November 22, 2024
GE HealthCare announces time-saving AI tool for doctors who treat cancer
GE HealthCare announced a new artificial intelligence application it said will save time for doctors who diagnose and treat cancer.

GE Healthcare booth is seen ahead of the 2022 China International Fair for Trade in Services (CIFTIS) at China National Convention Center on August 28, 2022 in Beijing, China. 

Yi Haifei | China News Service | Getty Images

GE HealthCare on Monday announced a new artificial intelligence application it said will save time for doctors who diagnose and treat cancer.

CareIntellect for Oncology, as the tool is called, will help oncologists get up to speed on a patient’s history and disease progression by quickly showing them the data they need, the company said. GE said it wants to spare oncologists the headache of digging through records so they can focus on caring for their patients.   

Health-care data is notoriously difficult to analyze, and as much as 97% of the data produced by hospitals goes unused, according to a Deloitte report. That information is stored across numerous vendors and file formats such as images, lab test results, clinical notes and device readings, which can be extremely taxing for doctors to sort through. 

“It’s very time-consuming, very frustrating for these clinicians,”  Dr. Taha Kass-Hout, GE HealthCare’s global chief science and technology officer, told CNBC in an interview.

CareIntellect for Oncology will be able to summarize clinical reports and identify when patients are deviating from their treatment plans, Kass-Hout said. The system can flag when a patient misses a lab test, for instance, so that their doctor can determine the best next steps. 

“For cancer patients, the treatment journey can last years and involve numerous doctor visits,” he said.

GE HealthCare’s CareIntellect for Oncology

Courtesy of GE HealthCare

CareIntellect for Oncology can also help identify relevant clinical trials that patients might be eligible for, saving oncologists hours of work, said Chelsea Vane, vice president of digital products at GE HealthCare. That process has traditionally required doctors to scroll through a database of available trials, memorize inclusion and exclusion criteria and dig through patient records to determine a good fit, Vane told CNBC.

“What we’ve done is remove that,” she said.

The purpose of the new app is to save oncologists time and effort, but if doctors want to dive into more detail, CareIntellect for Oncology allows them to view the original record that’s referenced, the company said.

GE HealthCare is planning to make CareIntellect for Oncology widely available to U.S. customers in 2025, and it will initially be optimized for prostate and breast cancers. Health organizations such as Tampa General Hospital are already evaluating it, the company said. Since the tool is cloud-based, it will drive recurring revenue for GE HealthCare, Kass-Hout said. 

The company is planning to introduce additional apps under the CareIntellect brand in the future, Kass-Hout said. The oncology tool is the first offering, and health-care organizations will be able to easily pick and choose the apps that they want to enable, he added.  

GE HealthCare is also hoping to integrate its CareIntellect products with some of the other early stage AI initiatives it teased on Monday.   

The company highlighted five new AI products that it is developing, including a collaborative team of AI agents, a tool to predict an aggressive type of breast cancer recurrence, and a tool to flag suspicious mammography scans to radiologists more quickly. 

GE HealthCare decided to preview the new tools to give customers an idea of the problems it’s trying to solve, Kass-Hout said. The company will solicit feedback from health-care organizations and work with regulators as necessary, he said. 

For instance, GE HealthCare is exploring how a group of AI agents can work together as a team to support doctors through its tool called Health Companion.

The agents in Health Companion will be trained as experts in specific domains, such as radiology, pathology or genomics, and offer insights based on their expertise, Kass-Hout said. The agents could identify whether a specific symptom is a side effect of treatment or a sign of disease progression, for example, and suggest next steps, he added. 

Ideally, the tool will give doctors the same kind of support they’d expect from working with a multidisciplinary team, Kass-Hout said. But while consulting a panel of experts can take days or weeks, Health Companion would be available immediately. 

“At the moment, it’s an early concept,” he said. “Our aim is to elevate the standard of care and get ahead of the overburden of clinicians trying to take care of their patient.”

WATCH: The Pulse of AI in Health Care