Neuroscience Institute, Virginia Mason Franciscan Health, Seattle, Washington, USA.
Department of Otolaryngology-Head Neck Surgery, Section of Otology/Neurotology, Virginia Mason Franciscan Health, Seattle, Washington, USA.
Laryngoscope. 2024 Feb;134(2):926-936. doi: 10.1002/lary.30894. Epub 2023 Jul 14.
The aim of the study was to train and test supervised machine-learning classifiers to predict acoustic hearing preservation after CI using preoperative clinical data.
Retrospective predictive modeling study of prospectively collected single-institution CI dataset.
One hundred and seventy-five patients from a REDCap database including 761 patients >18 years who underwent CI and had audiometric testing preoperatively and one month after surgery were included. The primary outcome variable was the lowest quartile change in acoustic hearing at one month after CI using various formulae (standard pure tone average, SPTA; low-frequency PTA, LFPTA). Analysis involved applying multivariate logistic regression to detect statistical associations and training and testing supervised learning classifiers. Classifier performance was assessed with numerous metrics including area under the receiver operating characteristic curve (AUC) and Matthews correlation coefficient (MCC).
Lowest quartile change (indicating hearing preservation) in SPTA was positively associated with a history of meningitis, preoperative LFPTA, and preoperative SPTA. Lowest quartile change in SPTA was negatively associated with sudden hearing loss, noise exposure, aural fullness, and abnormal anatomy. Lowest quartile change in LFPTA was positively associated with preoperative LFPTA. Lowest quartile change in LFPTA was negatively associated with tobacco use. Random forest demonstrated the highest mean classification performance on the validation dataset when predicting each of the outcome variables.
Machine learning demonstrated utility for predicting preservation of residual acoustic hearing in patients undergoing CI surgery, and the detected associations facilitated the interpretation of our machine-learning models. The models and statistical associations together may be used to facilitate improvements in shared clinical decision-making and patient outcomes.
3 Laryngoscope, 134:926-936, 2024.
本研究旨在利用术前临床数据训练和测试监督机器学习分类器，以预测人工耳蜗植入后患者的听力保留情况。
回顾性预测建模研究，对来自前瞻性收集的单一机构人工耳蜗植入数据集进行分析。
纳入了来自 REDCap 数据库的 175 名患者，这些患者年龄均大于 18 岁，共 761 名患者接受了人工耳蜗植入，并在术前和术后一个月进行了听力测试。主要结局变量是使用各种公式（标准纯音平均，SPTA；低频纯音平均，LFPTA）计算得出的术后一个月内听力最低四分位变化。分析包括应用多元逻辑回归来检测统计学关联，并训练和测试监督学习分类器。使用多种指标评估分类器性能，包括接收者操作特征曲线下的面积（AUC）和马修斯相关系数（MCC）。
SPTA 的最低四分位变化（表示听力保留）与脑膜炎病史、术前 LFPTA 和术前 SPTA 呈正相关。SPTA 的最低四分位变化与突发性听力损失、噪声暴露、耳闷和异常解剖呈负相关。LFPTA 的最低四分位变化与术前 LFPTA 呈正相关。LFPTA 的最低四分位变化与吸烟呈负相关。随机森林在预测每个结局变量时，在验证数据集中表现出最高的平均分类性能。
机器学习在预测人工耳蜗植入患者残余听力保留方面具有一定的应用价值，所检测到的关联有助于我们对机器学习模型的解释。这些模型和统计关联可一起用于促进共同的临床决策制定和改善患者结局。
3 级喉镜，134：926-936，2024。
