A Southeastern professor and two student researchers teamed up to develop beyond gold standard cochlear implant sound technology in only eight weeks, earning a national award.
Southeastern Professor of Physics Sanichiro Yoshida and graduate students Anthony Calmes and Conor McGibboney, both of Hammond, are making waves in research that could one day help others. Their efforts were recently recognized on a national level, earning an honorable mention in the first Cochlear Implant (CI) Hackathon. The team, with no prior experience in cochlear or sound technology, achieved better than the “gold standard,” a reference implementation of a fully featured cochlear implant sound coding strategy.
The CI Hackathon used a crowdsourced ranking structure to judge entries against each other and against a baseline, which essentially gives an approximation of what a cochlear implant sounds like today.
“Southeastern’s entry performed better than the baseline in the category of simple words. Essentially this means that their improved algorithm outperformed what the cochlear implant can do today in that sound category,” explained CI Hackathon Project Leader Leah Muller. “This is a huge feat in itself. We as organizers were both surprised and excited when we started seeing teams perform better than baseline, because that meant we were succeeding in the second part of our mission with the Hackathon: to generate promising new strategies that may improve hearing for real cochlear implant users.”
Cochlear implants are electronic devices that restore hearing to people who are born without hearing or who lose their hearing over time. The implants allow some deaf people to hear sounds by overriding the innate hearing mechanisms of the ear. Although cochlear implants can restore or improve hearing, some users may experience difficulty, especially with hearing speech in a noisy environment and enjoying music.
Held virtually, the CI Hackathon encouraged national participation with contestants from five continents. Top tier entries represented academic institutions, commercial entities, and individuals. The Southeastern team placed in the top four against sixty registered teams from universities all over the world, as well as private biotech firms and other tech companies.
The competition was a joint effort between Advanced Bionics, University of California San Francisco, and University of Minnesota with a goal to inspire members of the public to improve CI sound processing. The hackathon ranked entries based on their performance in four sound categories—a series of three unrelated one-syllable words, natural speech, speech in a noisy background, and music across a wide range of styles. Advanced Bionics, one of three FDA-approved CI manufacturers, sponsored the contest and provided the gold standard reference.
Advanced Bionics and the University of Minnesota further provided a software framework for contestants to develop and test their own algorithms, including an acoustic simulation of the sound percept produced by a CI in
order for normal-hearing listeners to be able to optimize and ultimately judge the sound quality of the CI sound
Yoshida, whose expertise is in optical interferometry and field theory, heard about the competition and was intrigued by it. The opportunities it presented to bring together many different discipline backgrounds and fields for a common purpose was particularly enticing. Yoshida connected with Calmes and McGibboney, who were current graduate students in his department, and formed a team to dig deeper into this project. Dean of the College of Science and Technology Dan McCarthy encouraged him to enter the competition.
McGibboney said that what inspired him to compete at the CI Hackathon is general curiosity about the human brain. “Humankind has progressed so far, yet a lot of the mystery about what life is and how humans will continue to evolve is locked inside the human brain,” he said. “It was an excellent opportunity to take some of the knowledge we learned in physics and other areas and see what we could uncover or how we could apply it to the human brain.”
Yoshida, Calmes, and McGibboney worked on the project over winter break. Yoshida explained that the software his team developed for the competition included basic strategies, such as increasing the gain of part of the control unit. Next, Calmes and McGibboney tested the software and reported to Yoshida with the results. He then modified the strategy and sent it back to the students.
“The goal of the software development was to improve the performance of the sensor, the part that receives the sound and converts it to electronic signals; of the controller and conditioner, the part that controls various parameters to condition the electric signals; and the actuator, the part that generates the output sound from the electric signal processed by the controller and conditioner unit,” Yoshida explained.
“The sensing/actuation algorithm is similar to that used by LIGO (Laser Interferometer for Gravitational-wave Observatory),” he continued. “I participated in the LIGO project for about 10 years. This experience helped me develop the software for the Cochlear Hackathon, since our software development was physics based.”
While Yoshida and his team were able to draw upon extensive knowledge from such other areas, they actually saw the lack of experience with CI as an advantage in that it helped them form a unique approach to the competition.
“The scientific content of the competition was similar to what I covered in a graduate physics course in the semester prior to the competition,” Yoshida said. “What set our entry apart from the others was that our software design is based on physics, whereas our competitors developed their versions based on computer science or medical science.”
While an advantage may have been achieved in this way, the team’s accomplishment without prior experience in CI and sound technology is highly significant.
“Southeastern was one of the few teams that entered the hackathon with no background in acoustics, hearing, or cochlear implants at all,” explained Muller. “Over the course of eight weeks from Hackathon opening to final
entry, they not only learned the necessary principles of cochlear implant technology and sound processing, but they also used that knowledge to come up with a top-ranked processing algorithm.”
“The fact that Dr. Yoshida and his team beat out major research universities when he and his students had never worked on cochlear implants or anything like them before is simply astounding,” McCarthy said.
Along with the feat achieved in CI research from Yoshida and his students working together, participating in this project also had a profound direct impact on and served as a great hands-on learning opportunity for the younger researchers.
“The programming was definitely the most interesting thing I learned while undertaking this project with the CI Hackathon,” Calmes said. “I had not really done much with the computer programming aspect of it before, but it definitely was the most intense area for me.”
For McGibboney, realizing the direct impact that science can have on people’s lives created the biggest spark in him. “Sometimes, especially during a pandemic, you can develop distances between people, and I am aware that being deaf can cause a disconnect,” he said. “However, the idea that we could share in the first experience of listening to music again—and when I first learned how that impacted people that were wearing the cochlear implants—that was the most amazing thing to me. That’s a real connection from science in the classroom to making it into the marketplace to improve people’s lives.”
By Tonya Lowentritt