Can Immunotherapy Cure MS?

An AI Clinical Tool Aims to Decode Hearing Loss and Tinnitus

A new clinical trial in the UK plans to evaluate whether artificial intelligence can accurately triage patients with hearing loss and tinnitus. The study, led by the Royal Cornwall Hospitals NHS Trust, will recruit 1,500 participants to determine if an AI system’s clinical recommendations match those of specialist doctors. This research responds to a pressing national health issue: over one million GP appointments in the UK are for tinnitus each year, and patients often wait more than 12 months for a hospital specialist assessment.

The Virtual Clinic Foundation

To manage overwhelming demand, the Royal Cornwall Hospitals NHS Trust previously developed a virtual ear, nose, and throat clinic. In this model, patients take a validated in-person hearing test and complete online questionnaires about their symptoms. An otolaryngologist then reviews the collected data to decide on management, which could be discharge with advice, a request for imaging, or scheduling a face-to-face consultation.

Initial results were positive. The virtual approach allowed the majority of patients to be managed without an in-person visit, dramatically cutting waiting times. However, a major bottleneck remained: a specialist doctor still had to personally review every single case. This constant demand on clinician time limits the system’s overall capacity and takes time away from patients who need more complex, direct care.

How the AI Trial Will Work

This new observational study adds an AI layer to the existing virtual clinic process. Researchers will use explainable AI methods to train a system on the clinical data from patients. The goal is for the AI to learn to make triage recommendations—such as “discharge,” “refer for scan,” or “see in clinic”—that match the standard set by the clinicians.

A core feature is explainability. The AI will be designed to provide clear, transparent reasoning for each recommendation it generates, rather than acting as a “black box.” During the trial, the AI will analyze each patient’s data and produce an outcome. A clinician will also independently assess the same case to determine the appropriate management path. The clinician will be blinded to the AI’s suggestion. Finally, researchers will compare the two sets of recommendations to measure their level of agreement.

The study is designed as a pilot comparative study. All patient care will continue to follow the clinician’s final decision, ensuring safety is not compromised during the research phase.

Who Can Participate in the Study

The trial plans to enroll 1,500 participants. To be eligible, individuals must be patients assessed through the Royal Cornwall Hospitals’ virtual hearing loss and tinnitus clinic, have symptoms of hearing loss or tinnitus, and be able to provide informed consent.

The exclusion criteria are specific: individuals under 18 years old, those unable to consent, patients without sufficient English proficiency where translation cannot be arranged, and cases where the submitted data quality is too poor for analysis. These criteria ensure the data used to train and test the AI is reliable and ethically sourced.

The Burden of Hearing Loss and Tinnitus

The trial addresses a significant public health challenge. Hearing loss affects an estimated 11 million people in the UK, and tinnitus impacts around 7 million. Both conditions are more than minor inconveniences; they are strongly associated with reduced quality of life, poorer mental health outcomes, and challenges with employment.

The COVID-19 pandemic exacerbated existing backlogs in specialist services. The long waits for assessment can leave patients in distress, uncertain about their condition and management options. An effective AI triage tool could help clear this backlog by ensuring patients are routed to the right care pathway more efficiently.

Current Status and Future Implications

As of now, this trial is listed as “NOT_YET_RECRUITING” on ClinicalTrials.gov. The research team is likely in the final stages of preparing to enroll the first of the 1,500 participants.

If the AI tool demonstrates high concordance with clinician judgments, its implementation could change patient pathways. It would not replace doctors but act as a clinical recommendation assistant. This could streamline the initial triage of routine cases, allowing otolaryngologists to focus their expertise on patients with more complex or unusual symptoms. The potential result is a sustainable system where access to initial assessment is faster for everyone, and specialist time is used more effectively.

For patients, this could mean shorter waits for answers and management plans. For the field of audiology and otolaryngology, it represents a practical application of AI to solve a well-defined capacity problem, with a model that could be adapted for other high-volume clinical services.

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)