Introduction: The acquisition of blood lactate concentration (BLC) during exercise is beneficial for endurance training, yet a convenient method to measure it remains unavailable. BLC and electrocardiogram (ECG) both exhibit variations with changes in exercise intensity and duration. In this study, we hypothesized that BLC during exercise can be predicted using ECG data.Methods: Thirty-one healthy participants underwent four cardiopulmonary exercise tests, including one incremental test and three constant work rate (CWR) tests at low, moderate, and high intensity. Venous blood samples were obtained immediately after each CWR test to measure BLC. A mathematical model was constructed using 31 trios of CWR tests, which utilized a residual network combined with long short-term memory to analyze every beat of lead II ECG waveform as 2D images. An artificial neural network was used to analyze variables such as the RR interval, age, sex, and body mass index.Results: The standard deviation of the fitting error was 0.12 mmol/L for low and moderate intensities, and 0.19 mmol/L for high intensity. Weighting analysis demonstrated that ECG data, including every beat of ECG waveform and RR interval, contribute predominantly.Conclusion: By employing 2D convolution and artificial neural network-based methods, BLC during exercise can be accurately estimated non-invasively using ECG data, which has potential applications in exercise training.

We hypothesized that blood lactate concentration([Lac]_blood) is a function of cardiopulmonary variables, exercise intensity and some anthropometric elements during aerobic exercise. This investigation aimed to establish a mathematical model to estimate [Lac]_blood noninvasively during constant work rate (CWR) exercise of various intensities. 31 healthy participants were recruited and each underwent 4 cardiopulmonary exercise tests: one incremental and three CWR tests (low: 35% of peak work rate for 15 min, moderate: 60% 10 min and high: 90% 4 min). At the end of each CWR test, venous blood was sampled to determine [Lac]_blood. 31 trios of CWR tests were employed to construct the mathematical model, which utilized exponential regression combined with Taylor expansion. Good fitting was achieved when the conditions of low and moderate intensity were put in one model; high-intensity in another. Standard deviation of fitting error in the former condition is 0.52; in the latter is 1.82 mmol/liter. Weighting analysis demonstrated that, besides heart rate, respiratory variables are required in the estimation of [Lac]_blood in the model of low/moderate intensity. In conclusion, by measuring noninvasive cardio-respiratory parameters, [Lac]_blood during CWR exercise can be determined with good accuracy. This should have application in endurance training and future exercise industry.

Background: Low peak oxygen consumption (V̇O ₂ ) is associated with higher cardiovascular and all-cause mortality, while improvements in peak V̇O ₂ reduce this risk. Although early detection allows timely intervention, practical screening tools remain lacking. As electrocardiograms (ECGs) reflect both cardiac and age-related changes, they may offer a viable screening approach. Objective: This study aimed to detect low peak V̇O ₂ using resting 12-lead ECGs analyzed by a trained neural network. Methods: The low peak V̇O ₂ estimation model was developed using data from 965 individuals (n=540, 56% with cardiac or pulmonary disorders) mainly at Chang Gung Memorial Hospital, Linkou, and validated in an independent cohort of 242 individuals (n=194, 80% with cardiac disorders) at the Keelung branch. Resting ECGs were recorded immediately before cardiopulmonary exercise testing. Low peak V̇O ₂ was defined as a peak V̇O ₂ of <14 mL/kg/min. Results: The mean peak V̇O ₂ was 17.5 (SD 6.1) and 15.4 (SD 3.9) mL/kg/min in the training and validation datasets, with 27% (261/965) and 38% (92/242) classified as low peak V̇O ₂ , respectively. Wavelet analysis improved model accuracy, underscoring its value for feature extraction. Three input models were compared: (1) individual characteristics (IC; age, sex, BMI, resting heart rate, and heart rate variability), (2) ECG alone, and (3) ECG plus IC. ECG alone outperformed IC, and combining both yielded the highest accuracy. For low peak V̇O ₂ prediction, ECG plus IC achieved mean area under the curve, precision, and recall values of 0.89, 0.72, and 0.72 in cross-validation, and 0.87, 0.67, and 0.61 in external validation. Conclusions: An artificial intelligence–driven ECG-based algorithm showed strong potential for screening low peak V̇O ₂ , enabling early identification of individuals with low peak V̇O ₂ and facilitating timely clinical intervention.