Application of Machine Learning to Predict Hearing Outcomes of Tympanoplasty

Abstract

ObjectiveThis retrospective study aimed to evaluate the performance of machine learning techniques in predicting air‐bone gap after tympanoplasty compared with conventional scoring models and to identify the influential factors.MethodsWe reviewed the charts of 105 patients (114 ears) with chronic otitis media who underwent tympanoplasty. Two numerical scoring systems (middle ear risk index [MERI] and ossiculoplasty outcome parameter staging [OOPS]) and three algorithms (random forest [RF], support vector machine [SVM], and k nearest neighbor [kNN]) were created. Experimental variables included age, preoperative air‐bone gap, soft‐tissue density lesion in the tympanic cavity in CT, otorrhea, surgical history, ossicular bone problems in CT, tympanic perforation location, perforation type (central or marginal), grafting material, smoking history, endoscopy use, and the operator whose experience was 20 years or longer, or shorter. Binary classification, postoperative air‐bone gap ≤15 or >15 dB, and multiclass classification, classification into seven categories by 10 dB, were performed, and the percentages of correct prediction were calculated. The importance of features in the RF model was calculated to identify influential factors.ResultsThe percentages of correct prediction in binary classification were 62.3%, 72.8%, 81.5%, 81.5%, and 81.5% in MERI, OOPS, RF, SVM, and kNN, respectively, and those in multiclass classification were 29.8%, 21.9%, 63.1%, 44.7%, and 50% in the same order. The RF model suggested larger preoperative air‐bone gap, and older age could make the postoperative air‐bone gap larger.ConclusionThe machine learning techniques, especially the RF model, are promising methods for precise postoperative air‐bone gap prediction.Level of Evidence4 Laryngoscope, 133:2371–2378, 2023

Prediction of Hearing Prognosis After Intact Canal Wall Mastoidectomy With Tympanoplasty Using Artificial Intelligence

“…Few studies have attempted to predict hearing outcomes after surgery using machine learning. 23 In contrast to previous studies that assessed only a few factors related to hearing outcomes, the present study analyzed over 60 factors through artificial intelligence, thereby determining their significance. By employing a machine learning technique, we were able to effectively capture the relationships between these factors and outcomes.…”

“…In our study, we leveraged this technology to predict hearing outcomes after tympanomastoidectomy. Few studies have attempted to predict hearing outcomes after surgery using machine learning 23 . In contrast to previous studies that assessed only a few factors related to hearing outcomes, the present study analyzed over 60 factors through artificial intelligence, thereby determining their significance.…”

Diagnosis, Treatment, and Management of Otitis Media with Artificial Intelligence

“…Although many epidemiological studies have been conducted on OM as a common respiratory complication in children, few studies have been able to analyse big data so that symptom severity, longitudinal patterns and variance in OM patients related to age and seasonality can be accurately described. In order to plan primary health care services and individualised treatments effectively, these factors are important ( Supplementary Table S6 ) [ 93 , 94 , 95 ]. NLP was used by Anthony Dowell et al, to assess primary care incidence and service utilisation of childhood respiratory illnesses in 77,582 children.…”

“…They found that the correct prediction percentages of their algorithm for predicting postoperative ABG ≤ 15 or >15 dB were 81.5%. They also determined the factors, such as age and ABG, influencing the outcome of tympanoplasty [ 95 ]. ENT specialists have also become interested in the study of intelligent follow-up regarding the presence and patency of tympanostomy tubes following their placement.…”