<![CDATA[Subjective, retrospection-based questionnaires of self-reports are not sufficient sources for early signs of psychological and fatigue conditions in athletes which opens the possibility of digital biomarkers. The review is on the basis of the PRISMA 2020 framework of the 799 records initially found through Scopus, of which 2 duplicates were excluded resulting in 797, among whom, 44 were considered for the inclusion of 16 and the rest were the exclusion of abstract titles. Papers in English describing sensor-based detection of psychological, affective, or fatigue states in sport or exercise were selected. Electrocardiograpy was the main modality used alongside heartbeat rate variability with the most studies focusing on the detection of fatigue and stress states, and there was an overall increase in the use of deep and machine-learning classifiers with some studies having classification accuracy beyond 90%. Multimodal fusion gave better results compared to the single-signal method. The limitations to the actual implementation of these systems were mainly the use of small sample sizes, weak external validation, inconsistent labelling of the reference signal. This review presents modalities and stresses the importance of standardized protocols, ecological validity, and transparent reporting both for researchers and practitioners.]]>
Copyrights © 2025
