Mika P. Tarvainen

Background: Atrial fibrillation (AF), the most prevalent cardiac arrhythmia, affects 2% to 4% of the global adult population and is associated with an increased risk of stroke. Early diagnosis of AF and atrial flutter (AFL) is crucial due to their association with stroke risk and the challenge posed by their often asymptomatic and episodic nature. Traditional electrocardiogram (ECG) interpretation requires substantial expert input and can be challenging, especially with poor-quality ECGs. Objective: This study aimed to evaluate the performance of a deep neural network (DNN) model in detecting AF/AFL from a large, heterogeneous set of long-term ambulatory ECG recordings, including clinical data collected over 6 months at a university hospital, and assess its effectiveness in a setting reflecting the diversity and complexity of real-world clinical data. Methods: The research combined public datasets totaling 10,248 patients, ECG data from our previous studies (648 patients), and authentic long-term ECG recordings from 4346 patients at Kuopio University Hospital for development of the DNN model. Its clinical accuracy and generalizability were assessed using a separate test dataset consisting of 1039 pseudonymized long-term ECG recordings from 1010 patients, all thoroughly reviewed and annotated by experts. Results: The DNN model demonstrated high effectiveness, achieving 96.4% sensitivity and more than 99.99% specificity for time-level AF and AFL detection. At the recording level, it identified AF and AFL with 100% sensitivity and 98.77% specificity, producing false positives in only 1.2% (11/897) of recordings, of which 81.8% (9/11) had other non–AF/AFL arrhythmias. The model maintained high performance across diverse patient characteristics, including varying ages, comorbidities, coexisting arrhythmias, and poor-quality ECG recordings. Conclusions: The results demonstrate that the proposed DNN model may support automated screening for AF and AFL in long-term ambulatory ECG recordings and may reduce manual review workload in clinical practice.

We investigated whether physically active adults with uncomplicated type 1 diabetes mellitus (T1DM) differ from healthy controls in cardiopulmonary responses during incremental cardiopulmonary exercise testing (CPET). Fifteen adults with uncomplicated T1DM (duration 15 ± 7 years) and 33 matched controls (age, sex, height, and BMI) performed an incremental cycling CPET and self‐reported their weekly physical‐activity levels. Pulmonary and cardiovascular variables were recorded at rest, ventilatory thresholds 1 and 2, and peak effort, with noninvasive cardiac function assessed by impedance cardiography. Weekly light‐intensity activity frequency differed between groups according to the Mann–Whitney U test ( p = 0.028), indicating a statistically significant distributional difference. However, the Hodges–Lehmann estimated median difference was not statistically significant because the 95% CI included zero (0–3). No other self‐reported physical activity variables differed between the groups. In the linear mixed‐effects models, no significant Group effects or Group × Stage interactions were observed for any cardiopulmonary variable during CPET. In this physically active and clinically selected cohort, adults with uncomplicated T1DM showed no statistically detectable differences in cardiopulmonary responses compared with healthy controls under the present CPET conditions. Larger studies are needed to confirm these findings and evaluate potential subclinical hemodynamic or autonomic differences.

Accurate monitoring of heart rate (HR) is critical for assessing cardiac functions in a wide range of health and wellness applications. Seismocardiography (SCG), which captures subtle chest vibrations using wearable accelerometers, provides a non-invasive and cost-effective approach for resting and nocturnal HR monitoring. This study presents a deep learning-based approach for accurate heartbeat detection and HR estimation from three-dimensional SCG signals. The model was trained on a large-scale dataset of resting SCG signals collected from 6600 subjects and evaluated on an independent cohort of 947 individuals. For short-term (≤5 min) resting SCG recordings, the model achieved robust performance in heartbeat detection (PPV: 0.979, sensitivity: 0.916, F1-score: 0.946). HR estimation showed high accuracy, with a mean absolute error (MAE) of 0.27 bpm, root mean square error (RMSE) of 1.02 bpm, and correlation of 0.996 with the reference HR. To assess real-world applicability, the model was further evaluated on 28 nocturnal recordings acquired using Apple Watch accelerometer, yielding an MAE of 1.10 bpm, an RMSE of 1.88 bpm, and a correlation of 0.982. The proposed SCG-based deep learning model demonstrates robust and highly accurate HR monitoring in both resting and nocturnal conditions, highlighting its potential for integration with consumer-grade wearable devices in a server-based analysis pipeline.

Background: Atrial fibrillation (AF) and atrial flutter (AFL) are common arrhythmias associated with the risk of ischemic stroke, which can be reduced with anticoagulation therapy. Thus, early diagnosis of AF and AFL is essential. However, diagnosis may be challenging due to the paroxysmal and asymptomatic nature of these arrhythmias. Objective: Current diagnostic workflows involve time-consuming and resource-intensive manual review of noisy signals and prolonged recordings. We evaluated a mobile system that combines a wireless wearable single-lead chest strap electrocardiogram (ECG) and a novel deep neural network (DNN)–based artificial intelligence (AI) method for detecting AF/AFL episodes, AF/AFL burden, and rhythm change and estimated the delay in the detection of rhythm change from AF/AFL to sinus rhythm. We also assessed the rhythm classification performance. Methods: A total of 116 patients with recent-onset AF or AFL undergoing cardioversion were monitored using a mobile single-lead chest strap ECG system. Simultaneously, a 3-lead Holter ECG served as the reference. The DNN-based AI analyzed the single-lead chest strap ECG data to detect AF/AFL, non-AF/AFLrhythm, and noninterpretable segments, as well as to estimate AF/AFL burden and detect rhythm change. Performance metrics included sensitivity, specificity, positive predictive value, negative predictive value, and intraclass correlation coefficient for AF and AFL burden estimation. Results: The sensitivity and specificity for detecting AF/AFL were 91.9% (204.9/223.0 h) and 99.6% (242.4/243.5 h), respectively. The sensitivity for detecting AF was 96.2% (191.5/199.0 h), whereas it was 55.8% (13.4/24.0 h) for detecting AFL. The positive predictive value and negative predictive value for AF/AFL detection were 99.5% (204.9/206.0 h) and 93.1% (242.4/260.5 h), respectively. The intraclass correlation coefficient between the AF/AFL burden estimated by the DNN-based AI method and that derived from the physician-interpreted reference ECG was 0.96 (95% CI: 0.94‐0.97; P <.001). Rhythm change detection occurred within 1 minute in most cases. Conclusions: The mobile single-lead chest strap ECG system powered by a DNN-based AI algorithm demonstrated strong performance in detecting AF, estimating AF burden, and recognizing rhythm change to sinus rhythm. This AI-driven approach enables automated and accurate rhythm analysis, supporting clinical decision-making. Further validation in real-world ambulatory settings is warranted.

UNSTRUCTURED Background and Aims: Atrial fibrillation (AF), the most prevalent cardiac arrhythmia, affects 2-4% of the global adult population and is linked with an increased risk of stroke. Early diagnosis of AF and atrial flutter (AFL) is crucial due to their association with stroke risk and the challenge posed by their often asymptomatic and episodic nature. Traditional ECG interpretation, requiring significant expert input, can be challenging, especially with poor-quality ECGs. This study evaluates the performance of a deep neural network (DNN) model in detecting AF/AFL from a large, heterogeneous set of long-term ambulatory ECG recordings, including real-world clinical data collected over six months at a university hospital. The aim was to assess its effectiveness in a setting reflecting the diversity and complexity of real-world clinical data. Methods: The research combined public datasets totaling 10,896 patients and authentic long-term ECG recordings from 4,346 patients at Kuopio University Hospital for the DNN model’s development. Its clinical accuracy and generalizability were assessed using clinical data and 1039 anonymized long-term ECG recordings from 1010 patients, all thoroughly reviewed and annotated by experts. Results: The DNN model demonstrated high effectiveness, achieving 96.35% sensitivity and over 99.99% specificity in time-level AF/AFL detection. At the patient level, it identified AF/AFL with 100% sensitivity and 98.77% specificity, producing false positives in only 11 (1.23%) patients, of which nine had other AF/AFL arrhythmias. The model was highly effective across diverse patient groups, including various ages, comorbidities, other arrhythmias, and poor-quality ECGs. Conclusions: This study showcases the potential of the DNN model to significantly enhance the interpretation of long-term ambulatory ECG recordings, offering an efficient tool for AF/AFL detection, which could decrease manual labor and streamline patient management.

Objective: The aim of the study was to compare the Polar H10 and Bittium Faros™ heart rate variability monitors for use in occupational and environmental health research, considering data collection, storage, transfer, resource needs, and limitations. Methods: A literature review, device testing, and expert interviews were conducted to assess device performance and usability. Results: Polar H10 is affordable, records RR intervals for up to 20 hours offline, and requires a smartwatch and app for extended storage. Bittium Faros™ is a certified electrocardiogram monitor that records full waveforms, supports long measurements, and enables automatic data transfer. Both devices can be worn with a chest strap, but placement issues may affect data quality. Patch electrodes improve stability but may cause skin irritation. Conclusions: This assessment highlights key device differences to guide HRV monitor selection for occupational and environmental epidemiological studies.

Purpose: Cardiovascular autonomic neuropathy remains underdiagnosed in type 1 diabetes mellitus, posing a risk for severe complications, particularly in patients with lowered V̇O_2max, compared to controls. This study aimed to determine whether heart rate variability during cardiovascular autonomic reflex tests reveals early signs of cardiovascular autonomic neuropathy in patients with uncomplicated type 1 diabetes mellitus and normal cardiovascular fitness, compared to healthy controls. Methods: A type 1 diabetes mellitus group (n = 14) with no other diagnosed diseases (diabetes duration 15 ± 7 years) and a control group (n = 31) underwent deep breathing test, passive orthostatic test, and cardiopulmonary exercise test. Participants were assessed for heart rate variability, heart rate, blood pressure, and V̇O_2max (mL/min/kg). Results: Participant characteristics, including V̇O_2max (mL/min/kg), showed no significant differences. The type 1 diabetes mellitus group had higher systolic blood pressure during the supine phase of the orthostatic test than healthy controls (131.6 ± 14.7 mmHg vs. 122.4 ± 10.8 mmHg, p = 0.022). After 5 mins in the upright position, systolic blood pressure (132.2 ± 20.6 mmHg vs. 118.7 ± 11.7 mmHg, p = 0.036), heart rate (85 (76; 89) bpm vs. 75 (72; 83) bpm, p = 0.013), and the root mean square of successive RR interval differences (20.22 (11.22; 27.42) vs. 27.11 (19.90; 35.52), p = 0.033) were significantly different compared to controls. Conclusion: Patients with uncomplicated type 1 diabetes mellitus, despite having normal cardiorespiratory fitness, exhibited higher systolic pressure and greater sympathetic activation in orthostatic tests, suggesting subclinically altered cardiovascular autonomic function.

Ventilatory thresholds (VT1 and VT2) are critical in exercise prescription and athletic training, delineating the transitions from aerobic to anaerobic metabolism. More specifically, VT1 signifies the onset of lactate accumulation whilst VT2 signifies the onset of metabolic acidosis. Accurate determination of these thresholds is vital for optimizing training intensity. Fractal correlation properties of heart rate variability (HRV), particularly the short-term scaling exponent alpha 1 of Detrended Fluctuation Analysis (DFA-α1), have demonstrated potential for this purpose. This study validates the accuracy of commercial ventilatory threshold estimation algorithm (VT-algorithm) developed by Kubios. The VT-algorithm employs instantaneous heart rate (HR) relative to HR reserve and respiratory rate (RF), along with the DFA-α1. Sixty-four physically active participants underwent an incremental cardiopulmonary exercise test (CPET) with inter-beat interval (RR) measurements. DFA-α1 and the Kubios VT-algorithm were used to assess HR and oxygen uptake (VO2) at ventilatory thresholds. On average VO2 at true VT, DFA-α1, and VT-algorithm derived ventilatory thresholds were 1.74, 2.00 and 1.89 l/min (VT1) and 2.40, 2.41 and 2.40 l/min (VT2), respectively. Correspondingly, average HRs at the true VT, DFA-α1, and VT-algorithm thresholds were 141, 151 and 142 bpm (VT1) and 169, 168 and 170 bpm (VT2), respectively. When compared to the true thresholds, Bland-Altman error statistics (bias ± standard deviation of error) for the DFA-α1 thresholds were -0.26±0.41 l/min or -10±16 bpm at VT1 and 0.00±0.34 l/min or 1±10 bpm at VT2, whereas the VT-algorithm errors were - 0.15±0.28 l/min or -1±11 bpm at VT1 and 0.01±0.20 l/min or -1±7 bpm at VT2. HRV based VT determination algorithms accurately estimate ventilatory thresholds, offering insights into training zones, internal loading, and metabolic transitions during exercise without the need of laboratory equipment. The Kubios VT-algorithm, which incorporates instantaneous HR and RF along with DFA-α1, provided higher accuracy for VO2 and HR values for both VT1 and VT2.

Background: Occupational well-being supports the retention of the nurse educator workforce and their ability to manage workload. There is a research gap regarding interventions promoting occupational well-being. Purpose: To evaluate the effectiveness of an 8-workweek digital occupational well-being intervention using self-conducted exercises among nurse educators in secondary vocational nursing schools in Finland. Methods: A quasi-experimental study design was applied using an intervention group (n = 37) and a control group (n = 40). Data were collected at baseline, post, and 1-month follow-up using a questionnaire and a heart rate sensor to assess the resource-workload-balance and its associating and promoting factors. Results: This study found no statistical effects on the resource-workload-balance. Positive effects were found on associating factors (general well-being) and promoting factors (recovery experiences and self-regulation). Conclusions: Self-Help INtervention for Educators (SHINE) has the potential to promote recovery experiences during working hours; however, the intervention needs more investigation.

Multidomain interventions based on bio-/neurofeedback have proven useful in improving executive functions. The present study aimed to explore the potential efficacy and feasibility of an intervention that combined Heart Rate Variability Biofeedback (HRV-BF) and Near Infrared Hemoencephalography Neurofeedback (nirHEG-NF) on inhibitory control (IC) of healthy older adults. Thirty-four participants were randomly assigned to two groups: the biofeedback group (received a 10-week combined intervention of HRV-BF and nirHEG-NF) and the active control group (received a similar protocol without real-time biofeedback). Besides cognitive outcomes, the study examined pre- and post-changes in autonomic regulation and prefrontal blood oxygenation at rest and during training. Results revealed training-induced inhibitory control gains in one of the two interference tasks, whereas no effect was found on response inhibition. After the intervention, HRV increased in participants with the lowest levels of HRV at baseline. Although older adults increased blood oxygenation during training, no significant pre- and post-changes were found in blood flow oxygenation. These findings not only suggest that HRV-BF and nirHEG-NF potentially improve performance in certain subcomponents of inhibition (i.e., interference vs. response inhibition), but it may also be beneficial for parasympathetic activity in participants with low HRV and for increasing blood flow oxygenation on prefrontal areas during training.