Recommendations for marathon runners: on the application of recommender systems and machine learning to support recreational marathon runners

Deep fit_predic: a novel integrated pyramid dilation EfficientNet-B3 scheme for fitness prediction system

This study introduces novel deep learning (DL) techniques for effective fitness prediction using a person's health data. Initially, pre-processing is performed in which data cleaning, one-hot encoding and data normalization are performed. The pre-processed data are then fed into the feature selection stage, where the useful features are extracted using the enhanced chameleon swarm (ECham-Sw) optimization technique. Then, a clustering process is performed using Minkowski integrated gravity center clustering (Min-GCC) to cluster the health profiles of each individual. Finally, the Pyramid Dilated EfficientNet-B3 (PyDi-EfficientNet-B3) technique is proposed to predict the fitness of each individual efficiently with enhanced accuracy of 99.8%.

Training patterns associated with injury in New York City Marathon runners

OBJECTIVE

Training patterns are commonly implicated in running injuries. The purpose of this study was to measure the incidence of injury and illness among marathon runners and the association of injuries with training patterns and workload.

METHODS

Runners registered for the New York City Marathon were eligible to enrol and prospectively monitored during the 16 weeks before the marathon, divided into 4-week 'training quarters' (TQ) numbered TQ1-TQ4. Training runs were tracked using Strava, a web and mobile platform for tracking exercise. Runners were surveyed at the end of each TQ on injury and illness, and to verify all training runs were recorded. Acute:chronic workload ratio (ACWR) was calculated by dividing the running distance in the past 7 days by the running distance in the past 28 days and analysed using ratio thresholds of 1.3 and 1.5.

RESULTS

A total of 735 runners participated, mean age 41.0 (SD 10.7) and 46.0% female. Runners tracked 49 195 training runs. The incidence of injury during training was 40.0% (294/735), and the incidence of injury during or immediately after the marathon was 16.0% (112/699). The incidence of illness during training was 27.2% (200/735). Those reporting an initial injury during TQ3 averaged less distance/week during TQ2 compared with uninjured runners, 27.7 vs 31.9 miles/week (p=0.018). Runners reporting an initial injury during TQ1 had more days when the ACWR during TQ1 was ≥1.5 compared with uninjured runners (injured IQR (0-3) days vs uninjured (0-1) days, p=0.009). Multivariable logistic regression for training injuries found an association with the number of days when the ACWR was ≥1.5 (OR 1.06, 95% CI (1.02 to 1.10), p=0.002).

CONCLUSION

Increases in training volume ≥1.5 ACWR were associated with more injuries among runners training for a marathon. These findings can inform training recommendations and injury prevention programmes for distance runners.