I love the stories emerging regularly in radiology where technology and creativity are ushering in new ways to improve the patient care offered by radiologists. Such is the case at the University of Miami Miller School of Medicine where the Department of Radiology is working with the University’s Institute for Data Science and Computing (IDSC) to design an artificial intelligence (AI) tool that could help them as they diagnose patients, and which could also play a role in providing personalized, precision-based patient care.
Traditionally, radiologists use their specialized training and experience to review radiological images, the patient’s electronic medical records and other data sources to reach a diagnosis. At the University of Miami, radiology faculty are working with the IDSC to develop an artificial intelligence toolbox to help them diagnose their patients based not only on imaging data but also by considering a patient’s unique background and circumstances. The Miami radiologists hope the toolbox will consider risk factors such as race and ethnicity, socioeconomic and educational status, and exposure in order to aid their radiologists' evaluations. They believe artificial intelligence in radiology is currently only able to make a binary decision such as positive or negative for one disease, rather than scanning for a host of disorders. They hope to change this.
According to Radiology Department Chair Alex McKinney, M.D., artificial intelligence should be contextual in nature, which will take in all of a patient’s risk factors, lab data, and past medical data. “It will become a form of augmented interpretation to help us take care of the patient," he said. The team in Miami notes that patient data that is used in a radiological diagnosis is typically not inclusive of a range of demographic groups, which can lead to a bias in care. Their AI toolkit is designed to alter that paradigm.
The radiologists plan to focus first on illnesses that have a high mortality or prevalence in the local population, like breast cancer, lung cancer, and prostate cancer, and to add other patient conditions over time. In an effort to remove bias, the team plans to add more images and data of all population groups in the community, as it is available.
This is an ambitious project. If proven successful, this project could lead to a new era in diagnostic radiology where radiologists can add to their understanding not just the radiological images but also the patient’s genetics, age, and various risk factors. We live in exciting times!