Transformative Ear Sensing for Fast, Low-Cost Diagnostics

The Challenge

Early identification of hearing impairment is critical for timely treatment and (re)habilitation to avoid missing the window of opportunity for normal speech, language, and cognitive development in children. While Australia leads the world in universal newborn hearing screening (UNHS) for this purpose, it is costly (~$22.6M/year), but cheaper options are less accurate. This project involves the development of a patented screening test (US Patent 11839467) that breaks the cost-accuracy trade-off, with potential savings of $7-11 million per year while maintaining high diagnostic performance. This will be a world-first transformational innovation that can also expand hearing screening access in remote First Nations communities and low-/middle-income countries. NSW is the ideal place to translate this innovation, with strong clinical networks, implementation capability, and industry partnerships, helping sustain Australia’s leadership in hearing screening while delivering an Australian-grown solution for global benefit.

The Solution

Hearing assessment spans the middle ear, inner ear, and auditory nerve/brain, but unlike the eye, these structures cannot be inspected directly, so clinicians use remote-sensing tests. Today this requires three devices and sequential exams: tympanometry (middle-ear), otoacoustic emissions, OAE (inner-ear), and auditory brainstem response, ABR (nerve/brain), adding cost, time, and loss-to-follow-up. This project’s innovative test completes all three hearing checks in 6-8 minutes, faster than the standard multi-device pathway, but still slower than Australia’s newborn screening benchmark. It will reduce recording and analysis to less than one minute using fast signal processing (e.g., deconvolution) and machine learning (e.g., random forests, adaptive filtering). It will deliver a production-ready “sense-to-decision” pipeline which converts raw ear-sensor recordings into three clinical outputs (middle-ear, inner-ear, neural), packaged for deployment on our industry partner’s existing hardware.

Industry Partners