An Automated Speech-in-Noise Test for Remote Testing: Development and Preliminary Evaluation

Breaking Sound Barriers: Exploring Tele-Audiology's Impact on Hearing Healthcare

Hearing impairment is a global issue, affecting billions of people; however, there is a gap between the population affected by hearing loss and those able to access hearing healthcare. Tele-audiology, the application of telemedicine in audiology, serves as a new form of technology which aims to provide synchronous or asynchronous hearing healthcare. In this article, we reviewed some recent studies of tele-audiology-related topics to have a glimpse of the current development, associated challenges, and future advancement. Through the utilization of tele-audiology, patients can conveniently access hearing healthcare, and thus save travel costs and time. Recent studies indicate that remote hearing screening and intervention are non-inferior to the performance of traditional clinical pathways. However, despite its potential benefits, the implementation of tele-audiology faces numerous challenges, and audiologists have varying attitudes on this technology. Overcoming obstacles such as high infrastructure costs, limited reimbursement, and the lack of quality standards calls for concerted efforts to develop effective strategies. Ethical concerns, reimbursement, and patient privacy are all crucial aspects requiring in-depth discussion. Enhancing the education and training of students and healthcare workers, along with providing relevant resources, will contribute to a more efficient, systematic hearing healthcare. Future research will aim to develop integrated models with evidence-based protocols and incorporating AI to enhance the affordability and accessibility of hearing healthcare.

Characterization of Synthetic Health Data Using Rule-Based Artificial Intelligence Models

The aim of this study is to apply and characterize eXplainable AI (XAI) to assess the quality of synthetic health data generated using a data augmentation algorithm. In this exploratory study, several synthetic datasets are generated using various configurations of a conditional Generative Adversarial Network (GAN) from a set of 156 observations related to adult hearing screening. A rule-based native XAI algorithm, the Logic Learning Machine, is used in combination with conventional utility metrics. The classification performance in different conditions is assessed: models trained and tested on synthetic data, models trained on synthetic data and tested on real data, and models trained on real data and tested on synthetic data. The rules extracted from real and synthetic data are then compared using a rule similarity metric. The results indicate that XAI may be used to assess the quality of synthetic data by (i) the analysis of classification performance and (ii) the analysis of the rules extracted on real and synthetic data (number, covering, structure, cut-off values, and similarity). These results suggest that XAI can be used in an original way to assess synthetic health data and extract knowledge about the mechanisms underlying the generated data.

Evaluation of Machine Learning Algorithms and Explainability Techniques to Detect Hearing Loss From a Speech-in-Noise Screening Test

PURPOSE

The aim of this study was to analyze the performance of multivariate machine learning (ML) models applied to a speech-in-noise hearing screening test and investigate the contribution of the measured features toward hearing loss detection using explainability techniques.

METHOD

Seven different ML techniques, including transparent (i.e., decision tree and logistic regression) and opaque (e.g., random forest) models, were trained and evaluated on a data set including 215 tested ears (99 with hearing loss of mild degree or higher and 116 with no hearing loss). Post hoc explainability techniques were applied to highlight the role of each feature in predicting hearing loss.

RESULTS

Random forest (accuracy = .85, sensitivity = .86, specificity = .85, precision = .84) performed, on average, better than decision tree (accuracy = .82, sensitivity = .84, specificity = .80, precision = .79). Support vector machine, logistic regression, and gradient boosting had similar performance as random forest. According to post hoc explainability analysis on models generated using random forest, the features with the highest relevance in predicting hearing loss were age, number and percentage of correct responses, and average reaction time, whereas the total test time had the lowest relevance.

CONCLUSIONS

This study demonstrates that a multivariate approach can help detect hearing loss with satisfactory performance. Further research on a bigger sample and using more complex ML algorithms and explainability techniques is needed to fully investigate the role of input features (including additional features such as risk factors and individual responses to low-/high-frequency stimuli) in predicting hearing loss.

Effects of word familiarity and receptive vocabulary size on speech-in-noise recognition among young adults with normal hearing

Having a large receptive vocabulary benefits speech-in-noise recognition for young children, though this is not always the case for older children or adults. These observations could indicate that effects of receptive vocabulary size on speech-in-noise recognition differ depending on familiarity of the target words, with effects observed only for more recently acquired and less frequent words. Two experiments were conducted to evaluate effects of vocabulary size on open-set speech-in-noise recognition for adults with normal hearing. Targets were words acquired at 4, 9, 12 and 15 years of age, and they were presented at signal-to-noise ratios (SNRs) of -5 and -7 dB. Percent correct scores tended to fall with increasing age of acquisition (AoA), with the caveat that performance at -7 dB SNR was better for words acquired at 9 years of age than earlier- or later-acquired words. Similar results were obtained whether the AoA of the target words was blocked or mixed across trials. Differences in word duration appear to account for nonmonotonic effects of AoA. For all conditions, a positive correlation was observed between recognition and vocabulary size irrespective of target word AoA, indicating that effects of vocabulary size are not limited to recently acquired words. This dataset does not support differential assessment of AoA, lexical frequency, and other stimulus features known to affect lexical access.

A Comparison between a remote testing and a laboratory test setting for evaluating emotional responses to non-speech sounds

OBJECTIVE

To evaluate remote testing as a tool for measuring emotional responses to non-speech sounds.

DESIGN

Participants self-reported their hearing status and rated valence and arousal in response to non-speech sounds on an Internet crowdsourcing platform. These ratings were compared to data obtained in a laboratory setting with participants who had confirmed normal or impaired hearing.

STUDY SAMPLE

Adults with normal and impaired hearing.

RESULTS

In both settings, participants with hearing loss rated pleasant sounds as less pleasant than did their peers with normal hearing. The difference in valence ratings between groups was generally smaller when measured in the remote setting than in the laboratory setting. This difference was the result of participants with normal hearing rating sounds as less extreme (less pleasant, less unpleasant) in the remote setting than did their peers in the laboratory setting, whereas no such difference was noted for participants with hearing loss. Ratings of arousal were similar from participants with normal and impaired hearing; the similarity persisted in both settings.

CONCLUSIONS

In both test settings, participants with hearing loss rated pleasant sounds as less pleasant than did their normal hearing counterparts. Future work is warranted to explain the ratings of participants with normal hearing.

User-Operated Audiometry Project (UAud) - Introducing an Automated User-Operated System for Audiometric Testing Into Everyday Clinic Practice

Hearing loss is the third leading cause of years lived with disability. It is estimated that 430 million people worldwide are affected, and the number of cases is expected to increase in the future. There is therefore increased pressure on hearing health systems around the world to improve efficiency and reduce costs to ensure increased access to quality hearing health care. Here, we describe the User-Operated Audiometry project, the goal of which is to introduce an automated system for user-operated audiometric testing into everyday clinic practice as a means to relieve part of this pressure. The alternative to the existing referral route is presented in which examination is executed via the user-operated system. This route is conceptualized as an interaction between the patient, the system, and the hearing care professional (HCP). Technological requirements of the system and challenges that are related to the interaction between patients, the user-operated system, and the HCPs within the specific medical setting are discussed. Lastly, a strategy for the development and implementation of user-operated audiometry is presented, which includes initial investigations, a validation study, and implementation in a real-life clinical situation.

Evaluation of a Novel Speech-in-Noise Test for Hearing Screening: Classification Performance and Transducers Characteristics

One of the current gaps in teleaudiology is the lack of methods for adult hearing screening viable for use in individuals of unknown language and in varying environments. We have developed a novel automated speech-in-noise test that uses stimuli viable for use in non-native listeners. The test reliability has been demonstrated in laboratory settings and in uncontrolled environmental noise settings in previous studies. The aim of this study was: (i) to evaluate the ability of the test to identify hearing loss using multivariate logistic regression classifiers in a population of 148 unscreened adults and (ii) to evaluate the ear-level sound pressure levels generated by different earphones and headphones as a function of the test volume. The multivariate classifiers had sensitivity equal to 0.79 and specificity equal to 0.79 using both the full set of features extracted from the test as well as a subset of three features (speech recognition threshold, age, and number of correct responses). The analysis of the ear-level sound pressure levels showed substantial variability across transducer types and models, with earphones levels being up to 22 dB lower than those of headphones. Overall, these results suggest that the proposed approach might be viable for hearing screening in varying environments if an option to self-adjust the test volume is included and if headphones are used. Future research is needed to assess the viability of the test for screening at a distance, for example by addressing the influence of user interface, device, and settings, on a large sample of subjects with varying hearing loss.

Reward Enhances Online Participants' Engagement With a Demanding Auditory Task

Online recruitment platforms are increasingly used for experimental research. Crowdsourcing is associated with numerous benefits but also notable constraints, including lack of control over participants' environment and engagement. In the context of auditory experiments, these limitations may be particularly detrimental to threshold-based tasks that require effortful listening. Here, we ask whether incorporating a performance-based monetary bonus improves speech reception performance of online participants. In two experiments, participants performed an adaptive matrix-type speech-in-noise task (where listeners select two key words out of closed sets). In Experiment 1, our results revealed worse performance in online (N = 49) compared with in-lab (N = 81) groups. Specifically, relative to the in-lab cohort, significantly fewer participants in the online group achieved very low thresholds. In Experiment 2 (N = 200), we show that a monetary reward improved listeners' thresholds to levels similar to those observed in the lab setting. Overall, the results suggest that providing a small performance-based bonus increases participants' task engagement, facilitating a more accurate estimation of auditory ability under challenging listening conditions.

Introduction for the 4th International Meeting on Internet and Audiology Special Issue of the American Journal of Audiology

[Figure: see text].