Carbohydrate supplementation is a common practice among endurance athletes participating in long-distance competitions. However, glycogen storage regulation, in-competition blood glucose levels, and their relationship with athletic performance are influenced by multiple factors. This review summarizes the recent research progress on carbohydrate supplementation, addressing its applications in the pre-, during-, and post-competition phases. It explores variables that influence the effectiveness of carbohydrate supplementation and provides a summary of strategies, based on six key aspects: carbohydrate properties, multi-nutrient interactions, gastrointestinal function, individual differences (such as age and gender), environmental conditions, and psychological factors. The combination of different types, ratios, and concentrations of carbohydrates has been demonstrated to enhance the efficiency of carbohydrate digestion and absorption. The synergistic combination of protein, sodium, and caffeine intake demonstrates enhanced efficacy in carbohydrate supplementation strategies. Gastrointestinal tolerance training for carbohydrate supplementation has been identified as an effective measure to alleviate gastrointestinal discomfort during high-dose carbohydrate intake. The adjustment of the carbohydrate-to-fat ratio and the type of carbohydrate intake has been found to mitigate the impact of gender and menstrual cycles on glycogen storage and substrate utilization. Modifying the timing of glycogen storage and regulating the concentration and temperature of carbohydrate solutions during competition have been demonstrated to facilitate coping with the elevated energy expenditure and metabolic substrate shift from fat to carbohydrates, triggered by a combination of environmental and psychological factors, including special environmental and climatic conditions (e.g., high altitude, high temperature, high humidity, and cold) and emotional states (e.g., pre-competition stress and anxiety during the competition). To achieve precise carbohydrate supplementation for athletes in major events under various competitive environments, it is necessary to quantitatively assess the effects of carbohydrate supplementation, supported by mechanistic studies. This can be achieved by utilizing wearable devices to monitor the entire competition, coupled with data collection technologies, such as high-throughput profiling. Furthermore, emerging data analytics techniques, such as machine learning and causal inference, should be leveraged to refine supplementation strategies.

Objective: This study developed and evaluated a machine learning model for predicting oxygen uptake (VO ₂ ) in dragon boat athletes using wearable signals and compared performance across signal combinations and device groups. Methods: Twelve male collegiate dragon boat athletes completed nine tests grouped into six task types (T1–T6): 200-, 500-, and 1000-m all-out paddling tests; an incremental test; a five-stage constant-load test; and four supramaximal-intensity tests pooled as T6. Heart rate, respiratory rate, device-derived minute ventilation, and muscle-oxygenation signals were collected. Fifteen signal combinations (G1–G15) and six device groups (A–F) were evaluated using a multilayer perceptron model. Performance was assessed using mean absolute error (MAE) and root mean square error (RMSE), with mean absolute percentage error (MAPE) also reported. Results: G13–G15 showed comparable performance, with no significant differences. G14 yielded the lowest MAE and MAPE (2.21 mL·kg ⁻¹ ·min ⁻¹ and 9.48%), whereas G13 yielded the lowest RMSE (3.19 mL·kg ⁻¹ ·min ⁻¹ ). No significant differences were observed among signal combinations within Groups A, C, D, E, or F. Group F showed the lowest MAPE but did not differ significantly from Group B, whereas Group C showed significantly higher MAPE than Groups B, D, E, and F. Performance did not differ significantly across exercise tasks. During the 1000-m test, predicted and measured VO ₂ were strongly correlated for G15 ( r = 0.88), with a mean prediction error of 0.44 mL·kg ⁻¹ ·min ⁻¹ and descriptive 95% limits of agreement from −6.32 to 7.20 mL·kg ⁻¹ ·min ⁻¹ . Conclusion: Cardiorespiratory signals provided useful information for wearable-based VO ₂ prediction, whereas the contribution of muscle-oxygenation signals depended on the input configuration. G13–G15 should be regarded as comparably performing configurations, and bilateral muscle-oxygenation monitoring was not consistently superior to unilateral monitoring. Multimodal wearable signals show potential for non-invasive VO ₂ monitoring in dragon boat athletes, although larger samples and independent validation are required.