Advanced Melanoma Therapy Improves Survival

A New AI Tool Could Streamline Care for Millions With Hearing Loss and Tinnitus

With hearing loss impacting about 11 million people in the UK and tinnitus affecting around 7 million, the demand on specialist ear, nose, and throat (ENT) services is immense. Tinnitus accounts for over one million GP appointments annually, and hospital referrals can involve waits exceeding a year. A new clinical trial, based at the Royal Cornwall Hospitals NHS Trust, is testing an artificial intelligence (AI) system designed to assist in triaging these patients. The goal is to determine if AI can match clinician assessments, potentially freeing specialists to concentrate on complex cases and reducing dangerous delays for patients.

Building on a Virtual Clinic Success

The trial builds directly on work already done by the Royal Cornwall Hospitals NHS Trust. To tackle long wait times, they developed a virtual ENT clinic. Patients with hearing loss or non-pulsatile tinnitus attend an in-person, validated hearing test and complete online questionnaires. A clinician then reviews this data to decide the next step, which could be discharge with advice, a referral for scans, or booking a face-to-face consultation. Initial results showed this virtual approach could manage most patients effectively, significantly cutting wait times.

However, a bottleneck remains. A clinician must still personally review every single case. This consumes time that could be spent on direct patient care for those with more complicated needs. The new study, officially titled “Artificial Intelligence for the Automated Diagnosis, Triage, and Assessment of Patients With Hearing Loss and Tinnitus,” aims to introduce an AI assistant into this workflow.

How the AI Clinical Trial Will Work

This is an observational pilot study designed to compare AI recommendations to the current gold standard: clinician judgment. The research team will use explainable AI methods to train a system on data from the virtual clinic. The AI will learn to suggest management pathways—like discharge, imaging, or further consultation—and provide clear reasoning for its suggestions, similar to a clinician’s notes.

During the trial, data from 1,500 participants will be processed by the AI tool. Crucially, the managing clinicians will remain unaware of the AI’s output. They will make their care decisions independently, as usual. Afterwards, researchers will compare the AI’s recommendations to the clinicians’ actual decisions. The primary measure is the level of concordance between the two. If the AI’s suggestions closely align with specialist judgment, it demonstrates potential for real-world use.

Who Can Participate in the Study?

The trial will enroll patients who are already being assessed through the Royal Cornwall Hospitals’ virtual hearing loss and tinnitus clinic. To be eligible, participants must be 18 or older, have symptoms of hearing loss or tinnitus, and be able to provide informed consent. The study excludes individuals who cannot consent, those under 18, patients without sufficient English proficiency where translation is unavailable, and cases where the collected data is of insufficient quality for analysis.

It is important to note this is not a trial of a new drug or device. No patient’s care will be determined by the AI during this study. All participants will continue to receive a full clinical review. The AI is being tested silently in the background.

The Potential Impact on Patients and Hearing Healthcare

Long waiting times for ENT and audiology services are a major problem across health systems, linked to increased patient anxiety and potentially worse outcomes. A validated AI clinical recommendation assistant could change this dynamic for hearing-related conditions.

If the AI tool proves effective, it could be integrated into the virtual clinic pathway. It would act as a first-pass filter, handling routine cases with high accuracy and flagging complex ones for immediate specialist attention. This would accelerate access to care for all patients. It could allow clinicians to focus their expertise on diagnostically challenging cases, referrals for urgent imaging, or patients requiring detailed counselling. For the average patient with straightforward hearing loss or tinnitus, it could mean a faster, more efficient journey to a resolution.

The trial status is currently NOT_YET_RECRUITING, with a target enrollment of 1,500 participants. The results will be closely watched by healthcare providers struggling with backlogs in audiology and ENT services. Success here could provide a model for managing other high-volume, data-rich specialty referrals.

This article is for informational purposes only. Consult a qualified professional for personalised advice.

Source:
AI-assisted Diagnosis, Triage and Assessment of Hearing Loss and Tinnitus (ClinicalTrials.gov: NCT07586098)