Computer Vision turns into a very different discipline, when the task is not just counting people or everyday objects but rather distinguishing healthy from cancerous tissue in seconds; or detecting a hairline defect before assembly; or helping a robot act on parts that vary in shape from cycle to cycle.
In these environments, strong results depend not just on model selection, but also on engineering the full decision pipeline, around data conditioning, feature engineering, training strategy, the handling of edge cases, etc.
Using Reveal Surgical as a featured example, this webinar examined how Raman spectroscopy and AI are combined, in order to help surgeons assess tissue in vivo within seconds, during procedures where margin decisions really matter.
We connected these same design principles to other private-sector settings beyond healthcare, such as semiconductor inspection, food grading and drone-based damage assessment: all areas where performance depends greatly on accuracy, consistency and robustness
This webinar explored how specialized AI systems are ideally built, when the signal is hard to read, the workflow is unforgiving, and a wrong decision has real consequences.
In this webinar, you will find:
- Why many real-world vision solutions fail with generic AI approaches.
- How to choose the right class of algorithm for difficult signals and constrained workflows.
- What it takes to build a stable, predictable pipeline, from data conditioning to edge-case handling.
- How these principles apply across healthcare, manufacturing, robotics and inspection environments.
Speakers:
Michel Dubois, Operating Partner and Head of the Artificial Intelligence Lab, Innovobot Labs
Michel Dubois brings more than 30 years of leadership in AI and machine learning across business, technology, academia, intellectual property, and strategic partnerships. He has held senior roles at Newtrax, IBM, Mila, and Innovobot, where he serves as Operating Partner and Chief AI Scientist. His strength is connecting technical innovation with commercial growth, helping organizations scale advanced technologies into practical, market-ready solutions.
Christian Sauvageau, CEO, Reveal Life Science
Christian Sauvageau leads the strategy, operations, and commercialization path for Reveal Life Science’s AI-enabled surgical technology platform. He brings more than 30 years of leadership across life sciences, medical technology, diagnostics, and healthcare innovation, including senior executive roles at Merck and the leadership of a publicly traded diagnostics company. His strength is aligning clinical development, regulatory execution, partnerships, and capital formation so complex science can become a disciplined, market-ready product.
François Daoust, Co-founder and Chief Technology Officer, Reveal Surgical
François Daoust is Co-founder and Chief Technology Officer at Reveal Surgical, where he leads the development of the company’s AI-powered Raman spectroscopy platform. He has been a key technical force behind the technology since 2016. His strength is translating complex photonics, signal processing, and AI into practical clinical tools, helping bring real-time tissue characterization into operating room workflows and clinical studies.