PURPOSE: The purpose of this study was to employ machine learning (ML) in order to identify individuals who currently report feelings of anxiety using self-paced walking gait and balance assessments. METHODS: Participants (n = 87) completed the testing in a single session. Self-reported feelings of anxiety were measured using the Profile of Mood States- Short Form (POMS-SF), the modified Clinical Test of Sensory Interaction and Balance (mCTSIB) was used to assess balance in 4 conditions (eyes open (EO), feet on ground (FI), eyes closed (EC) FI, EO foam surface (FO) and EC FO) and a self-paced 2-minute walking gait around a 6 m track was measured using the APDM mobility monitors. Participants were grouped into anxious (A) (n = 36, age = 23.31 ± 3.69 yrs, height = 174.13 ± 9.76 cm, weight = 74.86 ± 14.26 kg, male = 14), and not anxious (NA) (n = 51, age = 25.20 ± 4.24, height = 174.13 ± 9.76 cm, weight = 73.00 ± 15.83 kg, male = 21). The models were trained using classifiers with all features and top 5 features (using a 0.024 cut-off for feature importance). The models were trained in a 10-fold cross-validation manner in order to avoid problems such as overfitting or selection bias. Data was then randomly split into training set (90%) and test set (10%) and ran through each of the ML models 10,000 times. RESULTS: The top model was a Random Forrest with the top 5 features. The median accuracy of that model was 75% and the mean was 69.7%. Approximately 3% of the models had 100% accuracy. The top five features were mean angles of turns, variance of neck bending in the frontal plane, variance in arm swing speed, movement of the lumbar region in the sagittal plane and the maximal lumbar rotation in the transverse plane. CONCLUSIONS: The findings of this study suggest that gait variables are the most important variables in identifying individuals who report feelings of anxiety. The median accuracy of the models suggests that studies with a larger pool of subjects may yield significantly better models in order to identify individuals who feel anxious when using walking gait analysis.