GE HealthCare
(Nasdaq: GEHC)
today announced data showing its AI models’ ability to accurately predict patient responses to immunotherapies.
The study collected clinical data to accurately forecast the effectiveness and toxicity of cancer immunotherapy. Data demonstrated accuracy between 70% and 80% with the AI models in a pan-cancer cohort. Investigators published outcomes in the Journal of Clinical Oncology Clinical Cancer Informatics (JCO CCI).
According to a news release, the authors believe the approach is the first attempt to design AI model capable of assessing the risks and benefits of immunotherapy using only electronic health record (EHR) data. Inputs already exist in medical records, like diagnosis codes and medication, providing an advantage for the AI models. The study drew just two features (smoking status and number of prior immune checkpoint inhibitor (ICI) drugs) from manually collected data.
“We focused primarily on this routinely collected structured data to build predictive models with the goal that these models would be able to be implemented in any clinical setting,” Travis Osterman, associate VP for research informatics and associate chief medical information officer for Vanderbilt University Medical Center (VUMC), and director of cancer clinical informatics at Vanderbilt-Ingram Cancer Center, said in a news release.
How the GE HealthCare AI models work
To develop the models, GE HealthCare and VUMC retrospectively analyzed and correlated immunotherapy responses for thousands of VUMC cancer patients. They used the patients’ deidentified demographic, genomic, tumor, cellular, proteomic and imaging data.
GE HealthCare trained its models to predict efficacy outcomes and the likelihood of an individual patient developing an adverse reaction. This provided information that may help clinicians select the most appropriate treatment pathway sooner. It also potentially helps to spare unnecessary side effects and cost.
The company believes the models have the potential for wide deployment and adoption, thanks to a variety of input features. GE HealthCare plans to look at commercializing the models once it secures regulatory authorization. Potential fields include clinical decision support and drug development.
“We aim to partner with pharmaceutical companies, researchers, and clinicians to optimize and ultimately apply the AI models in therapy development and in clinical practice,” said Jan Wolber, global product leader – Digital at GE HealthCare’s Pharmaceutical Diagnostics segment. “We want to use AI to personalize predictions and provide decision support for the clinician in determining appropriate therapies.”
