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

BACKGROUND: Recessive GJB2 variants, the most common genetic cause of hearing loss, may contribute to progressive sensorineural hearing loss (SNHL). The aim of this study is to build a realistic predictive model for GJB2-related SNHL using machine learning to enable personalized medical planning for timely intervention. METHOD: Patients with SNHL with confirmed biallelic GJB2 variants in a nationwide cohort between 2005 and 2022 were included. Different data preprocessing protocols and computational algorithms were combined to construct a prediction model. We randomly divided the dataset into training, validation, and test sets at a ratio of 72:8:20, and repeated this process ten times to obtain an average result. The performance of the models was evaluated using the mean absolute error (MAE), which refers to the discrepancy between the predicted and actual hearing thresholds. RESULTS: We enrolled 449 patients with 2184 audiograms available for deep learning analysis. SNHL progression was identified in all models and was independent of age, sex, and genotype. The average hearing progression rate was 0.61 dB HL per year. The best MAE for linear regression, multilayer perceptron, long short-term memory, and attention model were 4.42, 4.38, 4.34, and 4.76 dB HL, respectively. The long short-term memory model performed best with an average MAE of 4.34 dB HL and acceptable accuracy for up to 4 years. CONCLUSIONS: We have developed a prognostic model that uses machine learning to approximate realistic hearing progression in GJB2-related SNHL, allowing for the design of individualized medical plans, such as recommending the optimal follow-up interval for this population.

Revisiting Genetic Epidemiology with a Refined Targeted Gene Panel for Hereditary Hearing Impairment in the Taiwanese Population.

Hearing impairment is one of the most common sensory disorders in children, and targeted next-generation sequencing (NGS)-based genetic examinations can assist in its prognostication and management. In 2020, we developed a simplified 30-gene NGS panel from the original 214-gene NGS version based on Taiwanese genetic epidemiology data to increase the accessibility of NGS-based examinations. In this study, we evaluated the diagnostic performance of the 30-gene NGS panel and compared it with that of the original 214-gene NGS panel in patient subgroups with different clinical features. Data on the clinical features, genetic etiologies, audiological profiles, and outcomes were collected from 350 patients who underwent NGS-based genetic examinations for idiopathic bilateral sensorineural hearing impairment between 2020 and 2022. The overall diagnostic yield was 52%, with slight differences in genetic etiology between patients with different degrees of hearing impairment and ages of onset. No significant difference was found in the diagnostic yields between the two panels, regardless of clinical features, except for a lower detection rate of the 30-gene panel in the late-onset group. For patients with negative genetic results, where the causative variant is undetectable on current NGS-based methods, part of the negative results may be due to genes not covered by the panel or yet to be identified. In such cases, the hearing prognosis varies and may decline over time, necessitating appropriate follow-up and consultation. In conclusion, genetic etiologies can serve as references for refining targeted NGS panels with satisfactory diagnostic performance.

Comprehensive Etiologic Analyses in Pediatric Cochlear Implantees and the Clinical Implications.

Cochlear implantation is the treatment of choice for children with profound sensorineural hearing impairment (SNHI), yet the outcomes of cochlear implants (CI) vary significantly across individuals. To investigate the CI outcomes in pediatric patients with SNHI due to various etiologies, we prospectively recruited children who underwent CI surgery at two tertiary referral CI centers from 2010 to 2021. All patients underwent comprehensive history taking, next generation sequencing (NGS)-based genetic examinations, and imaging studies. The CI outcomes were evaluated using Categories of Auditory Performance (CAP) and Speech Intelligibility Rating (SIR) scores. Of the 160 pediatric cochlear implantees (76 females and 84 males) included in this study, comprehensive etiological work-up helped achieve clinical diagnoses in 83.1% (133/160) of the patients, with genetic factors being the leading cause (61.3%). Imaging studies identified certain findings in 31 additional patients (19.3%). Four patients (2.5%) were identified with congenital cytomegalovirus infection (cCMV), and 27 patients (16.9%) remained with unknown etiologies. Pathogenic variants in the four predominant non-syndromic SNHI genes (i.e., SLC26A4, GJB2, MYO15A, and OTOF) were associated with favorable CI outcomes (Chi-square test, p = 0.023), whereas cochlear nerve deficiency (CND) on imaging studies was associated with unfavorable CI outcomes (Chi-square test, p < 0.001). Our results demonstrated a clear correlation between the etiologies and CI outcomes, underscoring the importance of thorough etiological work-up preoperatively in pediatric CI candidates.