In this study, we sought to evaluate the efficiencies of multiple machine learning algorithms in detecting neonates’ Frequency Following Responses (FFRs). We recorded continuous brainwaves from 43 American neonates in response to a pre-recorded monosyllable/i/with a rising frequency contour. Recordings were classified into response and no response categories. Six response features were extracted from each recording and served as predictors in FFR identification. Twenty-three supervised machine learning algorithms with mean efficiency values of 86.0%, 94.4%, 97.2%, and 97.5% when 1, 10, 100, and 1000 random iterations were implemented, respectively. These high efficiency values obtained from the neonatal FFRs demonstrate that machine learning algorithms can help assess pitch processing in neonates and can be applied to auditory screening and intervention services for neonates at risk for disorders associated with decreased pitch processing.

No abstract

To improve the efficiency and timeliness in frequency-following response (FFR) testing, the purpose of this study was to investigate the capabilities of machine learning in the detection of an FFR. Continuous brain waves were recorded from 25 Chinese adults in response to a pre-recorded Mandarin monosyllable \yi2\ with a rising frequency contour. A total of 8000 artifact-free sweeps were recorded from each participant. Continuous brain waves accumulated (from the first sweep) up to the first 500 sweeps were considered FFR absent; brain waves accumulated (from the first sweep) up to the last 1000 sweeps (i.e., from 7001 to 8000 sweeps) were considered FFR present. Six response features (frequency error, slope error, tracking accuracy, spectral amplitude, pitch strength, and root-mean-square amplitude) were extracted from each recording and served as key predictors in the identification of a response. Twenty-three supervised machine-learning algorithms, with a 10-fold cross-validation procedure, were implemented via a Classification Learner App in MATLAB. Two algorithms yielded 100% efficiency (i.e., 100% sensitivity and 100% specificity) and 14 others produced efficiency ≥99%. Results indicated that a majority of the machine-learning algorithms provided efficient and accurate predictions in whether an FFR was present or absent in a recording.

No abstract

No abstract