The Growing Complexity of Modern Oncology
Cancer care has evolved dramatically in recent years. What oncologists once treated as unified diseases are now recognized as numerous subtypes, each demanding tailored treatment approaches based on rapidly evolving clinical guidelines. This increasing complexity creates significant challenges for healthcare providers who must navigate an ever-expanding landscape of treatment options.
Oncologists today face the dual challenge of managing diverse cancer varieties while simultaneously keeping pace with constantly changing best practices. The burden is substantial and growing as medical research advances.
The Guideline Challenge in Cancer Treatment
A major hurdle in modern oncology is managing the sheer volume and complexity of clinical guidelines. Leading organizations like the National Comprehensive Cancer Network (NCCN), American Society of Clinical Oncology (ASCO), and American Cancer Society regularly update their recommendations—sometimes hundreds of times annually—based on new clinical trials, emerging therapies, and evolving treatment paradigms.
These guidelines often lack standardization across organizations, and individual cancer centers frequently add their own expertise layers. This creates a challenging landscape for clinicians striving to consistently apply the latest evidence-based practices.
Compounding this problem is the growing shortage of specialized oncologists, as Dr. Travis Zack, assistant professor of medicine at UCSF, explains: “Many regions face oncology specialist shortages, forcing general practitioners to handle more initial cancer workups and treatment planning. Unfortunately, GPs typically lack both time and specialized training to stay updated on the latest oncology guidelines, leading to care inconsistencies and treatment delays.”
Dr. Zack adds, “There’s also the fundamental challenge of navigating unstructured patient data and the time required to aggregate and review that information according to updated guidelines to make optimal patient recommendations.”
Innovative AI Solution Development
Recognizing these challenges, the University of California at San Francisco collaborated with health technology company Color to develop an artificial intelligence system that automates the process of aggregating, structuring, and applying the latest clinical guidelines alongside comprehensive patient information.
“Our goal was creating a decision support system that seamlessly integrates national guidelines and patient data with local institutional best practices, ensuring every patient receives the most current, evidence-based care possible—without adding cognitive burden to already overworked clinicians,” notes Dr. Zack.
This solution addresses the fundamental challenge of providing oncologists with quick, reliable access to up-to-date, evidence-based recommendations while optimizing physician time, making world-class oncology expertise more accessible, efficient, and scalable across all care settings.
How the AI System Works
The developed AI system combines a large language model informed by applicable national and local institutional guidelines with transparent logic, allowing clinicians to understand precisely how and why the AI generates its recommendations.
The system was designed with two core functions:
- Aggregating and structuring clinical data – The AI pulls and organizes relevant patient information from electronic health records to create a comprehensive view of the patient’s condition. If critical data—such as biopsy results, molecular testing, or staging scans—is missing, the system flags it before the oncology consultation to prevent unnecessary delays.
- Integrating national and local clinical guidelines – The AI incorporates both standard guidelines (from sources like NCCN, ACS, and ASCO) and institution-specific protocols, ensuring physicians receive the most relevant, up-to-date treatment recommendations tailored to each specific case.
“For example, if a patient is referred for suspected lung cancer, the system automatically assesses whether all necessary diagnostic steps have been completed,” explains Dr. Zack. “If a key test is missing, it prompts the referring physician to order it before the patient’s oncology visit. During the consultation, the AI provides an evidence-based framework for decision making, reducing the cognitive burden while ensuring adherence to best practices.”
The overarching goal isn’t replacing human judgment but enhancing it—allowing oncologists to focus on personalized treatment decisions rather than spending valuable time retrieving and verifying information.
Real-World Implementation and Testing
The AI technology was deployed in oncology workflows to support both general practitioners and oncologists, ensuring evidence-based insights guided each step in the patient journey.
For their published study, Color clinicians analyzed 100 de-identified patient cases from UCSF—50 breast cancer and 50 colon cancer cases. Each case included two sets of records: diagnosis records containing all available information up to and including the diagnosis date, and treatment records encompassing all information up to the treatment initiation date.
The evaluation process involved:
- Diagnosis run type: 100 patient cases using only records available up to the diagnosis date
- Treatment run type: 100 patient cases with records included up to the treatment initiation date
A primary care physician reviewed the AI-generated output and made adjustments where necessary. The system’s performance was assessed by tracking modifications in three key areas: accuracy of extracted decision factors, relevance of recommended workups, and completeness of relevant workups. Additionally, the study recorded the time required for clinicians to finalize each workup plan using the AI.
The AI System in Practice
In practical application, the system functions through:
- Data aggregation and structuring: Before oncology consultations, the AI automatically compiles relevant clinical information and identifies missing diagnostic steps
- Guideline-based recommendations: At the point of care, the system provides tailored recommendations based on current guidelines and institutional policies
- Continuous learning and updates: The AI dynamically incorporates the latest clinical research and guideline updates
“By reducing administrative tasks and eliminating care inconsistencies, the AI allows oncologists to focus on patient interactions and treatment planning, resulting in faster and more effective cancer care,” Dr. Zack notes.
Impressive Clinical Results
The implementation of AI in oncology workflows has yielded significant improvements in efficiency and decision-making quality. One notable outcome is the dramatic reduction in record review time for oncologists.
“Previously, reviewing patient records and clinical guidelines could take one to two hours, particularly for complex cases,” Dr. Zack explains. “With the AI system, this time has been reduced to approximately 10-15 minutes in most cases. By automating data aggregation and structuring relevant clinical information, oncologists can focus on decision-making rather than manual data retrieval.”
The study also revealed a 95% concordance between AI-generated recommendations and clinical decisions made by oncologists based on standard guidelines. This high alignment suggests the AI effectively synthesizes and applies guidelines in a way that supports clinical decision-making. While human oversight remains essential, this agreement level indicates the AI serves as a reliable tool for reinforcing evidence-based care.
Additionally, the system has improved treatment initiation timeliness. Delays in ordering essential diagnostic tests—such as biopsies or genomic testing—can extend the time between diagnosis and treatment by weeks or months. By identifying missing but necessary workups earlier, the AI system has helped reduce these delays, ensuring patients progress to treatment more quickly.
Implementation Advice for Healthcare Organizations
For healthcare organizations looking to integrate AI into oncology or other specialties, Dr. Zack advises a strategic approach:
“Ensure the AI system has access to comprehensive and accurate patient data. AI-driven decision support tools rely on complete datasets to generate clinically meaningful recommendations. Interoperability challenges between electronic health records and other data sources can result in incomplete clinical pictures, potentially affecting AI output reliability. Addressing these gaps through effective data integration and standardization should be a priority.”
Dr. Zack also emphasizes maintaining balance between AI recommendations and clinical judgment: “AI should support, not replace, healthcare providers. Organizations should ensure clinicians remain actively engaged in interpreting AI-generated insights and can override recommendations when necessary.”
He concludes, “AI systems should provide transparent and explainable decision pathways, allowing users to understand how recommendations were generated. Clear visibility into the underlying logic builds trust in AI-assisted decision-making and promotes adoption among clinicians.”
Discover the latest Provider news updates with a single click. Follow DistilINFO HospitalIT and stay ahead with updates. Join our community today!
Leave a Reply