Search Results (555)

The “athlete’s heart” represents a physiological condition consequent to long-term adaptation to sport-specific training loads. Cardiac imaging is a pivotal tool in the assessment of athletes. The study of the autonomic nervous system (ANS) using heart rate variability is becoming increasingly popular in sports fields, given its fundamental role in cardiovascular adaptation to exercise. Unfortunately, many barriers limit the clinical use of this methodology. In this observational study of 330 elite athletes, we used a composite index of ANS control (ANSI), designed to overcome pitfalls, following the hypothesis that studying the ANS could help address cardiac adaptation to high-volume training. In athletes subdivided into three groups considering the level (LOW/HIGH) of combined static/dynamic exercise and the presence [+]/absence [−] of left ventricular hypertrophy (LVH), we found that ANSI showed a progressive increase (LOW: 38.2 ± 27.6%; HIGH-LVH [−]: 52.1 ± 27.2%; HIGH-LVH [+]: 64.4 ± 24.9%, J-T test p < 0.001), with significant differences between all groups considered (p < 0.001). After adjustment and within the HIGH group, ANSI showed the strongest association with LVH and the highest explanatory power among autonomic variables. In conclusion, ANSI was able to differentiate elite athletes characterized by different cardiac remodeling and workloads (as suggested by different sport disciplines), corroborating the hypothesis for a wider use of ANS evaluation in the sports field. Full article

The association of HDL and its subfractional profile with cardiovascular health, particularly atherosclerosis, is well established; however, its association with post-exercise heart rate response remains underexplored. This cross-sectional study investigated whether HDL subfraction distribution is associated with post-exercise heart rate dynamics. We analyzed 304 adults, stratifying HDL into ten subfractions and 3 subclasses using the Lipoprint^® system. Heart rate was measured at rest (HR_rest), immediately after the YMCA 3-min step test (HR_aft), and during recovery (HR_5min and HR_10min) to calculate ΔHR. Multiple regressions were applied with False Discovery Rate correction. Participants with a more favorable post-exercise heart rate profile exhibited higher ApoA-I levels and favorable lipid ratios. Subfractions spanning the large and intermediate ranges (HDL-3 to HDL-5) were inversely associated with HR_aft, HR_5min, and ΔHR. In contrast, smaller, lipid-poor subfractions (HDL-7 to HDL-10) were associated with higher heart rates and a less favorable post-exercise response. Total HDL-C and subclass-level concentrations showed no significant association. These findings suggest that HDL particle size distribution may provide exploratory insight into exercise-related cardiovascular responses beyond conventional lipid metrics. Although limited by the use of a submaximal field test and manual heart rate assessment, these results support further investigation of HDL subfraction profiling in relation to post-exercise heart rate dynamics. Full article

This study proposes a multivariate analytical framework for transabdominally recorded intrauterine acoustic signals (TRIAS) and evaluates spectral changes across gestation. Using a digital stethoscope, 60 recordings were obtained from 44 pregnant women (≥14 weeks). Power spectral density (PSD) was calculated (30–2000 Hz) and analyzed using ordinary least squares (OLS) regression, principal component analysis (PCA), and partial least squares regression (PLSR). A consistent spectral peak was observed at 300–400 Hz. OLS revealed a significant decrease in PSD levels within 400–600 Hz (specifically 478.5 Hz and 521.5 Hz) during the second trimester (t ≈ −3.5), while changes were minimal in the third trimester. PCA identified maternal heart rate as the primary contributor to the first principal component, and PLSR showed that maternal cardiovascular dynamics and gestational progression were associated with distinct, largely independent components in the multivariate space. These results suggest that TRIAS may capture a composite maternal–fetal acoustic environment influenced by both maternal circulation and gestational uterine alterations. Specifically, the observed second-trimester spectral changes may reflect fluid dynamic or structural transitions—such as variations in amniotic fluid volume—and may serve as a baseline for future monitoring research. Full article

As the pace of modern life continues to accelerate, the pressure participants face is growing, and mental health issues are becoming increasingly prominent. Against this backdrop, meditation, as a proven method for stress relief and relaxation, has garnered widespread attention. However, many people face challenges during meditation, such as difficulty entering a meditative state quickly or achieving sub-optimal outcomes. This is particularly true for beginners, who often struggle to accurately gauge the rhythm of meditation and thus fail to fully harness its regulatory effects on both body and mind. To address these issues, this study proposes a handheld meditation device. By making contact with the body via sensors, the device can measure multiple physiological metrics in real time, including skin conductance, electromyography, and heart rate. Based on these measurements, the device can monitor the user’s emotional fluctuations in real time. When emotional changes are detected, it uses the data to play music, release specific scents, or adjust lighting ambiance, thereby dynamically regulating the user’s psychological state. This helps users better immerse themselves in a meditative state and effectively enhances the benefits of meditation. This paper provides an in-depth analysis of the device’s design principles, detailing its hardware components—including various sensors and emotional regulation modules—and explaining the operational logic of its software algorithms. The effectiveness and reliability of the device were verified through rigorous experiments. The study also thoroughly examines the application prospects and potential value of this handheld meditation device, exploring new approaches and methods for the development of meditation technology and related equipment. Full article

During acute coronary occlusion, ischemia is a major determinant of the cell response to subsequent reperfusion and is the major precursor of the typical “ischemia–reperfusion injury” (IRI). Therefore, elucidating the full IRI process primarily relies on a good understanding of ischemia-induced alterations. Ischemic arrhythmias frequently arise during the acute phase of a myocardial infarction (MI) and originate in the terminal arborisations of the cardiac conduction system. These ventricular arrhythmias are triggered by abnormal Ca²⁺-dependent depolarisations (DADs) of Purkinje cells (Pcells) due to increased spontaneous Ca²⁺ release by the sarcoplasmic reticulum (SR). This early alteration of the conduction tissue is also likely to provide a substrate for IRI-related arrhythmogenicity. Recent evidence associates the ischemic phase of the MI with a significant increase in SERCA2 pump expression in Pcells, suggesting that enhanced SR-Ca²⁺ release results from an augmentation of Ca²⁺ sequestration by the SR in those cells. We examined this hypothesis by assessing the impact of ischemia on the dynamics of SR-Ca²⁺ uptake in live Pcells by high-resolution confocal microscopy in a classical canine model of LAD coronary ligation. Pcells from five normal hearts were compared with cells from five hearts 48 Hrs after coronary occlusion. Purkinje-specific Ca²⁺ events, namely peripheral Ca²⁺ wavelets (Wlets) and central cell-wide waves (CWWs), were analysed to assess the regional SR-Ca²⁺ transport of Pcells. A total of 83 normal and 126 MI Wlets, along with 10 normal and 30 MI CWWs, were analysed to compare the peripheral and central SR-Ca²⁺ transports of Pcells between normal and ischemic hearts. Forty-eight hours following the onset of ischemia, individual SR-Ca²⁺ release sites exhibited a 60% increase in Ca²⁺ spark firing rate. However, the site density remained unchanged, indicating an acceleration of intra-SR-Ca²⁺ cycling rather than direct alteration of the SR-Ca²⁺ release channels. While central CWWs remained unchanged, a 37% acceleration of resting Ca²⁺ restoration was readily visible in peripheral Wlets, consistent with enhanced SR-Ca²⁺ uptake at the cell periphery. Computational modelling reproduced these findings when the Ca²⁺ uptake rate was numerically increased by 35%, confirming that augmented SERCA activity is sufficient to explain the pro-arrhythmic SR-Ca²⁺ release of Pcells after MI. Our findings confirm that the augmentation of Ca²⁺ pump density in the periphery of Pcells is associated with an increase in SR-Ca²⁺ uptake, explaining the arrhythmogenicity of Purkinje fibres in an ischemic heart. This ischemia-mediated pro-arrhythmic remodelling of intracellular Ca²⁺ handling in the conduction system is also likely to contribute to triggered activity during subsequent reperfusion. Full article

Background/Objectives: Ischemic injury resulting from the blockage of the left coronary artery or its terminal left anterior descending and circumflex branches is associated with relatively high rates of morbidity and mortality. Variations in left coronary artery size and the number of terminal branches may in part influence the development of such blockages due to altered intraluminal fluid dynamics, while also potentially complicating medical imaging interpretation or preoperative planning for interventions of coronary circulation. While left coronary artery morphology has been extensively reported on, the results reveal substantial geographical variability, with the Eastern United States being largely underrepresented. Methods: We aimed to establish baseline measurements of left coronary artery length and width and the prevalence of variable branching patterns by examining 101 cadaveric hearts from adult, whole-body donors (59 F, 42 M) that were ethically willed to the State of Virginia. Results: The most common terminal branching pattern was left coronary artery bifurcation (n = 54), followed by trifurcation (n = 45), tetrafurcation (n = 1), and aberrant circumflex artery origin, with the left coronary artery traveling in the anterior interventricular sulcus without branching (n = 1). Compared to females, the left coronary artery of males was significantly longer ($\overline{\mathrm{x}}$ = 14.87 mm ± 5.73 mm vs. $\overline{\mathrm{x}}$ = 12.64 mm ± 4.79; p = 0.04) and wider ($\overline{\mathrm{x}}$ = 6.31 mm ± 1.10 mm vs. $\overline{\mathrm{x}}$ = 5.54 mm ± 0.90 mm; p < 0.01), regardless of terminal branching pattern. Conclusions: These results can support the interpretation of coronary angiograms and the preoperative planning of surgical interventions involving coronary vasculature. Full article

High-dynamic amusement ride conditions involving impacts, rapid rotations, and abrupt posture changes introduce severe motion artifacts that degrade vital sign quality and destabilize physiological state recognition. This study aims to develop an engineering-ready closed-loop framework for robust passenger monitoring and intelligent diagnosis. A multimodal sensing and modeling pipeline was designed to jointly leverage physiological signals such as heart rate and SpO₂ and kinematic measurements, including acceleration, angular rate, velocity, and attitude. Inertial and PPG signals were preprocessed into supervised samples through wavelet multiresolution denoising and coordinate frame unification, while a strapdown inertial navigation system was used to propagate a 12-channel physical quantity sequence. To ensure interpretability and standards compliance, constraints from GB 8408-2018 were translated into executable threshold rules, enabling standards-driven auto-labeling and rule-based early warning. Building on this foundation, three learning modules were developed: a fusion model for high-dynamic heart rate estimation, a CNN–LSTM dynamic-threshold-enhanced network TAPNet for rapid kinematic anomaly screening, and an attention-augmented hybrid model HS-BANet integrating one-dimensional residual blocks, bidirectional LSTM, and multi-head attention for fine-grained arrhythmia classification. Experimental results demonstrated accurate and consistent heart rate estimation with RMSE of 1.18 bpm on HSSH-I and 1.24 bpm on the independent HSSH-II set, strong agreement with training and testing correlations of 0.9928 and 0.9865, and near-zero bias in Bland–Altman analysis. TAPNet achieved 96.9% validation accuracy and 98.2% test accuracy for kinematic anomaly recognition, maintaining robust generalization under class imbalance. HS-BANet enabled multi-class identification of PVC, PAC, VT, SVT, and AF, achieving an accuracy of 92.37%, an F1-score of 86.87%, a precision of 88.45%, a sensitivity of 88.14%, and a specificity of 89.42%. Overall, the proposed two-stage multimodal closed-loop—fast, interpretable early warning based on physical quantity thresholds followed by fine-grained diagnosis from physiological signals—supports stable feature extraction and reliable decision-making under strong motion artifacts and non-stationary dynamics, balancing responsiveness and diagnostic credibility, while showing potential for practical safety early warning and future deployment-oriented operational support in amusement ride scenarios. Full article

Background: Research on the acute effects of β-alanine supplementation has primarily focused on performance outcomes, with limited attention to the underlying physiological mechanisms. This study aimed to investigate the acute effects of two β-alanine doses on performance, mechanical output, and acid–base balance during a 10 min cycling time trial (10’-TT), and to explore the relationship between buffering-related variables and performance. Methods: Eighty-five recreational cyclists performed a 10’-TT under indoor conditions before (control) and following the acute ingestion of β-alanine (moderate-dose β-alanine 10 g—BAM; high-dose β-alanine 20 g—BAH) or placebo (PLA), with each condition tested on separate days. Data were analyzed using two-way repeated-measures ANOVA and correlation analyses. Results: No significant differences were observed in performance variables (distance, speed, cadence, or heart rate; p ≥ 0.751). However, total external mechanical work (kJ) was significantly reduced following acute supplementation (p = 0.028). Notably, the BAM condition reduced the mechanical cost of exercise without impairing performance, and this effect was moderately associated with changes in bicarbonate levels. Conclusions: Acute β-alanine supplementation did not improve performance outcomes but may alter buffering-related physiological responses associated with reduced mechanical work during high-intensity cycling exercise. These findings highlight the relevance of buffering-related mechanisms, particularly bicarbonate dynamics, in modulating the mechanical cost (work performed relative to performance achieved) of high-intensity exercise. Full article

Gas Electron Multiplier (GEM) detectors have become an indispensable component of modern tracking systems. The heart of a GEM detector is a thin polyimide foil (∼50 µm) clad with copper (∼5 µm) on both sides and containing an array of regularly spaced holes (typically diameter of ∼70 µm and pitch of ∼140 µm) fabricated using photolithographic techniques. The presence of the dielectric substrate (polyimide) within the amplification region introduces a time dependent response when the detector is exposed to external irradiation, a phenomenon commonly referred to as the charging-up effect. This effect arises from the accumulation of charge on the insulating polyimide surfaces, leading to a gradual modification of the local electric field configuration inside the GEM holes and, consequently, a variation in the detector gain over time. The charging-up behaviour has been systematically investigated for triple GEM chamber prototypes using an Fe-55 radioactive source (5.9 keV X-rays) with an activity of ∼20 mCi. The characteristic charging-up time constant has been extracted, and its dependence on detector gain and irradiation rate has been examined. In addition, the uniformity of detector performance in terms of count rate, gain, and energy resolution has been studied both before and after the charging-up process. In this review article, the experimental setup, data acquisition methodology, and analysis procedures developed and carried out by our group are summarised. The key findings reported by other groups, relevant Monte Carlo simulation efforts, and future outlook for the charging-up investigation on GEM based detectors are also discussed in this article. The investigations and their outcomes reviewed here provide valuable insight into the charging-up dynamics of GEM detectors and their dependence on operational parameters. Full article

Soccer is a complex sport with significant physical, physiological, psychological, technical, and tactical demands on players. This review presents an ergonomics-based model of soccer performance, emphasizing that no single component operates in isolation. Building on the foundational ergonomic framework, this review integrates contemporary evidence on training load monitoring, ecological dynamics, and cognitive-perceptual performance dimensions not systematically addressed in prior frameworks. Elite outfield players cover 9–14 km·h⁻¹ per match, with high-speed running (19.8–24.8 km·h⁻¹) making up about 20% of total distance and sprinting (>25 km·h⁻¹) around 2%. These outputs vary by playing position, tactical formation, possession dynamics, and environmental conditions. Longitudinal data from the English Premier League indicate a 35% increase in high-speed running over the past decade, suggesting intensifying physical demands. Physiological responses, including average heart rates of 156–175 bpm, reflect the aerobic and anaerobic demands on players. The review also examines benchmarks like VO₂max, sprint velocity, and anthropometry, highlighting their utility and limitations as performance indicators. Regarding training load management, the review evaluates frameworks such as the Acute:Chronic ratio and high-speed running exposure protocols, noting limitations and risks of over-relying on external load metrics. Periodization approaches, including tactical periodization, are discussed for integrating physical, technical, tactical, and psychological components in training. The proposed ergonomic model conceptualizes elite soccer performance as an emergent property of interacting physical, physiological, tactical, psychological, and environmental subsystems, with direct implications for training design, selection, and load management. Selection decisions should consider cognitive and perceptual competencies like decision-making, anticipation, and situational awareness, alongside physical and physiological profiles, aligned with the team’s game model. Full article

Background: Sleep-disordered breathing (SDB) is frequently accompanied by autonomic nervous system (ANS) dysfunction, which is closely associated with an increased incidence of cardiovascular diseases and elevated mortality risk. Heart rate variability (HRV) serves as a classic metric for evaluating sympathovagal balance; however, the specific impacts of four distinct types of respiratory events—obstructive apnea (OA), central apnea (CA), mixed apnea (MA), and hypopnea (HYP)—on HRV remain underinvestigated. Utilizing ultra-short-term HRV analysis, this study aimed to evaluate the immediate effects of different respiratory events on ANS function, while further exploring the modulatory roles of arousal, Apnea–Hypopnea Index (AHI) severity and sleep stages (non-rapid eye movement [NREM] vs. rapid eye movement [REM]). Methods: A total of 108 patients with SDB undergoing overnight polysomnography (PSG) were included. A total of 19,862 respiratory events, including obstructive apnea (OA), central apnea (CA), mixed apnea (MA), and hypopnea (HYP), were analyzed using 15 s ECG segments. Linear mixed-effects models (LMMs) and estimated marginal means (EMMs) with Sidak-adjusted pairwise comparisons were constructed to evaluate differences in ECG-derived features and to analyze differences between event types. Results: Central apnea (CA) was associated with significantly reduced HRV and heart rate indices, including Standard Deviation of Successive Differences (SDSD), Root Mean Square of the Successive (RMSSD), Standard Deviation 1 (SD1), and heart rate (HR), compared with other respiratory event types (all p < 0.05). Across all event types, HRV metrics exhibited consistent dynamic changes before, during, and after respiratory events (all p < 0.001), characterized by a decrease during the event followed by post-event recovery. In the interaction effect of sleep stage, SDSD was significantly lower in CA compared with both OA (estimate = −11.67, 95% CI −18.78 to −4.59, p < 0.001) and HYP (estimate = −11.38, 95% CI −18.55 to −4.20, p < 0.001) during NREM sleep. No significant differences in HRV parameters, heart rate, or QRS duration were observed between OA and HYP (all p > 0.05). Conclusions: This study is the first to elucidate the differential impacts of four distinct types of sleep respiratory events on ultra-short-term HRV, confirming that CA events exert the most profound effects on autonomic function. These findings suggest that the proportion of CA occurrences could serve as a more precise biomarker for identifying individuals at high risk for cardiovascular diseases within the SDB population. Full article

This paper presents the design and development of an adaptive augmented reality (AR) assistance system for complex mechanical assembly tasks. Integrating a wrist-worn optical heart rate sensor to evaluate the user’s cognitive state, the system is intended to run as a standalone application on the Meta Quest 3 headset. The system displays instructions and visual cues directly overlaid on the user’s physical workspace and constantly monitors their heart rate variability through the sensor as an estimate of their cognitive load. When the system detects an overload, it dynamically adjusts the presentation of information—for example, it slows down pacing, simplifies instructions, or switches to a different interaction modality (audio)—as an attempt to reduce the overload. The paper makes three contributions: first, it provides a documented standalone integration of physiological sensing with adaptive interface logic on a mixed reality headset without external compute infrastructure; second, it provides a systematic characterization of platform-specific tracking incompatibilities on the Meta Quest 3, documenting the progression through four spatial registration strategies and the specific failure condition that triggered each transition; third, it reports spatial interface design observations from iterative developer testing in the current prototype configuration, including panel height ranges not previously reported in the AR interface literature at this level of specificity. The paper also discusses the within-subjects evaluation protocol that is planned for final system testing with actual users. The work is intended as an engineering and design contribution that establishes the foundation for subsequent empirical evaluation of adaptive AR guidance in industrial assembly contexts. Full article

Background: High-intensity interval training (HIIT) is increasingly implemented in school-based physical education, yet little is known about how exercise intensity changes across repeated sessions and whether such trajectories are associated with physiological adaptations in adolescents. Methods: This study, conducted within the PEER-HEART project (ClinicalTrials.gov: NCT06431230), included 145 adolescents from the experimental arms of a broader school-based trial, stratified by sex and training modality: male HIPT (n = 24), male HIIT (n = 45), female HIPT (n = 46), and female HIIT (n = 30). The 8-week intervention was delivered twice weekly, and exercise intensity was monitored during 16 sessions using heart rate sensors. Intensity trajectories were examined using visual trajectory plots and slope-based statistical analyses, and pre-to-post changes in body fat percentage, systolic and diastolic blood pressure, and predicted VO₂max were analyzed. Results: Visual analyses indicated sex- and modality-specific intensity patterns, with the female HIIT group showing the most dynamic trajectory. Mean changes ranged from −0.63 ± 2.37% to −1.54 ± 2.66% for body fat, from −2.73 ± 5.28 to −5.37 ± 4.32 mmHg for systolic blood pressure, from −0.22 ± 5.00 to −2.62 ± 6.25 mmHg for diastolic blood pressure, and from 0.41 ± 3.25 to 3.81 ± 4.67 mL/kg/min for predicted VO₂max across subgroups. Although most delta values showed no significant main effects, a sex × modality interaction was observed for body fat percentage, and a main effect of sex was observed for predicted VO₂max. Greater intensity progression was associated with larger reductions in diastolic blood pressure in the female HIIT (β = −708.0, p < 0.001, R² = 0.39) and male HIIT groups (β = −377.0, p = 0.014, R² = 0.13) and with greater improvements in predicted VO₂max in the female HIPT (β = 154.0, p = 0.029, R² = 0.104) and male HIIT groups (β = 315.0, p = 0.029, R² = 0.106). Conclusions: Individual intensity trajectories may help explain variability in physiological responses to school-based interval training and may provide additional insight beyond group-level comparisons alone. Full article

Objective: Rapid vertical manoeuvres and intermittent vibration in autonomous electric vertical take-off and landing (eVTOL) aircraft can provoke pronounced psychological anxiety in passengers. To address this, we propose a closed-loop adaptive system that integrates an in-ear wearable sensor with dynamic regulation of the cabin microenvironment, enabling real-time monitoring of each passenger’s autonomic state and delivering individualised mitigation through a continuous sense–analyse–intervene–feedback loop. Methods: The system is built around a pair of custom in-ear modules that integrate dual-wavelength photoplethysmography (PPG; 525 nm green and 940 nm infrared), galvanic skin response (GSR), and a six-axis inertial measurement unit (IMU) sampled at 200 Hz. To suppress the 20–80 Hz vibration generated by the distributed electric propulsion system, a compliant silicone damping sleeve attenuates high-frequency components at the hardware level, while a Kalman filter fuses the IMU and PPG streams and an adaptive notch filter removes residual rotor harmonics. The pipeline raises the heart-rate-variability (HRV) signal-to-noise ratio (SNR) to 24.1 dB, with a Pearson correlation of 0.96 against a medical-grade chest strap. A hybrid CNN–LSTM network—two convolutional layers (32 filters each) followed by two LSTM layers (128 hidden units)—predicts impending anxiety from HRV time-domain features (RMSSD, pNN50) and frequency-domain features (LF/HF ratio), triggering intervention 8.2 s in advance on average. According to the predicted anxiety level (mild/moderate/severe), a fuzzy controller modulates transcutaneous auricular vagus nerve stimulation (1–5 mA), the binaural-beat frequency (4–8 Hz, theta band), and the cabin lighting colour temperature (2700–6500 K) in real time. The intervention parameters are continuously refined by SPSA-based stochastic optimisation of the HRV recovery rate (step size 0.01; updated every 30 s). Results: In a randomised controlled experiment conducted in a simulated flight environment (N = 50; aged 22–45 years; 1:1 sex ratio), the active group reached physiological recovery in 52.3 s on average, compared with 98.6 s for the sham-controlled group—a 47% reduction (Cohen’s d = 1.24, p < 0.001). User acceptance reached 94%. Conclusions: The proposed in-ear platform enables closed-loop adaptive regulation of anxiety in the eVTOL cabin and overcomes the limitations of conventional passive mitigation strategies. By combining vibration-tolerant physiological sensing with multimodal environmental control, the work offers a practical pathway for improving passenger experience in urban air mobility and provides a useful reference for human-factors standards governing autonomous aircraft. Full article

Remote Photoplethysmography (rPPG) provides a non-contact alternative to traditional heart rate monitoring. Estimating physiological signals from facial videos has recently attracted significant research interest. However, rPPG performance is sensitive to illumination variation and environmental interference, which can distort the extracted physiological signal. Since the background and face are affected by similar conditions, the effect of these conditions can be extracted from the background and isolated from the result. This paper proposes the Triple-Head Spatio-Temporal Transformer (TH-STT). TH-STT is a multi-task architecture designed to separate rPPG signals from environmental interference. In addition to facial tokens, a background anchor token is used as an environmental reference. Facial tokens and background anchor are processed using a shared transformer backbone. The proposed architecture has two auxiliary tasks to help purify the resulting rPPG. The Reaction-Driven Gating (RDG) mechanism was introduced, which tracks facial muscular activity. Furthermore, a Dynamic Anchor Locking (DAL) strategy is proposed to cancel environmental illumination interference. Experimental results on three benchmark datasets demonstrate improved and stable performance, with the TH-STT achieving a Mean Absolute Error (MAE) of 0.42 bpm on UBFC-rPPG and 1.08 on COHFACE. Full article