الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 112 |  |  |
| | | from | 112 | | |
Abstract
Hearing loss in children is associated with delayed communication and language development. It leads to poorer educational achievements and later employment prospects. Cochlear implant (CI) is considered a widely accepted rehabilitation choice for those children who receive little or no benefit through traditional amplification methods, and have favorable outcomes in listening, spoken language, literacy, and social/emotional well-being. However, cochlear implantation procedure is more expensive than hearing aids, although health insurance may cover some or all of the costs.
In pediatric cochlear implantation (PCI), there is a common clinical inquiry about ”Who achieved success?”. Each patient’s outcome changes and varies depending on their baseline preoperative assessment and post-operative rehabilitation. Although there is no standardized score for judging the success of the CI process, language development, educational, and social tests are commonly used to determine progress.
None of the previous work addressed the cocktail of variables found in each child in real life. So still the decision of success is made based on subjective judgment of the CI team for each patient with no objective tool (score) controlling this process. But due to the heterogeneity of the variables affecting the CI outcome, setting a scoring system with the ordinary statistical methods was difficult. So we planned to use artificial intelligence (AI) and machine learning (ML) to help us generate this predictive score.
Three outcome measures were assessed; language age deficit, phonological deficit, and social deficit. Classification predictive scores, and regression predictive scores were calculated through ML models.