iNav: How AI-Driven Navigation Is Accelerating Pancreatic Cancer Care With Daniel King, MD, PhD, and Waqas Haque, MD, MPH

Pancreatic cancer remains one of oncology’s most challenging diseases—fast-moving, hard to detect early, and often diagnosed after curative options have narrowed. In this episode of Exploring AI in Oncology, Dr. Waqas Haque, Hematology/Oncology Fellow at the University of Chicago, speaks with Dr. Daniel King, Gastrointestinal Medical Oncologist at Northwell Health, Assistant Professor at Northwell’s Feinstein Institutes for Medical Research, and co-inventor of iNav, an AI-enabled navigation platform designed to expedite diagnosis and treatment for patients with suspected pancreatic cancer.

Identifying Patients Earlier Through Radiology Report Analysis

Dr. King’s clinical experience motivated the development of iNav at Northwell’s Feinstein Institutes for Medical Research. He explained, "Unfortunately, pancreatic cancer is a very aggressive, rapidly moving disease, and anything we can do to start treatment faster improves outcomes." The key challenge was finding ways to intercept patients before they entered the oncology clinic—often the point when the disease is already advanced. 

iNav focuses on a strategic trigger: radiology reports. Rather than interpreting scans directly, the platform uses a natural language processing (NLP) model to scan radiology reports for language suggestive of pancreatic malignancy. This approach revealed patients who had radiologic suspicion of pancreatic cancer but did not yet have a confirmed diagnosis.

The Crucial Link Between AI Triggers and Human Navigation

A common pitfall of AI tools is the lack of a true workflow bridge between detection and actual care; iNav was intentionally built with this connection in mind. Dr. King emphasized, “What was really important for us to do was develop a workflow that not only found the patient who is at risk, but linked that identification to a clinical implementation workflow."

Once the NLP model flags a suspicious radiology report, that case is added to a daily review list managed by Dr. King, a navigator, and a coordinator. The navigator then reaches out to the involved clinicians and the patient to ensure that recommended next steps are scheduled without delay. This human-in-the-loop model is central to iNav’s impact. "The navigators are very good about ensuring expedited healthcare delivery. They make sure people are not getting left behind," Dr. King said.

Real-World Improvements in Time to Treatment 

Early data from iNav’s first published study offer encouraging signs. Although the pilot included only 71 patients and was not powered for statistical significance, multiple metrics improved:

• Time from suspicious imaging to biopsy dropped from 12 days to 6

• Time from imaging to oncology consultation decreased from 27 days to 17

• Time from imaging to treatment initiation improved from approximately 8 weeks to 5

These reductions matter in pancreatic cancer, where delays can transform a borderline-resectable lesion into an unresectable one. Dr. King shared that the system helps ensure "patients who have lesions that are resectable make their way to resection," preventing cases where patients "present later with lesions that are no longer resectable."

Balancing Early Detection With False-Positive Concerns

One question many clinicians have about AI detection systems is the risk of overloading teams with false positives. Dr. King acknowledged that radiologic suspicion is a spectrum, and some flagged patients ultimately do not have malignancy. 

However, iNav’s workflow does not mandate biopsies or unnecessary procedures—it simply prompts a navigator-led evaluation. "Our technology does require the person in the middle; there's a human step in between," Dr. King explained. He added, "I think that sacrifice is worth it, because the benefit is that we are improving care delivery for patients with known pancreas cancer, and ultimately we hope to improve clinical outcomes."

Funding, Research, and Innovation 

The initial pilot helped Dr. King’s team secure support from the American Society of Clinical Oncology (ASCO) Conquer Cancer Foundation to launch a larger randomized trial within the health system. Northwell’s internal innovation programs, including a “Shark Tank”–style competition, also played a pivotal role, providing early funding and visibility. iNav’s design and early results even earned recognition as one of TIME’s Best Inventions of 2024. 

Advice for Clinicians Exploring AI in Oncology

Dr. King encourages oncologists with an interest in digital innovation to seek out disease-specific foundations, federal grants, or philanthropic support. "Our mission is to show that AI could be a real force for good when it comes to improving healthcare delivery," he said. 

What’s Next for iNav? 

Enrollment for the randomized trial is 40% complete, with results expected in the next 18 months. If findings confirm early signals, iNav may represent a scalable model for AI-driven navigation across other malignancies. For pancreatic cancer, where every day matters, AI-enhanced navigation may become a powerful tool for bringing patients to care faster and supporting multidisciplinary coordination.

Key Takeaways 

• AI as a Workflow Trigger: iNav uses natural language processing to scan radiology reports for suspicious findings, identifying high-risk patients before they even reach an oncologist.

• Human-in-the-Loop Navigation: The platform’s success relies on linking AI detection to a dedicated navigation team.

• Significant Time Savings: Early pilot data showed that AI-triggered navigation could reduce the time from initial imaging to treatment initiation by approximately three weeks (from 8 weeks to 5).

• Preventing "Lost to Follow-Up": By proactively flagging suspicious cases, the system helps ensure that patients with resectable lesions do not progress to unresectable stages due to administrative or coordination delays.

• Clinical Innovation Pathways: Dr. King highlights that clinicians can drive AI innovation through health system challenges and foundation grants. "I feel lucky to be part of a healthcare system that truly try to inspire its providers to come up with innovative solutions." 

About the Speakers

Daniel King, MD, PhD, is a Gastrointestinal Medical Oncology at Northwell Health Cancer Institute, where he is also the Director of Research and Development in the Center for Genomic Medicine. Dr. King is passionate about development targeted therapies and breakthrough technologies for improving cancer outcomes. He is the co-inventor of iNav, an AI-powered pancreatic cancer detection tool which was included in TIME’s Best Inventions of 2024.

Waqas Haque, MD, MPH, is a second-year Hematology/Oncology Fellow at the University of Chicago. He recently completed his Internal Medicine Residency at New York University (NYU) in a Clinical Investigator Track. Dr. Haque has balanced his patient care work with a variety of research projects. During his fellowship training at the University of Chicago, he plans to further his work in innovative clinical trial design, value-based care delivery to cancer patients, and clinical investigation.