Hearing loss can be considered the most prevalent health impairment worldwide, with adequate diagnosis and treatment lacking for most. Computational audiology, the augmentation of traditional hearing health care by digital methods, has potential to dramatically advance audiological precision and efficiency to address the global burden of hearing loss.
Modern machine learning and data collection techniques combined with mobile technology will transform hearing health care. Quantifiable aspects of audiology make it well suited for computational augmentation, automation and clinical decision support. These advances can improve access to care, personalize diagnoses and treatments, and lead to better quality services. Here, we present recent examples that illustrate the potential of computational audiology and raise awareness for risks associated with big-data processing and emerging medical artificial intelligence (AI) applications.
Computational audiology can be applied along the entire spectrum of hearing health care: from fundamental auditory science to more clinical and translational work on diagnosis, treatment, and rehabilitation across the lifespan. Computational assistance can replace some of the tasks currently performed by experts, reduce testing burdens and, most importantly, improve access, quality, and uniformity of hearing health care on a global scale. Computational audiology may initially serve as an assistive tool in the hands of the expert but we foresee that modern mobile phones, readily available in billions of pockets across the world, will catalyze the democratization of audiology. However, a potential risk of highly autonomous and self-learning clinical decision support systems is that, depending on the design, they behave like black boxes. Such systems are not interpretable and unsuitable for an informed-decision process. Also, it hinders independent inspection, collaborative research, and awareness for biases in data. In order to accomplish a fair, affordable, and safe digital transformation of audiology, we must join forces with experts in computational sciences, agree on global standards and evidence-based procedures, and carefully consider the possible challenges and risks of big data and AI technology.