Predicting Common Audiological Functional Parameters (CAFPAs) as Interpretable Intermediate Representation in a Clinical Decision-Support System for Audiology

Abstract

The application of machine learning for the development of clinical decision-support systems in audiology provides the potential to improve the objectivity and precision of clinical experts' diagnostic decisions. However, for successful clinical application, such a tool needs to be accurate, as well as accepted and trusted by physicians. In the field of audiology, large amounts of patients' data are being measured, but these are distributed over local clinical databases and are heterogeneous with respect to the applied assessment tools. For the purpose of integrating across different databases, the Common Audiological Functional Parameters (CAFPAs) were recently established as abstract representations of the contained audiological information describing relevant functional aspects of the human auditory system. As an intermediate layer in a clinical decision-support system for audiology, the CAFPAs aim at maintaining interpretability to the potential users. Thus far, the CAFPAs were derived by experts from audiological measures. For designing a clinical decision-support system, in a next step the CAFPAs need to be automatically derived from available data of individual patients. Therefore, the present study aims at predicting the expert generated CAFPA labels using three different machine learning models, namely the lasso regression, elastic nets, and random forests. Furthermore, the importance of different audiological measures for the prediction of specific CAFPAs is examined and interpreted. The trained models are then used to predict CAFPAs for unlabeled data not seen by experts. Prediction of unlabeled cases is evaluated by means of model-based clustering methods. Results indicate an adequate prediction of the ten distinct CAFPAs. All models perform comparably and turn out to be suitable choices for the prediction of CAFPAs. They also generalize well to unlabeled data. Additionally, the extracted relevant features are plausible for the respective CAFPAs, facilitating interpretability of the predictions. Based on the trained models, a prototype of a clinical decision-support system in audiology can be implemented and extended towards clinical databases in the future.

Interpretable Clinical Decision Support System for Audiology Based on Predicted Common Audiological Functional Parameters (CAFPAs)

“…For the purpose of providing a quantitative link between audiological measures and CAFPAs, Saak et al [ 33 ] established regression models based on the data set [ 34 ] and the expert-estimated CAFPAs from [ 31 ]. To incorporate different degrees of interpretability vs. flexibility of the models, the regression was performed using lasso regression, elastic nets and random forests [ 39 ].…”

“…Hence, CAFPAs could not be estimated for new patients. Therefore, as a next step towards a CDSS operable for individual patients, Saak et al [ 33 ] established regression models that allow automatic prediction of CAFPAs given the measurement outcomes (cf. Figure 1 B, middle and right part).…”

“…Figure 1 B, middle and right part). Three different models (lasso regression, elastic net, and random forest) were investigated in [ 33 ], and it was shown that all models provide adequate to good predictive performance. In addition, measurements employed in the prediction of CAFPAs were analyzed by means of feature importance, which provided interpretability of the CAFPA prediction and revealed audiologically plausible relationships between measurement outcomes and CAFPAs [ 33 ].…”

“…Therefore, by combining these previous steps [ 32 , 33 ], the current study aims at exploring whether a CDSS can be constructed for audiology which is as interpretable as expert knowledge-based classification and as data-driven as machine learning-based classification. In the CDSS, CAFPAs are to be estimated from the available audiological measurements, and classification of diagnostic cases (audiological findings or treatment recommendations) needs to be performed based on predicted CAFPAs (cf.…”

“…Figure 1 B, right part). In addition, the interpretability of the CDSS needs to be assured, which can partly be combined from the interpretability aspects from the classification of [ 32 ] and CAFPA prediction of [ 33 ].…”