Researchers developed an AI-based model for accurate diagnosis of ear infections.
The model uses deep learning techniques and was trained using a dataset of more than 600 images of tympanic membranes from children aged 18 years or younger who were taking treatment at Mass Eye and Ear due to recurrent ear infections or fluid in the ears. The AI-based model proved to be 80 percent more accurate than the accuracy of clinicians reported in the medical literature. This development in diagnosing tools can assist doctors with better clinical decision-making. The model named OtoDX was more than 95 percent accurate in diagnosing an ear infection in a set of 22 test images compared to 65 percent accuracy among a group of clinicians consisting of ENTs, pediatricians, and primary care doctors, who reviewed the same images.
“Ear infections are incredibly common in children yet frequently misdiagnosed, leading to delays in care or unnecessary antibiotic prescriptions,” said Matthew Crowson, MD, lead author of the study and an otolaryngologist and artificial intelligence researcher at Mass Eye and Ear in a press release. “This model won’t replace the judgment of clinicians but can serve to supplement their expertise and help them be more confident in their treatment decisions.”
Ear infections are most common in children, it causes due to buildup of bacteria inside the middle ear. However, the previous studies reveal the inefficiency of conventional methods for diagnosing ear infections in children from a physical exam with an accuracy below 70 percent. This may result in the delay of left untreated ear infections that can lead to hearing loss, development delays, complications like meningitis, etc. Contrary to this if a child is overtreated who doesn’t have any ear infections may lead to antibiotic resistance and will result in the ineffectiveness of medications for future infections. The second problem is of most significant public health importance.
“Since clinicians would rather stay on the side of caution, it’s pretty easy to see why parents typically walk out of urgent care with a prescription for antibiotics,” he said.
The researchers are working to enhance their work and validate AI-based diagnostic tools by using more than 1,000 intraoperative images of the tympanic membrane collected by the Mass Eye and Ear team. They envision providing clinicians and doctors a viable tool that can assist them in better diagnosis of ear infections in kids.
Click here for the Published Research Paper.
