Large language model–based prediction of speech intelligibility after Vibrant Soundbridge implantation using multidimensional outcome data: Part 2 of a prospective study

Abstract

Abstract Active middle ear implants (AMEIs) such as the Vibrant Soundbridge (VSB) offer an effective treatment option for patients with mixed or conductive hearing loss and large air–bone gaps, where conventional hearing rehabilitation often fails. However, postoperative outcomes—particularly speech intelligibility at 65 dB SPL in free-field (WRS 65dB )—show high interindividual variability. This study aimed to develop a predictive model for WRS 65dB based on four clinically relevant parameters: postoperative bone conduction thresholds (BC PTA4 ), unaided preoperative maximum speech intelligibility (WRS max ), Vibrogram threshold (VIB PTA4 ), and age. Data from 20 patients were analyzed. Spearman’s correlation revealed significant associations between WRS 65dB and postoperative BC PTA4 , preoperative WRS max , and age. Using a seven-step approach supported by GPT-4o, we developed a sigmoid-transformed linear regression model. The final model included BC PTA4 , WRS max , and age and achieved an R² of 0.51, r  = 0.71, RMSE = 6.18, and MAE = 4.67. Model performance was assessed by means of residual and outlier analysis. This model provides a transparent and clinically applicable tool for preoperative outcome estimation in VSB candidates. Further validation in larger, multicenter cohorts is needed to confirm its generalizability.