Comparison of High- and Low-Dose Rivaroxaban Regimens in Elderly East Asian Patients With Atrial Fibrillation.

BACKGROUND: In the Rivaroxaban Once-daily oral direct factor Xa inhibition Compared with vitamin K antagonism for prevention of stroke and Embolism Trial in Atrial Fibrillation (ROCKET AF) trial, rivaroxaban 20 mg was the on-label dose, and the dose-reduction criterion for rivaroxaban was a creatinine clearance of < 50 mL/min. Some Asian countries are using reduced doses label according to the J-ROCKET AF trial. The aim of this study was to assess the safety and efficacy of a high-dose rivaroxaban regimen (HDRR, 20/15 mg) and low-dose rivaroxaban regimen (LDRR, 15/10 mg) among elderly East Asian patients with atrial fibrillation (AF) in real-world practice. METHODS: This study was a multicenter, prospective, non-interventional observational study designed to evaluate the efficacy and safety of rivaroxaban in AF patients > 65 years of age with or without renal impairment. RESULTS: A total of 1,093 patients (mean age, 72.8 ± 5.8 years; 686 [62.9%] men) were included in the analysis, with 493 patients allocated to the HDRR group and 598 patients allocated to the LDRR group. A total of 765 patients received 15 mg of rivaroxaban (203 in the HDRR group and 562 in the LDRR group). There were no significant differences in the incidence rates of major bleeding (adjusted hazard ratio [HR], 0.64; 95% confidential interval [CI], 0.21-1.93), stroke (adjusted HR, 3.21; 95% CI, 0.54-19.03), and composite outcomes (adjusted HR, 1.13; 95% CI, 0.47-2.69) between the HDRR and LDRR groups. CONCLUSION: This study revealed the safety and effectiveness of either dose regimen of rivaroxaban in an Asian population for stroke prevention of AF. Considerable numbers of patients are receiving LDRR therapy in real-world practice in Asia. Both regimens were safe and effective for these patients. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT04096547.

Mechanical and Pharmacological Revascularization Strategies for Prevention of Microvascular Dysfunction in ST-Segment Elevation Myocardial Infarction: Analysis from Index of Microcirculatory Resistance Registry Data.

OBJECTIVES: We aimed to identify mechanical and pharmacological revascularization strategies correlated with the index of microcirculatory resistance (IMR) in ST-elevation myocardial infarction (STEMI) patients. BACKGROUND: Microvascular dysfunction (MVD) after STEMI is correlated with infarct size and poor long-term prognosis, and the IMR is a useful analytical method for the quantitative assessment of MVD. However, therapeutic strategies that can reliably reduce MVD remain uncertain. METHODS: Patients with STEMI who underwent primary percutaneous coronary intervention (PCI) were enrolled. The IMR was measured with a pressure sensor/thermistor-tipped guidewire immediately after primary PCI. High IMR was defined as values ≥66th percentile of IMR in enrolled patients (IMR > 30.9 IU). RESULTS: A total of 160 STEMI patients were analyzed (high IMR = 54 patients). Clinical factors for Killip class (P=0.006), delayed hospitalization from symptom onset (P=0.004), peak troponin-I level (P=0.042), and multivessel disease (P=0.003) were associated with high IMR. Achieving final thrombolysis in myocardial infarction myocardial perfusion grade 3 tended to be associated with low IMR (P=0.119), whereas the presence of distal embolization was significantly associated with high IMR (P=0.034). In terms of therapeutic strategies that involved adjusting clinical and angiographic factors associated with IMR, preloading of third-generation P2Y12 inhibitors correlated with reducing IMR value (ß = -10.30, P < 0.001). Mechanical therapeutic strategies including stent diameter/length, preballoon dilatation, direct stenting, and thrombectomy were not associated with low IMR value (all P > 0.05), and postballoon dilatation was associated with high IMR (ß = 8.30, P=0.020). CONCLUSIONS: In our study, mechanical strategies were suboptimal in achieving myocardial salvage. Preloading of third-generation P2Y12 inhibitors revealed decreased IMR value, indicative of MVD prevention.

Ivabradine-Induced Torsade de Pointes in Patients with Heart Failure Reduced Ejection Fraction.

Ivabradine is a selective inhibitor of the sinoatrial node "funny" current, prolonging the slow diastolic depolarization. As it has the ability to block the heart rate selectively, it is more effective at a faster heart rate. It is recommended for the treatment of heart failure reduced ejection fraction in the presence of beta-blocker therapy for the further reduction of the heart rate. However, previous reports have shown the association of Torsade de pointes (TdP) with concurrent use of ivabradine and drugs resulting in QT prolongation or blockage of the metabolic breakdown of ivabradine. In this article, we report two cases of patients with heart failure reduced ejection fraction who developed TdP after ivabradine use. Our report highlights the need to exercise caution with the administration of ivabradine in the presence of a reduced repolarization reserve, such as QT prolongation or metabolic insufficiency.

Prognostic Impact of Left Atrial Minimal Volume on Clinical Outcome in Patients with Non-Obstructive Hypertrophic Cardiomyopathy.

Maximal left atrial volume (LAVmax) has been suggested to be an important indicator of left ventricular (LV) diastolic function and a prognosticator in patients with hypertrophic cardiomyopathy (HCM). However, LAVmax can be influenced by LV longitudinal systolic function, which causes systolic descent of the mitral plane. We investigated the prognostic role of LAVmin in patients with HCM and tested if LAVmin is better than LAVmax in predicting clinical outcome in these patients. A total of 167 consecutive patients with HCM were enrolled (age = 64.7 ± 13.5 years, male: female = 120:47). Clinical parameters and conventional echocardiographic measurement including tissue Doppler measurement were evaluated. Left atrial maximal and minimal volumes were measured just before mitral valve opening and at mitral valve closure respectively using the biplane disk method. The relationship between LAVmin and the clinical outcome of hospitalization for heart failure (HF), stroke or all-cause mortality was evaluated. During a median follow-up of 25.0 ± 17.8 months, the primary end point of HF hospitalization, stroke or death occurred in 35 patients (21%). Indexed LAVmin was predictive of HF, stroke or death after adjustment for age, diabetes, hypertension, atrial fibrillation, LV ejection fraction, and E/e'in a multivariate analysis (P = 0.001). The model including indexed LAVmin was superior to the model including indexed LAVmax in predicting a worse outcome in patients with HCM (P = 0.02). In conclusion, LAVmin was independently associated with increased risk of HF, stroke, or mortality in patients with HCM and was superior to LAVmax in predicting clinical outcome in this population.

A Multi-Resolution Mode CMOS Image Sensor with a Novel Two-Step Single-Slope ADC for Intelligent Surveillance Systems.

In this paper, we present a multi-resolution mode CMOS image sensor (CIS) for intelligent surveillance system (ISS) applications. A low column fixed-pattern noise (CFPN) comparator is proposed in 8-bit two-step single-slope analog-to-digital converter (TSSS ADC) for the CIS that supports normal, 1/2, 1/4, 1/8, 1/16, 1/32, and 1/64 mode of pixel resolution. We show that the scaled-resolution images enable CIS to reduce total power consumption while images hold steady without events. A prototype sensor of 176 × 144 pixels has been fabricated with a 0.18 µm 1-poly 4-metal CMOS process. The area of 4-shared 4T-active pixel sensor (APS) is 4.4 µm × 4.4 µm and the total chip size is 2.35 mm × 2.35 mm. The maximum power consumption is 10 mW (with full resolution) with supply voltages of 3.3 V (analog) and 1.8 V (digital) and 14 frame/s of frame rates.

Impact of area strain by 3D speckle tracking on clinical outcome in patients after acute myocardial infarction.

BACKGROUND: Three-dimensional (3D) speckle tracking echocardiography (STE) has been developed to overcome the limitations of two-dimensional (2D) STE and has been applied in the several clinical settings. However, no data exist about the prognostic value of 3DSTE-based strain on clinical outcome after myocardial infarction (MI). This study was designed to investigate the prognostic value of area strain (AS) by 3D speckle tracking in predicting clinical outcome after acute MI. METHODS: We assessed 96 patients (62±14 years, 72% male) with acute MI and who had undergone a coronary angiography. Clinical parameters and conventional echocardiographic measurements including the left atrial (LA) size and tissue Doppler measurements were evaluated. The global left ventricular (LV) AS was measured using 3D speckle tracking software. The relationship between the AS and clinical outcome of death or hospitalization for heart failure (HF) was assessed. RESULTS: During a median follow-up of 33±10 months, primary endpoint of death or HF occurred in 12 patients (12.5%). AS was predictive of death or HF after adjustment for age, gender, peak CK-MB, LA volume, LV end-systolic volume, LV mass, the ratio of early mitral inflow velocity to early mitral annular velocity, and LV ejection fraction in a multivariate Cox model (HR 1.23, 95% CI 1.02-1.47, P=.03). In addition, AS added incremental value in predicting death or heart failure on a model based on clinical and standard echocardiographic measures (P=.008). CONCLUSION: AS is independently associated with increased risk of death or HF after acute MI, suggesting that it can be a useful prognostic parameter in the patients following MI.

Comparison of joint kinematics and pedaling force in the young and the elderly.

[Purpose] Proper pedaling posture can improve muscle strength and cardiopulmonary function. To investigate proper pedaling posture for the elderly, this study compared the pedaling efficiency of the elderly with that of the young by using an index of effectiveness (IE) and kinematic results. [Subjects and Methods] Eight adults in their twenties and eight in their seventies participated in 3-min, 40 rpm cycle pedaling tests, with the same load and cadence. The joint angle, range of motion (ROM), and IE were compared by measuring 3-dimensional motion and 3-axis pedal-reaction force during 4 pedaling phases (Phase 1: 330-30°, Phase 2: 30-150°, Phase 3: 150-210°, and Phase 4: 210-330°). [Results] The knee and ankle ROM, maximum knee extension, and maximum ankle dorsiflexion in the elderly were significantly decreased compared with those in the young. Moreover, there were significant differences in IE for the total phase, Phase 1, and Phase 4 between the elderly and young. IE of the young was greater than that of the elderly, except in Phase 3. [Conclusion] Joint movement in the elderly during pedaling was limited. This study provides information that will facilitate the proposal of an efficient pedaling method for the elderly.

Fall- and BBS-related differences in muscle strength and postural balance of the elderly.

[Purpose] The purpose of this study was to compare the differences in muscle strength and postural balance between fallers and non-fallers. We also compared the difference between normal and impaired balance groups using the same subjects and the same variables. [Subjects and Methods] Seventy-one healthy elderly females (age: 75.1 ± 75 years; weight: 57.3 ± 57 kg; height: 150.1 ± 15 cm) who had high levels of physical activity participated [25 fallers (FG) vs. 46 non-fallers (NG); and 52 healthy balance group (HBG) and 19 impaired balance group (IBG) subjects]. To compare the groups, the muscle strengths of 9 muscle groups, and 20 variables of the instrumented standing balance assessment (2 area variables, 9 time-domain variables, and 9 frequency-domain variables) were assessed. [Results] The FG and NG could only be categorized based on the frequency-domain variables of the instrumented standing balance assessment. On the other hand, there were significant differences between HBG and IBG in height, 6 muscle strength, and 2 time-domain variables of the instrumented standing balance assessment. [Conclusion] These results suggest that muscle strength and standing balance are reflected in physical balance ability (i.e., BBS); however they are in sufficient for determining the actual occurrence of falls.

The development and evaluation of a program for leg-strengthening exercises and balance assessment using Kinect.

[Purpose] In this study, a program was developed for leg-strengthening exercises and balance assessment using Microsoft Kinect. [Subjects and Methods] The program consists of three leg-strengthening exercises (knee flexion, hip flexion, and hip extension) and the one-leg standing test (OLST). The program recognizes the correct exercise posture by comparison with the range of motion of the hip and knee joints and provides a number of correct action examples to improve training. The program measures the duration of the OLST and presents this as the balance-age. The accuracy of the program was analyzed using the data of five male adults. [Results] In terms of the motion recognition accuracy, the sensitivity and specificity were 95.3% and 100%, respectively. For the balance assessment, the time measured using the existing method with a stopwatch had an absolute error of 0.37 sec. [Conclusion] The developed program can be used to enable users to conduct leg-strengthening exercises and balance assessments at home.

A study on balance assessment according to the levels of difficulty in postural control.

[Purpose] This study aimed to determine appropriate measures for assessing balance ability according to difficulty level during standing tasks. [Subjects and Methods] The subjects were 56 old (>65 years) and 30 young (20-30 years) adults. By using the Berg balance scale, the subjects were divided into three groups: 29 healthy older (Berg score≥52), 27 impaired older (Berg score≥40), and 30 healthy young (Berg score≥55). One inertial measurement unit sensor was attached at the waist, and the subjects performed standing tasks (1 min/task) with six difficulty levels: eyes open and eyes closed on firm ground, one foam, and two foams. Thirty-nine (24 time-domain, 15 frequency-domain) measures were calculated by using acceleration data. The slope of each derived measure was calculated through the least-squares method. [Results] Five (95% ellipse sway area, root mean squares [anterior-posterior and resultant directions], and mean distance [anterior-posterior and resultant directions] in time domain) of the 39 measures showed significant differences among the groups under specific standing conditions. The slopes of derived measures showed significant differences among the groups and significant correlations with the Berg scores. [Conclusion] The slope according to the difficulty level can be used to assess and discriminate standing balance ability.

Relationship between J Waves and Vagal Activity in Patients Who Do Not Have Structural Heart Disease.

BACKGROUND: J waves are associated with increased vagal activity in patients with idiopathic ventricular fibrillation in several studies to date. However, the relationship between J waves and autonomic nervous activity in patients without structural heart disease remains under investigation. We investigated whether the presence of a J wave on the surface electrocardiogram (ECG) was related to increased vagal activity in patients without structural heart disease. METHODS: This retrospective study included 684 patients without structural heart disease who had undergone Holter ECG and surface ECG monitoring. Based on the presence of J waves on the surface ECG, patients were divided into two groups: those with J waves (group 1) and those without J waves (group 2). We compared heart rate variability (HRV), reflecting autonomic nervous activity, using 24-h Holter ECG between the groups. RESULTS: J waves were present in 92 (13.4%) patients. Heart rate (HR) in group 1 was significantly lesser than that in group 2 (P = 0.031). The ratio of low-frequency (LF) components to high-frequency (HF) components (LF/HF) in group 1 was significantly lower than that in group 2 (P = 0.001). The square root of the mean squared differences of successive NN intervals in group 1 was also significantly higher than that in group 2 (P = 0.047). In a multivariate regression analysis, male sex, HR, and LF/HF ratio remained independent determinants for the presence of J waves (P = 0.039, P = 0.036, and P < 0.001, respectively). CONCLUSION: In patients without structural heart disease, the presence of a J wave was associated with a slow HR, male sex, and increased vagal activity, independently.

A high speed CMOS image sensor with a novel digital correlated double sampling and a differential difference amplifier.

In order to increase the operating speed of a CMOS image sensor (CIS), a new technique of digital correlated double sampling (CDS) is described. In general, the fixed pattern noise (FPN) of a CIS has been reduced with the subtraction algorithm between the reset signal and pixel signal. This is because a single-slope analog-to-digital converter (ADC) has been normally adopted in the conventional digital CDS with the reset ramp and signal ramp. Thus, the operating speed of a digital CDS is much slower than that of an analog CDS. In order to improve the operating speed, we propose a novel digital CDS based on a differential difference amplifier (DDA) that compares the reset signal and the pixel signal using only one ramp. The prototype CIS has been fabricated with 0.13 µm CIS technology and it has the VGA resolution of 640 × 480. The measured conversion time is 16 µs, and a high frame rate of 131 fps is achieved at the VGA resolution.

Effects of low light on the stability of the head and pelvis of the healthy elderly.

[Purpose] The purpose of this study was to evaluate the changes in body stability of the elderly while walking on even surface ground under low light. [Subjects] Ten young males and ten elderly males participated in this experiment. [Methods] Each subject walked along a 7 m walkway five times at their preferred walking speed under normal (>300 lux, NORM) and low light conditions (<5 lux, LOW). To compare the changes in body stability, the root mean square of acceleration (RMSacc) at the head and pelvis was used. [Results] The results show that the body stability of young adults showed a similar RMSacc in all directions at the head and pelvis between the normal and low light walking conditions. In contrast, the RMSacc in all directions at the head and pelvis during low light walking by elderly adults was significantly greater than that of normal light walking. [Conclusion] It was confirmed that, despite walking on even ground, low light condition affects the body stability of the elderly. To clearly evaluate the effect of low light with aging on gait pattern, further study will be necessary to perform additional experiments under various environmental conditions to investigate walking speed, multi-tasking, stairs, and uneven walkway performance.

Carvedilol analog modulates both basal and stimulated sinoatrial node automaticity.

The membrane voltage clock and calcium (Ca(2+)) clock jointly regulate sinoatrial node (SAN) automaticity. VK-II-36 is a novel carvedilol analog that suppresses sarcoplasmic reticulum (SR) Ca(2+) release but does not block the ß-receptor. The effect of VK-II-36 on SAN function remains unclear. The purpose of this study was to evaluate whether VK-II-36 can influence SAN automaticity by inhibiting the Ca(2+) clock. We simultaneously mapped intracellular Ca(2+) and membrane potential in 24 isolated canine right atriums using previously described criteria of the timing of late diastolic intracellular Ca elevation (LDCAE) relative to the action potential upstroke to detect the Ca(2+) clock. Pharmacological interventions with isoproterenol (ISO), ryanodine, caffeine, and VK-II-36 were performed after baseline recordings. VK-II-36 caused sinus rate downregulation and reduced LDCAE in the pacemaking site under basal conditions (P < 0.01). ISO induced an upward shift of the pacemaking site in SAN and augmented LDCAE in the pacemaking site. ISO also significantly and dose-dependently increased the sinus rate. The treatment of VK-II-36 (30 µmol/l) abolished both the ISO-induced shift of the pacemaking site and augmentation of LDCAE (P < 0.01), and it suppressed the ISO-induced increase in sinus rate (P = 0.02). Our results suggest that the sinus rate may be partly controlled by the Ca(2+) clock via SR Ca(2+) release during ß-adrenergic stimulation.

The relationship between J wave on the surface electrocardiography and ventricular fibrillation during acute myocardial infarction.

We investigated whether the presence of J wave on the surface electrocardiography (sECG) could be a potential risk factor for ventricular fibrillation (VF) during acute myocardial infarction (AMI). We performed a retrospective study of 317 patients diagnosed with AMI in a single center from 2009 to 2012. Among the enrolled 296 patients, 22 (13.5%) patients were selected as a VF group. The J wave on the sECG was defined as a J point elevation manifested through QRS notching or slurring at least 1 mm above the baseline in at least two leads. We found that the incidence of J wave on the sECG was significantly higher in the VF group. We also confirmed that several conventional risk factors of VF were significantly related to VF during AMI; time delays from the onset of chest pain, blood concentrations of creatine phosphokinase and incidence of ST-segment elevation. Multiple logistic regression analysis demonstrated that the presence of J wave and the presence of a ST-segment elevation were independent predictors of VF during AMI. This study demonstrated that the presence of J wave on the sECG is significantly related to VF during AMI.

Establishment of a Dataset for the Traditional Korean Medicine Examination in Healthy Adults.

We established a protocol for the traditional Korean medicine examination (KME) and methodically gathered data following this protocol. Potential indicators for KME were extracted through a literature review; the first KME protocol was developed based on three rounds of expert opinions. The first KME protocol's feasibility was confirmed, and data were collected over four years from traditional Korean medicine (KM) hospitals, focusing on healthy adults, using the final KME protocol. A literature review identified 175 potential core indicators, condensed into 73 indicators after three rounds of expert consultation. The first KME protocol, which was categorized under questionnaires and medical examinations, was developed after the third round of expert opinions. A pilot study using the first KME protocol was conducted to ensure its validity, leading to modifications resulting in the development of the final KME protocol. Over four years, data were collected from six KM hospitals, focusing on healthy adults; we obtained a dataset comprising 11,036 healthy adults. This is the first protocol incorporating core indicators of KME in a quantitative form and systematically collecting data. Our protocol holds potential merit in evaluating predisposition to diseases or predicting diseases.

Resting energy expenditure differs among individuals with different levels of perceived thermal sensitivity: A cross-sectional study.

Metabolic rate has been used in thermophysiological models for predicting the thermal response of humans. However, only a few studies have investigated the association between an individual's trait-like thermal sensitivity and resting energy expenditure (REE), which resulted in inconsistent results. This study aimed to explore the association between REE and perceived thermal sensitivity. The REE of healthy adults was measured using an indirect calorimeter, and perceived thermal intolerance and sensation in the body were evaluated using a self-administered questionnaire. In total, 1567 individuals were included in the analysis (women = 68.9%, age = 41.1 ±â€…13.2 years, body mass index = 23.3 ±â€…3.3 kg/m2, REE = 1532.1 ±â€…362.4 kcal/d). More women had high cold intolerance (31.8%) than men (12.7%), and more men had high heat intolerance (23.6%) than women (16.1%). In contrast, more women experienced both cold (53.8%) and heat (40.6%) sensations in the body than men (cold, 29.1%; heat, 27.9%). After adjusting for age, fat-free mass, and fat mass, lower cold intolerance, higher heat intolerance, and heat sensation were associated with increased REE only in men (cold intolerance, P for trend = .001; heat intolerance, P for trend = .037; heat sensation, P = .046), whereas cold sensation was associated with decreased REE only in women (P = .023). These findings suggest a link between the perceived thermal sensitivity and REE levels in healthy individuals.

Use of artificial intelligence-powered ECG to differentiate between cardiac and pulmonary pathologies in patients with acute dyspnoea in the emergency department.

BACKGROUND: Acute dyspnoea is common in acute care settings. However, identifying the origin of dyspnoea in the emergency department (ED) is often challenging. We aimed to investigate whether our artificial intelligence (AI)-powered ECG analysis reliably distinguishes between the causes of dyspnoea and evaluate its potential as a clinical triage tool for comparing conventional heart failure diagnostic processes using natriuretic peptides. METHODS: A retrospective analysis was conducted using an AI-based ECG algorithm on patients ≥18 years old presenting with dyspnoea at the ED from February 2006 to September 2023. Patients were categorised into cardiac or pulmonary origin groups based on initial admission. The performance of an AI-ECG using a transformer neural network algorithm was assessed to analyse standard 12-lead ECGs for accuracy, sensitivity, specificity and area under the receiver operating characteristic curve (AUC). Additionally, we compared the diagnostic efficacy of AI-ECG models with N-terminal probrain natriuretic peptide (NT-proBNP) levels to identify cardiac origins. RESULTS: Among the 3105 patients included in the study, 1197 had cardiac-origin dyspnoea. The AI-ECG model demonstrated an AUC of 0.938 and 88.1% accuracy for cardiac-origin dyspnoea. The sensitivity, specificity and positive and negative predictive values were 93.0%, 79.5%, 89.0% and 86.4%, respectively. The F1 score was 0.828. AI-ECG demonstrated superior diagnostic performance in identifying cardiac-origin dyspnoea compared with NT-proBNP. True cardiac origin was confirmed in 96 patients in a sensitivity analysis of 129 patients with a high probability of cardiac origin initially misdiagnosed as pulmonary origin predicted by AI-ECG. CONCLUSIONS: AI-ECG demonstrated superior diagnostic accuracy over NT-proBNP and showed promise as a clinical triage tool. It is a potentially valuable tool for identifying the origin of dyspnoea in emergency settings and supporting decision-making.

Artificial intelligence-enhanced electrocardiography for early assessment of coronavirus disease 2019 severity.

Despite challenges in severity scoring systems, artificial intelligence-enhanced electrocardiography (AI-ECG) could assist in early coronavirus disease 2019 (COVID-19) severity prediction. Between March 2020 and June 2022, we enrolled 1453 COVID-19 patients (mean age: 59.7 ± 20.1 years; 54.2% male) who underwent ECGs at our emergency department before severity classification. The AI-ECG algorithm was evaluated for severity assessment during admission, compared to the Early Warning Scores (EWSs) using the area under the curve (AUC) of the receiver operating characteristic curve, precision, recall, and F1 score. During the internal and external validation, the AI algorithm demonstrated reasonable outcomes in predicting COVID-19 severity with AUCs of 0.735 (95% CI: 0.662-0.807) and 0.734 (95% CI: 0.688-0.781). Combined with EWSs, it showed reliable performance with an AUC of 0.833 (95% CI: 0.830-0.835), precision of 0.764 (95% CI: 0.757-0.771), recall of 0.747 (95% CI: 0.741-0.753), and F1 score of 0.747 (95% CI: 0.741-0.753). In Cox proportional hazards models, the AI-ECG revealed a significantly higher hazard ratio (HR, 2.019; 95% CI: 1.156-3.525, p = 0.014) for mortality, even after adjusting for relevant parameters. Therefore, application of AI-ECG has the potential to assist in early COVID-19 severity prediction, leading to improved patient management.

Artificial intelligence-estimated biological heart age using a 12-lead electrocardiogram predicts mortality and cardiovascular outcomes.

Background: There is a paucity of data on artificial intelligence-estimated biological electrocardiography (ECG) heart age (AI ECG-heart age) for predicting cardiovascular outcomes, distinct from the chronological age (CA). We developed a deep learning-based algorithm to estimate the AI ECG-heart age using standard 12-lead ECGs and evaluated whether it predicted mortality and cardiovascular outcomes. Methods: We trained and validated a deep neural network using the raw ECG digital data from 425,051 12-lead ECGs acquired between January 2006 and December 2021. The network performed a holdout test using a separate set of 97,058 ECGs. The deep neural network was trained to estimate the AI ECG-heart age [mean absolute error, 5.8 ± 3.9 years; R-squared, 0.7 (r = 0.84, p < 0.05)]. Findings: In the Cox proportional hazards models, after adjusting for relevant comorbidity factors, the patients with an AI ECG-heart age of 6 years older than the CA had higher all-cause mortality (hazard ratio (HR) 1.60 [1.42-1.79]) and more major adverse cardiovascular events (MACEs) [HR: 1.91 (1.66-2.21)], whereas those under 6 years had an inverse relationship (HR: 0.82 [0.75-0.91] for all-cause mortality; HR: 0.78 [0.68-0.89] for MACEs). Additionally, the analysis of ECG features showed notable alterations in the PR interval, QRS duration, QT interval and corrected QT Interval (QTc) as the AI ECG-heart age increased. Conclusion: Biological heart age estimated by AI had a significant impact on mortality and MACEs, suggesting that the AI ECG-heart age facilitates primary prevention and health care for cardiovascular outcomes.