When Hend Alqaderiwas studying how saliva could predict the risk of diabetes or the severity of a coronavirus infection, she collected a lot of saliva samples-thousands, measuring hundreds of bacteria samples for each patient.
She and her colleagues couldn't analyze every single bacterium in each saliva sample, so they had to choose which ones to use. It was difficult to decide, knowing they might not end up with a representative sample.
A few years ago, Alqaderi learned that some of her colleagues were using machine learning, a type of artificial intelligence, to analyze large amounts of data. "I became interested," says Alqaderi, who is an assistant professor in the Department of Public Health and Community Service at Tufts University School of Dental Medicine.
She attended a conference and saw that machine learning could analyze thousands of samples and make disease predictions in minutes. Alqaderi was hooked. She took two courses on AI at MIT three years ago and continues to analyze vast datasets. Now, Alqaderi is director of the Dental AI Lab, a joint appointment with the Tufts Institute for Artificial Intelligence (TIAI). In that role, she's teaching dental students to use artificial intelligence at the Dental AI Lab and through a new Artificial Intelligence in Dentistry class.
While research first drew Alqaderi to AI, she also appreciates the technology's ability to augment patient care at dental practices. AI can help with scheduling, billing, insurance, patient charts, and documentation-the time-consuming work that supports dental care.
Dentists "spend a lot of time documenting every single procedure and treatment planning," Alqaderi says. "If we can use AI to free some time from dentists, they can focus more on the most important part, which is patient treatment and patient communication."
Predictive power of AI
AI can also be a powerful tool in preventive oral medicine.
We know some risk factors, such as socioeconomic status, diet, or smoking, but we still don't have the ability to build software that can predict the disease before it is happening."
Hend Alqaderi, Assistant Professor, Department of Public Health and Community Service, School of Dental Medicine, Tufts University
That's where AI comes in, analyzing electronic dental records and creating an algorithm to predict which patients are most likely to get cavities, oral cancer, or other conditions.
Alqaderi likens the system to the recommendations for users created by Netflix, Google, or Amazon, where an algorithm derived from their views, searches, and purchases prompts recommendations for things they may like.
The vision of a streamlined dental office with AI offering pre-diagnostic care is seeding the projects underway at Tufts Dental AI Lab. The lab, soft-launched in July with the mission of research and education, is part of TIAI and the Department of Public Health and Community Service. It is funded by Beyond Limits, a California-based tech company.
For now, the young lab is focused on research and publishing academic papers. "In the future, we want to translate the research into practice. We want to see all these projects coming to life," Alqaderi says. If lab members build an AI model to predict a disease, they eventually want to build software that will both determine which patients will develop particular conditions, and integrate into an office's electronic health system.
One current project is developing AI models that can read X-rays and pick up on anything suspicious, "and the dentist can cross-check or verify after AI, this will save a lot of time" Alqaderi says.
AI in the curriculum
Each project includes not only dental students or faculty, but also data or computer scientists from Tufts Institute for Artificial Intelligence. "We are the domain expertise, and they are the data scientists," Alqaderi says. And while those from TUSDM don't build the AI models, they need to understand the basics of AI analysis to work and communicate with their more technical counterparts.
To that end, Alqaderi led Tufts' first Artificial Intelligence in Dentistry course during the fall semester, where 240 third-year dental students attended 10 lectures covering major concepts in AI, ethics, dental treatment planning, and how AI can increase dental care access for people living in remote areas.
Students evaluate existing dental AI applications according to criteria such as ethics and data quality. They also work in teams on a capstone project, in partnership with the same scientists available to the Dental AI Lab.
AI is coming to dentistry, Alqaderi says, for patient care, research, and academics. "We want our students to be ready to deal with AI in their clinics, to consider ethical considerations, and to be ready to critique and use it."
An AI organization for dental students
The role of AI in dentistry is widely misunderstood compared to other fields like engineering, systemic medicine, and business, says Yash Brahmbhatt, D27, founder of the Artificial Intelligence in Dental Research & Education Society.
Brahmbhatt, the first Tufts Dental student intern at the Tufts Institute of Artificial Intelligence, launched the club with the intent of encouraging dental students and faculty to learn more about AI and develop more avenues for AI-based dental research. "During the launch of the society, I was inspired by the lack of organizations that focused on dentistry, research, and AI combined," he says.
"I wish more people understood that AI is not about replacing dentists, but about augmenting our skills to improve diagnosis, treatment planning, and patient outcomes," Brahmbhatt says.
"I am a firm believer that AI has the power to help our patients, and AI is already shaping health care. I also believe that properly trained AI models have the power to help patient outcomes immensely and can often catch diagnoses that humans might miss due to external factors like fatigue, stress, or accidental oversight," he adds.
MRI scans reveal human brain growth from pregnancy through birth