infarction
Electrocardiogram (ECG) changes after primary percutaneous coronary intervention (PCI) in ST-segment elevation myocardial infarction (STEMI) patients are associated with prognosis. This study investigated the feasibility of predicting left ventricular (LV) dysfunction in STEMI patients using an artificial intelligence (AI)-enabled ECG algorithm developed to diagnose STEMI. Serial ECGs from 637 STEMI patients were analyzed with the AI algorithm, which quantified the probability of STEMI at various time points. The time points included pre-PCI, immediately post-PCI, 6 h post-PCI, 24 h post-PCI, at discharge, and one-month post-PCI. The prevalence of LV dysfunction was significantly associated with the AI-derived probability index. A high probability index was an independent predictor of LV dysfunction, with higher cardiac death and heart failure hospitalization rates observed in patients with higher indices. The study demonstrates that the AI-enabled ECG index effectively quantifies ECG changes post-PCI and serves as a digital biomarker capable of predicting post-STEMI LV dysfunction, heart failure, and mortality. These findings suggest that AI-enabled ECG analysis can be a valuable tool in the early identification of high-risk patients, enabling timely and targeted interventions to improve clinical outcomes in STEMI patients.
- 17 July 2024
- Published by Springer Nature in Scientific Reports
- Vol. 14 (1) , 1-9
Open AccessAI-enabled ECG index for predicting left ventricular dysfunction in patients with ST-segment elevation myocardial infarction
Electrocardiogram (ECG) changes after primary percutaneous coronary intervention (PCI) in ST-segment elevation myocardial infarction (STEMI) patients are associated with prognosis. This study investigated the feasibility of predicting left ventricular (LV) dysfunction in STEMI patients using an artificial intelligence (AI)-enabled ECG algorithm developed to diagnose STEMI. Serial ECGs from 637 STEMI patients were analyzed with the AI algorithm, which quantified the probability of STEMI at various time points. The time points included pre-PCI, immediately post-PCI, 6 h post-PCI, 24 h post-PCI, at discharge, and one-month post-PCI. The prevalence of LV dysfunction was significantly associated with the AI-derived probability index. A high probability index was an independent predictor of LV dysfunction, with higher cardiac death and heart failure hospitalization rates observed in patients with higher indices. The study demonstrates that the AI-enabled ECG index effectively quantifies ECG changes post-PCI and serves as a digital biomarker capable of predicting post-STEMI LV dysfunction, heart failure, and mortality. These findings suggest that AI-enabled ECG analysis can be a valuable tool in the early identification of high-risk patients, enabling timely and targeted interventions to improve clinical outcomes in STEMI patients.