AI bringing different data together to form new insights

A new interdisciplinary project is to combine structured and unstructured health data from a variety of sources to create large datasets that could potentially generate new findings in health care.

The project, a collaboration between University of Illinois Chicago (UIC), University of Iowa, University of Missouri and Loyola University, as well as Microsoft and Tackle AI, will involve finding ways to combine data and free-text notes from nurses, physical and occupational therapists, speech and language pathologists and physicians.

These notes often provide additional, valuable information about a patient’s progress, particularly as their care moves outside the hospital.

The project will focus on two complex patient populations: patients who have experienced injuries related to a fall and infants transitioning from the neonatal intensive care unit to home. Both groups rely on the care provided by a variety of health professionals.

“Health care is an interdisciplinary process, but existing data tools and infrastructure ignore most of the team,” said Andrew Boyd, one of the project’s principal investigators and professor of biomedical and health information sciences at UIC.

“Other professions see patients more frequently and provide very high-fidelity data that gets closer to the reality of the patient, instead of just the brief snapshots in time that you get from data documented by physicians.”

Researchers will use advanced computational methods on the new, combined datasets to create all-team care summaries and new AI applications. They also hope to use the data to make new scientific discoveries that will improve care and treatment for patients. The research will test whether large language models can be trained to help understand and connect text data across professions.

Unifying data can help improve communication

Catherine K. Craven, a principal investigator and biomedical informatician at the University of Missouri, said: “Falls and NICU patients require all-team care while in the hospital and via outpatient clinics. But fragmented, siloed documentation impedes communication. By unifying this data, we can improve communication between health care providers, the patient and their care partners and generate novel scientific insights that improve patient outcomes.”

The value of multidisciplinary data is particularly clear when it comes to managing adult fall injuries. Falls are difficult to prevent and can lead to poor health outcomes in older adults. The top predictor of fall risk is the number of previous falls, but patients may not tell their physicians about all their falls. Reports on falls from emergency-room visits or outpatient therapy sessions may be overlooked in the flood of information in a patient’s health record.

Physical and occupational therapists also collect detailed information relevant to fall risk, such as strength and balance assessments. Because these reports are often subjective and text-based, they are hard to combine with physician notes or numerical data such as test results.

“Data is gold, but until it can be used, it is meaningless,” said Tanvi Bhatt, professor of physical therapy and rehabilitation sciences at UIC and co-investigator on the project. “The text-based notes that we have are more narrative and descriptive, compared to lab measures. But if that text is lost, there is no continuum of care.”

Integrating the text-based notes with other sources could help clinicians identify the cause of a patient’s falls and thereby offer the most appropriate interventions to prevent future injuries. It could also help researchers design and test new prediction models of fall risk and share those insights with patients in clear language.

Another advantage of integrating the data is that it could help involve the patient in health care decisions, said Mary Khetani, professor of occupational therapy and rehabilitation sciences at UIC and a co-investigator on the grant.

The narrative notes taken by physical and occupational therapists often come directly from interviews with a patient and their family. Organising the data to share with patients and their caregivers can help them feel more informed and engaged as they navigate different health care services outside of the hospital.