Leveraging Microsoft’s Azure OpenAI, Ontada streamlines oncology data analysis, reducing processing times by 85% to support clinical decisions and improve patient care for over 3000 oncologists across 30 states, according to Wanmei Ou, PhD, vice president of product, data analytics, and artificial intelligence (AI) at Ontada.
Ontada, a McKesson business unit, has revolutionized oncology data management with its flagship electronic health record (EHR) system, iKnowMed, used by more than 3000 community-based oncologists and supporting millions of patients. Rigorous model development and ongoing monitoring ensure data quality, while future innovations focus on expanding real-world data applications in oncology research and addressing health equity in rural communities.
Wanmei Ou, PhD, vice president of product, data analytics, and artificial intelligence (AI) at Ontada, discusses how by partnering with Microsoft, Ontada has reduced oncology data processing from 6 months to just weeks, alleviating documentation burdens and enhancing clinical decision support.
Ou explained the importance of using AI and natural language processing (NLP) to unlock unstructured data in oncology, as 70% to 80% of critical patient information is embedded in nonstandardized documents due to the interdisciplinary nature of oncology care. These data span across diverse systems and involve various specialists, including oncologists, surgeons, and primary care providers.
AI applications in this field serve 3 primary use cases, according to Ou. The first use is extracting real-world evidence from unstructured data for organized access. Second, AI applications reduce documentation burdens, allowing physicians to focus more on patient care than data entry. Third, these applications provide clinical decision support by integrating patient context, treatment guidelines, and insurance policies to recommend optimal next steps.
Ontada processes a massive amount of unstructured data—around 150 million documents from 2 to 3 million patients’ longitudinal records, said Ou. Due to the sheer scale, even Microsoft hadn’t previously encountered such a high-volume processing need within a short time frame. Through close collaboration and multiple on-site meetings with Microsoft’s technical specialists, Ontada leveraged the new Azure OpenAI, reducing processing time from an expected 6 months to just 2 to 3 weeks. This partnership has enabled Ontada to handle vast data more efficiently, and the solution may benefit other organizations facing similar high-volume data challenges, Ou noted.
Stuck in Prior Auth Purgatory: The Hidden Costs of Health Care Delays
June 19th 2025Delays, denials, and endless paperwork—prior authorization isn’t just a headache for providers; it’s a barrier for patients who need timely care, explains Colin Banas, MD, MHA, chief medical officer with DrFirst.
Listen
From Red Tape to Relief: Rewriting the Rules of Prior Authorization
June 23rd 2025Up to 257 million Americans could benefit from these prior authorization reforms that could have cross-market implications on health care plans administered through commercial insurers, Medicare Advantage, and Medicaid.
Read More
Culture Key to Data Collection of Sexual Orientation and Gender Identity in Cancer Care
June 23rd 2025This research highlights the key factors, like leadership buy-in, mandatory protocols, and electronic health record workflows, that influence the effective collection of data on sexual orientation and gender identity in outpatient oncology clinics to improve patient-centered care.
Read More