Innovating with AI to Advance Colonoscopy

AUTHOR: Anthony Borrelli

Artificial intelligence (AI) is seemingly everywhere. We see it on our smart phones with virtual assistants like Siri and Alexa, we see it in advanced security and surveillance systems, and we’re increasingly seeing it implemented in the medical field.

I recently had the pleasure of being a guest on Gastro Broadcast, an independent GI-focused podcast, and was interviewed by Dr. Fred Rosenberg about CAD EYE®, Fujifilm’s novel AI detection system for endoscopic imaging. That’s right — Fujifilm created an AI-powered system to elevate performance and quality in colonoscopy.

During our conversation, Dr. Rosenberg and I discussed the evolution of endoscopy and how AI has great potential to help clinicians detect pre-cancerous lesions regardless of size, shape, and color.

Here are some of the topics we discussed:

  • Evolution of our Endoscopy Business — At Fujifilm, we introduced our first endoscope in 1985. Since then, we’ve made tremendous strides in image quality, enhancements, and ergonomics. Our focus was originally on ambulatory surgical centers, but since 2019 Fujifilm has introduced more than 20 new and innovative endoscopic technologies, and we are now a trusted vendor of state-of-the-art GI imaging for academic medical centers and community hospitals across the
  • AI Terms to Know — Dr. Rosenberg and I discussed the importance of understanding the language associated with AI. For example, physicians often ask about the difference between machine learning versus deep learning. When it comes to AI detection tools, machine learning refers to a computer’s ability to analyze optical information and look for variances. Deep learning has more to do with the evolution of that algorithm. For example, if it can educate itself, it has the potential to do more. At Fujifilm, our job is to help software evolve and develop AI algorithms to improve or even detect other disease states.
  • CAD EYE: What the Data Shows — We recently received 510(k) clearance for our first of its kind AI system for endoscopic imaging, CAD EYE. Studies show that CAD EYE is helping clinicians to decrease missed rates, reduce removal of innocuous polyps, and improve adenoma detection rates 1. Physicians are leveraging CAD EYE to be more efficient. It’s supporting them to find potentially cancerous lesions and adenomas as quickly as possible.
  • Clinical and Economic Benefits of AI — We talked about the role AI will play in the future of endoscopy. Fujifilm anticipates enormous clinical value as AI evolves. For example, we may see AI tools that diagnose as well as detect, others that optimize the passage of the scope or optimize passage of the needle into lesions, and still others that tell clinicians how effective they are in sampling lesions. As for economic impact, if clinicians can make diagnoses faster or more accurately due to AI, patients can get the treatment they need sooner. That, in turn, minimizes the downstream impact of that patient needing to be readmitted or treated again.

It’s an exciting time for endoscopy. Though we discussed AI for colonoscopy specifically, the fact is, that anywhere an endoscope goes there’s an opportunity for AI to benefit that procedure. There are so many different aspects of AI that can be implemented down the line that can support many clinical procedures and ultimately, enhance patient outcomes.

You can learn more about Fujifilm’s novel AI detection system for endoscopic imaging by visiting here. To learn more, listen to the podcast on Gastro Broadcast.

You can also stream the podcast on the following services:

1. Neumann H et al. 2021. PLoS One 16(8):e0255955, PMID: 34437563.


Anthony Borrelli, Executive Director of Endoscopy Product and Marketing for FUJIFILM Healthcare Americas Corporation.