
Medtronic Chief Technology and Innovation Officer Ken Washington (left) and Medtronic Endoscopy Chief AI Officer Ha Hong pose with a scope used with the GI Genius Intelligent Endoscopy Module at the company’s facilities in Santa Clara, California. [Photo by Hardy Wilson for Medical Design & Outsourcing]
“It’s not easy to innovate at such speed, because they worked on several different algorithms and for each algorithm, you need to obtain high quality data and also high quality for each data point,” he said in an interview with Medical Design & Outsourcing. “That means the right amount of diversity, and also for each data point, a high quality label.”
MDO also interviewed Cosmo Intelligent Medical Devices President Nhan Ngo Dinh and Science, AI and Data SVP Andrea Cherubini to learn more about the system’s development and what tips and advice they had for other device developers. In our second post from that interview (read the first one here), we discussed how the Cosmo AI team developed the latest version of GI Genius so quickly.
And because Cosmo Intelligent Medical Devices was a startup called Linkverse before its acquisition by Cosmo Pharmaceuticals, we also asked how other startups should approach the medtech space.
(Editor’s note: While MDO prefers live interviews to offer readers medtech expertise without the filter of marketing or public relations staff, we conducted this Q&A by email through a Medtronic representative because Cherubini and Ngo Dinh live and work in Italy. The following has been lightly edited for space and clarity.)
MDO: Ha Hong said he was impressed by how quickly your team came through on this project. Can you offer some details about how long the Medtronic project took and how that breaks down into the different stages/phases?
![A photo of the Medtronic GI Genius ColonPro polyp detection system flagging a potential sign of colon cancer during a colonoscopy. [Photo courtesy of Medtronic]](https://www.medicaldesignandoutsourcing.com/wp-content/uploads/2024/07/Medtronic-GI-Genius-ColonPro-300x195.jpg)
The Medtronic GI Genius ColonPro polyp detection system identifies potential signs of colon cancer in real-time during colonoscopies. [Photo courtesy of Medtronic]
Related: 9 tips for implementing AI in medical devices from Medtronic’s Ha Hong
MDO: When expanding the data sets, how did you balance data quantity and quality? And how did you get this Medtronic project over the finish line as quickly as you did?

Andrea Cherubini is SVP of science, AI and data at Cosmo Intelligent Medical Devices [Photo courtesy of Cosmo]
“Transparency was a cornerstone of our approach. We communicated the purposes of data collection clearly to all stakeholders, fostering trust and enabling continuous feedback. Privacy and ethical concerns were addressed rigorously through anonymization protocols, protecting patient identities. Our project management framework emphasized agile methodologies and iterative development, allowing us to swiftly integrate feedback and optimize our algorithms. This approach ensured the efficiency of the project, which of course is still ongoing to further improve our AIs.”
Related: Medtronic leaders pick three technologies that are key for the future of endoscopy
MDO: Hong was saying that your team has maintained and benefitted from a small startup mentality, and he encouraged me to ask what advice you would share to help small startups approach the medical device field?

Cosmo Intelligent Medical Devices President Nhan Ngo Dinh [Photo courtesy of Cosmo]
“We learned that startups should focus on maintaining agility and embracing rapid prototyping. This allows for quick adjustments based on real-time feedback and emerging clinical needs. Integrating regulatory compliance from the beginning is essential to avoid costly setbacks later. Building strong connections with healthcare professionals for continuous feedback and validation is critical. These relationships provide insights into practical challenges and help tailor your product to meet clinical needs effectively. Establishing a multidisciplinary team that combines expertise in medical science, engineering, regulatory affairs, and business development will help navigate the complexities of the medical device landscape. Maintaining flexibility and continuously updating your knowledge base with evolving clinical and regulatory standards will ensure that your solutions remain relevant and effective.”
Look for more from Cherubini and Ngo Dinh at MDO in the weeks ahead, including how device developers can work with Cosmo on GI Genius apps and tips for using AI in medtech.