ASU’s Embedded Machine Intelligence Lab integrates AI into wearable technology to personalize and assist users with health and safety monitoring.
The lab aims to develop complex and compact technologies that people can carry and wear anywhere to improve portable wellness systems. Hassan Qassemzadeh, director of EMIL and professor in the School of Health Solutions, said the lab designs, develops and validates technologies that have the potential to be successful and widely adopted.
“The focus was on building systems that use AI-powered sensors in the health field to improve health, detect health events, provide automated interventions, and extract biomarkers related to various health outcomes,” said Qassemzadeh.
Qassemzadeh said he has been researching digital health since 2007 and established EMIL at the university in 2021. He added that the lab strives to be transparent and provide user-centered feedback and assistance, while also allowing those using the technology to maintain autonomy and control.
He said that to improve AI systems, EMIL focuses on personalization and consideration of individual differences. Such factors include age, demographics, and user data that changes in real time.
“(We) make sure that we do all these predictions and health assessments while taking into account the user’s situation,” Qassemzadeh said. “What will happen to them today? What will the situation be like over time? And within that context, you can make predictions.”
Reza Rahimi Azgan, a doctoral student studying computer science, said his lab focuses on algorithm development. He primarily works on “lightweight” systems that can be integrated into small devices, such as wristbands, that don’t have the resources to run large programs.
“The reason we want to run these models on these devices is because we want to deploy these models and deploy these systems and machines into mobile applications,” said Rahimi Azghan. “About the types of devices that interact directly with humans.”
He said the role of machines and algorithms is to be able to predict or identify what activities users are performing at a given time. This goal spans the collection of heart rate data, electrodermal activity levels, skin conductance, and more.
read more: AI in Healthcare: Balancing Innovation and Results
A common problem with most machine learning or AI models is that, although they are good at generalizing, they have difficulty retaining large amounts of data for long periods of time (also known as catastrophic forgetting). However, EMIL is interested in continuous learning of AI.
Rahimi Azgan said that rather than teaching models to search for new knowledge, EMIL focuses on increasing the retention power of models to build and remember users’ data on personal devices.
One of EMIL’s current projects, HeatMind, studies heat stress and heat-related illnesses encountered in the high-intensity environments that firefighters regularly experience. Ashley Nakawatase, a fourth-year student studying biomedical informatics and data science, began working on the project as part of her capstone.
Nakawatase said the lab aims to study how to monitor firefighters’ health indicators in real time through wearable sensors. This education and awareness will allow first responders to serve the public more safely and efficiently, she said.
“They are first responders, people who are trying to protect us, and they need to be protected as well,” Nakawatase said. “Building bonds with firefighters creates a new level of connection within the community and strengthens the entire community emergency response team.”
The team is also working on projects related to metabolic, mental and cardiovascular health, among others. Mr. Qasemzadeh said EMIL will continue its efforts to build a health system with a wide range of uses, usage options, and public trust.
“How do we scale the technology?” Qassemzadeh said. “We want to make sure the systems are robust, we want to make sure there’s transparency. We want to make sure there’s confidence in these systems in general.”
Edited by Kate Gore, Senna James and Pippa Huang.
Contact the reporter at dforres5@asu.edu.
Like and follow The State Press on Facebook @statepress With X.
Continue to support student journalism, donate In today’s national news agency.
