Western University uses AI to predict ear growth and develops 3D printed earmolds

Applications of AI


For families of children with hearing loss, it’s a relentless cycle of getting new ear impressions, months of doctor visits, and then coming back because the child’s ears have grown back. The ALLEars project, a major collaboration between Western University and Boystown National Research Hospital in Nebraska, is setting out to break this cycle once and for all by using AI to predict ear growth and 3D printing to manufacture ear impressions before they are needed.

The project is supported by a four-year, US$4.4 million grant from the Oberkotter Foundation, which supports initiatives focused on language development and literacy in children with hearing loss.

“By leveraging innovative and rapidly evolving technology to address long-standing challenges in providing quality care, this project will accelerate solutions and make a tangible difference in the lives of children with hearing loss,” said Teresa Carraway, CEO of the Oberkotter Foundation.

A problem that no one could solve

The World Health Organization estimates that 34 million children worldwide are deaf or hard of hearing. For the majority of hearing aid users, soft custom earmolds are not an option, they are what makes the device work. However, early childhood, when auditory intervention is most important, is also the time when physical growth is fastest.

“Children go through a period of very rapid growth during the first few years of life,” says Susan Scoley, an audiologist and professor at Western University’s School of Health Sciences and a leader on the ALLEars project. “That growth can lead to repeated interruptions in hearing aid use during critical periods of language development.”

Results will be visible quickly. One parent, Emily, described what the actual replacement cycle is like for her 8-year-old son, who has been wearing hearing aids since he was 6 months old: “Hearing aids are a lot of work. Sometimes we have to replace the earmold due to wear and tear, but after taking an impression, it can take 14 to 21 days to put the earmold back in. Two weeks is a long time for my son to get his hearing back to where it needs to be.”

Susan Scoley, audiologist, professor in the School of Health Sciences, and principal investigator of the ALLEars project. Photo courtesy of Western Health Sciences
Susan Scoley, audiologist, professor in the School of Health Sciences, and principal investigator of the ALLEars project. Photo courtesy of Western Health Sciences

Forecast, print and prepare

The ALLEars project completely reverses the existing model. Rather than reacting to growth as it happens, the system scans a digital impression of a child’s ear, uses AI to predict how it will change, and 3D prints future ear impressions in advance.

The AI ​​engine at the heart of the project is being developed by Soodeh Nikan, professor of engineering and AI lead at ALLEars. “AI learns ear characteristics by examining large earprint datasets and can translate this to predict future ear shape,” Nikan explained. “This is the first project of its kind to use AI technology for predictive earmold modeling.”

The research team is also developing mirroring technology that uses AI to infer the shape of one ear from the other. This reduces the number of impressions the infant has to undergo. “AI can help reduce repeat ear impressions, so if a child takes an impression on the left ear, there is no need to repeat it on the right ear.”

At Boys Town, a parallel research stream led by Vice President of Research Ryan McCreary is applying machine learning to acoustic prediction to determine how sound changes within the growing ear canal, ensuring children receive the appropriate amplification from their devices as they grow.

From digital files to physical earmolds

Once the AI ​​generates the predicted earmold shape, the file goes to Joshua Pearce’s engineering lab at Western University. There, postdoctoral researcher Alessia Romani is developing the 3D printing workflow needed to transform digital models into physical objects that are accurate, comfortable, and reproducible at scale.

The manufacturing challenges are considerable. “The earmolds we’re making are very small, so we’re trying to develop new ways of software, firmware and hardware to make very small but resilient earmolds that kids can use,” Pearce said.

Importantly, the team designed the entire workflow to be open source and freely available to audiologists and healthcare providers around the world. A particular focus is on low- and middle-income countries where access to earmold manufacturers is limited or non-existent.

For Scoley, that impact extends far beyond a single university or clinic. “If we can reduce appointments, expand global access to earmold manufacturing, and solve everyday clinical challenges for audiologists, we can make a world of difference. This project is a once-in-a-lifetime opportunity.”

AI and 3D printing come together in healthcare

ALLEars reflect the broader strategic opportunities that are taking shape across the healthcare sector. The combination of the predictive power of AI and the on-demand manufacturing capabilities of 3D printing is moving healthcare from a reactive to a preventive model.

Several initiatives are already showing where this is heading. Researchers at Washington State University have developed an AI-guided 3D printing process that can create detailed organ replicas optimized for each patient’s anatomy. This allows surgeons to receive a patient’s MRI in the morning, print a model in 30 minutes, and spend the remaining time preparing for surgery.

Elsewhere, Axial3D raised $18.2 million to expand its AI platform that converts patient CT and MRI scans into detailed 3D printable files to enable workflows for personalized surgical plans, custom implants, and patient-specific devices, and the company’s CEO says the technology has the power to impact millions of patients around the world. Together, these efforts aim for the same goal that ALLEars is pursuing: a healthcare model that knows exactly what patients need.

The 3D printing industry is seeking speakers for the 2026 Additive Manufacturing Applications (AMA) series covering energy, healthcare, automotive and mobility, aerospace, space and defense, and software. Each online event focuses on real-world operational implementation, certification, and supply chain integration. Practitioners interested in contributing can Please fill out the speaker recruitment form here.

Don’t forget to subscribe to get the latest 3D printing news. 3D Printing Industry Newsletter or follow us linkedin.

Explore fully The future of 3D printing and executive survey 3D Printing Industry Series. Featuring perspectives from CEOs, engineers, and industry leaders. Industrialization of additive manufacturing, 3D Printing Industry Trends in 2026qualifications, supply chain, Analysis of additive manufacturing industry.

Featured image show Susan Scoley, audiologist, professor in the School of Health Sciences, and principal investigator of the ALLEars project. Photo courtesy of Western Health Sciences



Source link