City of Hope uses Microsoft Azure and AI to rapidly onboard thousands of patients per year

With the rise of AI, healthcare institutions like City of Hope are seizing the opportunity to transform physician workloads and empower more focused, personalized patient care. By harnessing the power of Microsoft Azure, City of Hope has developed an innovative generative AI platform that supports oncology doctors in preparing for new patient appointments and building treatment plans. This cutting-edge solution helps doctors spend less time reviewing documents and more time engaging with patients face-to-face, revolutionizing the way care is delivered.

Transforming cancer care with AI: City of Hope’s legacy of innovation

Founded in 1913, City of Hope is one of the largest and most advanced cancer research and treatment organizations in the United States. With a rich history of using cutting-edge technology and innovations, City of Hope continues to stay at the forefront of advances in data science and AI to bring breakthroughs to its patients.

Recognizing the potential of recent AI advancements, City of Hope set out to tackle one of the healthcare industry’s biggest challenges: the administrative burden of patient referrals. In 2024 alone, City of Hope onboarded more than 150,000 patients. Doctors often see new patients with extensive medical histories spanning 10–20 years of test results, notes, and treatment reports. Doctors need to spend significant time—often during nonworking hours—to review hundreds of document pages, leading to less time interacting directly with patients due to extensive medical history review.

“One of our physician executives jokingly told me that because this is such a big pain point, if we could come up with a solution to digest all this information and provide an accurate summary, my statue would be placed in the City of Hope hospital,” says Nasim Eftekhari, Chief AI Officer at City of Hope. “That’s how we started working on this use case.”

City of Hope aimed to build a responsible large language model (LLM) platform to give valuable time back to doctors and their patients. From the outset, City of Hope’s decision makers knew they wanted to use Azure OpenAI Service and the optical character recognition technique, which recognizes printed and handwritten text in image files, as instrumental parts of the solution. “We want to work with the best and brightest out there who can bring the right technologies to bear for us,” says Simon Nazarian, Chief Digital and Technology Officer at City of Hope. “That’s where Microsoft came in. It’s a very mutually agreeable collaboration, where both sides are dedicated to advancing what is best for healthcare.”

Building a time-saving solution in three months

City of Hope created minimum viable capabilities for its physician-facing LLM platform in just three months. Then, it worked closely with Microsoft cloud engineers from customer success programs to bring the groundbreaking generative AI solution to life. “City of Hope has closed the loop of going from data and information to action and continuous optimization,” says Eftekhari. “It’s not just conceptual; our clinicians are using this solution now.”

City of Hope’s solution is one of the first uses of AI in general oncology, and the team worked diligently to implement it responsibly. The solution underwent multiple rounds of approvals, regulatory controls, and adoption scenarios to ensure accurate summaries of patient information. Remarkably, the AI platform provided a clean and comprehensive view of patients’ histories.

With this solution, doctors at City of Hope can drastically reduce the time spent reviewing patient referral documents and thus focus more on their patients and personal lives. Detailed and searchable information at their fingertips helps doctors quickly determine the best treatment options. “The main driver for this project was improving provider and patient experiences,” says Eftekhari. “Doctors can now look their patients in the eyes, already knowing their history, rather than trying to scramble to look at PDFs.”

This solution not only boosts physician productivity but also reduces costs. Doctors can potentially see more patients, thanks to the time-saving features, and avoid ordering redundant tests, scans, or biopsies by identifying prior labs and tests.

Patients benefit directly from the AI solution as well. Typically, patients and doctors share the burden of providing and verifying medical history. However, City of Hope’s Azure-powered platform shifts this responsibility away from patients, helping them to receive more personalized care and engage in focused, future-looking discussions with their doctors. Now, patients can concentrate on their primary responsibility: recovery.

Empowering high-quality patient care

Building an innovative generative AI platform with support from Microsoft has been a rewarding journey for City of Hope’s data and technology teams. Not only did they develop new guidelines, processes, and tools to create a solution with wide-ranging impacts in the healthcare industry, but they are also aiming to make a direct, positive impact on the quality of patient care as well as the patient and provider experience. “You can work on algorithms anywhere,” says Eftekhari. “But bringing them to life and having them improve and save lives—that’s not something you can do everywhere.”

Looking ahead, City of Hope plans to continually enhance its AI solution to empower even better experiences for doctors and patients. The institution’s next step is exploring more applications of the LLM platform, such as clinical-trial matching, and incorporating multimodal AI models to work with medical images and genomics in addition to text and electronic medical record data.

“We want to take advantage of innovative technologies to support real-time decision making,” says Nazarian. “We want to do it responsibly. We want to do it together. And, at the end of the day, we want to help improve healthcare for all.”

Discover more about City of Hope on Facebook, Instagram, LinkedIn, and X/Twitter.