Machine learning-based longitudinal prediction for GJB2-related sensorineural hearing loss

Pey-Yu Chen, Ta-Wei Yang, Yi-Shan Tseng, Cheng-Yu Tsai, Chiung-Szu Yeh, Yen-Hui Lee, Pei-Hsuan Lin, Ting-Chun Lin, Yu-Jen Wu, Ting-Hua Yang, Yu-Ting Chiang, Jacob Shujui Hsu, Chuan-Jen Hsu, Pei-Lung Chen, Cheng-Fu Chou, Chen-Chi Wu

Abstract: Highlights•GJB2 mutations are a leading cause of hearing loss worldwide.•Slow progression is a hallmark of GJB2-related hearing impairment.•Machine learning allows accurate prediction hearing outcome of GJB2-related hearing impairment.•The LSTM model accurately predicts hearing thresholds up to 4 years.