• AiNews.com
  • Posts
  • Color Health and OpenAI Transform Cancer Care with AI Copilot

Color Health and OpenAI Transform Cancer Care with AI Copilot

A futuristic hospital setting where doctors are using advanced AI technology. One doctor interacts with a high-tech screen displaying a patient's medical data and personalized treatment plan, highlighting the AI copilot assisting in cancer treatment

Color Health and OpenAI Transform Cancer Care with AI Copilot

Color Health is collaborating with OpenAI to pioneer a new approach to cancer treatment, leveraging the reasoning capabilities of GPT-4o to improve patient outcomes. Their innovative copilot application identifies missing diagnostics and creates tailored workup plans, enabling healthcare providers to make evidence-based decisions about cancer screening and treatment.

Improving Access to Cancer Care

For over a decade, Color Health has been committed to enhancing access to healthcare, serving more than 7 million patients. In 2023, they partnered with the American Cancer Society to support employers and health plans in managing cancer care, a leading cause of death and a major healthcare expense in the United States.

Personalized, Comprehensive Treatment Plans

Using OpenAI’s APIs, Color Health integrates patient medical data with clinical knowledge to create a copilot application that assists clinicians in developing customized treatment plans. This tool extracts, processes, and normalizes patient information, including family history and individual risk factors, alongside clinical guidelines. It generates personalized screening plans and required documentation for diagnostic workups.

“Color’s vision is to make cancer expertise accessible when it can most impact a patient’s healthcare decisions,” says Othman Laraki, CEO of Color Health. “Technology that improves access and equity must also support patient safety and privacy, which OpenAI's HIPAA-compliant standards ensure.”

How the Copilot Works

The copilot application functions with a clinician-in-the-loop model, ensuring that its outputs are reviewed and, if necessary, modified by a clinician before being presented to the patient. The application’s capabilities include:

  • Extracting and normalizing patient information from various formats.

  • Answering key questions to identify missing diagnostics.

  • Generating necessary documentation for diagnostics and treatment plans.

  • Addressing Delays in Cancer Treatment

Cancer screening and treatment are complex and time-consuming processes. Delays can significantly impact patient outcomes, with a four-week treatment delay increasing mortality risk by 6-13%. Color Health’s copilot helps mitigate these delays by expediting the creation of personalized screening and treatment plans.

Dr. Keegan Duchicela, a primary care physician at Color, highlights the complexities of developing individualized cancer screening plans. “Guidelines are constantly evolving, and individual risk factors aren't always immediately clear,” he explains.

Building a Proof of Concept with OpenAI

Color Health began collaborating with OpenAI in 2023 to address the challenges of cancer screening, diagnosis, and treatment. They aimed to:

  • Interpret inconsistently-formatted patient data.

  • Analyze dense healthcare guidelines.

  • Protect patient data privacy.

  • Support clinician-in-the-loop workflow design.

  • Integrate with electronic health records (EHRs) and hospital systems.

Through rapid experimentation, including using GPT-4o to extract information from complex clinical guidelines, Color developed effective proofs of concept with OpenAI's guidance. This collaboration led to the creation of the copilot application, incorporating HIPAA-compliant data protection standards and retrieval-augmented generation (RAG) to enhance output quality.

Reducing Time to Treatment

To measure the copilot’s impact, Color Health is partnering with the University of California, San Francisco Helen Diller Family Comprehensive Cancer Center (UCSF HDFCCC). They plan to conduct a retrospective evaluation followed by a targeted rollout, potentially integrating the copilot into clinical workflows for all new cancer cases at UCSF.

“UCSF is a leader in implementing technology to improve patient care,” says Dr. Alan Ashworth, President of UCSF HDFCCC. “We seek tools that can improve pre-visit charting efficiency and reduce delays in treatment initiation.”

Dr. Karen Knudsen, CEO of the American Cancer Society, adds, “Combining AI technologies with digitally-enabled clinical workflows could expedite the treatment process, benefiting patients, clinicians, and payers.”

Future Plans

Color Health is taking a phased approach to roll out the copilot, starting with its own clinicians and applying the tool to a limited number of cases. The copilot has already demonstrated significant improvements:

  • Identifying 4x more missing labs and diagnostic results.

  • Reducing the time to analyze patient records from weeks to an average of 5 minutes.

By the end of 2024, Color Health aims to use the copilot to provide AI-generated personalized care plans, under physician oversight, for over 200,000 patients.