Hamid Motallebzadeh, PhD, an instructor in Otolaryngology–Head and Neck Surgery (OHNS) at Harvard Medical School and an investigator at Mass Eye and Ear, was recently awarded a R21 research grant by the National Institutes of Health (NIH) for his research on using AI to automate the noninvasive diagnosis of middle-ear pathologies.
NIH R21 grants are awarded to scientists who are in the early, conceptual stages of project development. Dr. Motallebzadeh’s grant is worth more than $600,000 over three years, all of which will fund his research on machine-learning techniques to drive differential diagnostic automation.
In most cases, the diagnosis and confirmation of middle-ear pathologies currently rely on imaging and exploratory surgery. Both techniques can be costly, harmful and subjective in nature. Since the symptoms of many middle-ear disorders overlap, scientists have long sought an objective, fast and noninvasive way of differentiating one pathology from another. Wideband tympanometry, a safe way of measuring the acoustic response of the middle ear, has the potential to identify different pathologies, but not enough datasets of confirmed pathological cases exist to verify its diagnostic capabilities.
With the funding of his R21 grant, Dr. Motallebzadeh will try to create synthetic datasets that could be used to train a machine-learning algorithm to infer middle-ear status. According to Dr. Motallebzadeh, objectively and accurately identifying middle-ear pathologies could reduce the need for exploratory surgery. It could also improve the specificity of preoperative preparations and provide a low-cost means of postoperative monitoring.