Predicting the outcomes of ICBT with ANN and SVM (Rodrigo et al., 2022)

Purpose: Internet-based cognitive behavioral therapy (ICBT) has been found to be effective for tinnitus management, although there is limited understanding about who will benefit the most from ICBT. Traditional statistical models have largely failed to identify the nonlinear associations and hence find strong predictors of success with ICBT. This study aimed at examining the use of an artificial neural network (ANN) and support vector machine (SVM) to identify variables associated with treatment success in ICBT for tinnitus.
Method: The study involved a secondary analysis of data from 228 individuals who had completed ICBT in previous intervention studies. A 13-point reduction in Tinnitus Functional Index (TFI) was defined as a successful outcome. There were 33 predictor variables, including demographic, tinnitus, hearing-related and treatment-related variables, and clinical factors (anxiety, depression, insomnia, hyperacusis, hearing disability, cognitive function, and life satisfaction). Predictive models using ANN and SVM were developed and evaluated for classification accuracy. SHapley Additive exPlanations (SHAP) analysis was used to identify the relative predictor variable importance using the best predictive model for a successful treatment outcome.
Results: The best predictive model was achieved with the ANN with an average area under the receiver operating characteristic value of 0.73 ± 0.03. The SHAP analysis revealed that having a higher education level and a greater baseline tinnitus severity were the most critical factors that influence treatment outcome positively.
Conclusions: Predictive models such as ANN and SVM help predict ICBT treatment outcomes and identify predictors of outcome. However, further work is needed to examine predictors that were not considered in this study as well as to improve the predictive power of these models.
Supplemental Material S1. Characteristics of the study 740 participants and summary statistics for successful and unsuccessful treatment groups. Quantitative variables have been analyzed using a two-sample t test, while categorical data have been analyzed with chi-square or Fisher’s exact (denoted by an asterisk [*]). A threshold of .05 has been used.
Supplemental Material S2. Demographic variables (7 variables).
Supplemental Material S3. Tinnitus and hearing-related variables (15 variables).
Supplemental Material S4. Treatment-related variables (4 variables).
Supplemental Material S5. Clinical factors (7 variables).
Rodrigo, H., Beukes, E. W., Andersson, G., & Manchaiah, V. (2022). Predicting the outcomes of internet-based cognitive behavioral therapy for tinnitus: Applications of artificial neural network and support vector machine. American Journal of Audiology. Advance online publication. https://doi.org/10.1044/2022_AJA-21-00270

目的：基于互联网的认知行为疗法（ICBT）在耳鸣管理方面已被证实具有显著效果，尽管关于谁将最大程度地从ICBT中受益的理解仍显不足。传统的统计模型在识别非线性关联方面大多未能奏效，因此也未能找到ICBT成功预测的强有力指标。本研究旨在探讨人工神经网络（ANN）和支撑向量机（SVM）在识别与ICBT耳鸣治疗成功相关的变量中的应用。方法：本研究对先前干预研究中完成ICBT的228名个体数据进行二次分析。将耳鸣功能指数（TFI）降低13点定义为成功的结果。共有33个预测变量，包括人口统计学、耳鸣、听力相关和治疗相关变量，以及临床因素（焦虑、抑郁、失眠、过度敏感、听力障碍、认知功能和生命满意度）。开发了使用ANN和SVM的预测模型，并对其分类精度进行评估。使用SHapley Additive exPlanations（SHAP）分析识别最佳预测模型中相对预测变量的重要性。结果：最佳预测模型为ANN，其平均受试者工作特征曲线下面积为0.73 ± 0.03。SHAP分析揭示，较高的教育水平和较高的基线耳鸣严重程度是影响治疗结果的最关键因素。结论：如ANN和SVM等预测模型有助于预测ICBT的治疗结果并识别结果预测因素。然而，仍需进一步研究本研究所未考虑的预测因素，以及提高这些模型的预测能力。补充材料S1：研究特征740名参与者以及成功和未成功治疗组的汇总统计数据。定量变量使用双样本t检验进行分析，而分类数据使用卡方检验或Fisher确切检验（以星号[*]表示）。阈值设为0.05。补充材料S2：人口统计学变量（7个变量）。补充材料S3：耳鸣和听力相关变量（15个变量）。补充材料S4：治疗相关变量（4个变量）。补充材料S5：临床因素（7个变量）。Rodrigo, H., Beukes, E. W., Andersson, G., & Manchaiah, V. (2022). 基于互联网认知行为疗法耳鸣治疗效果的预测：人工神经网络和支撑向量机的应用。美国听觉学杂志。在线预发表。https://doi.org/10.1044/2022_AJA-21-00270