Dr. Claire WestonDr. Claire Weston is an accomplished and dedicated scientific leader with a track record of success in cancer research.

We’ve been happy to share excerpts from interviews Sean Ferrel and the Managed Solutions team have conducted with Todd Stewart, VP of global infrastructure and IT operations for Western Digital; Gavriella Schuster, corporate vice president, one commercial partner with Microsoft; Jeremy Giacosa, IT director with Accriva Diagnostics; and Ken Lawonn, senior vice president and chief information officer with Sharp HealthCare; Matt Webb, IT director with Mirati Therapeutics; and Helen Norris, CIO with Chapman University.

Now, check out Sean’s interview with Dr. Claire Weston, CEO and founder with Reveal Biosciences.

Here are excerpts from the interview:

In simple words, how do you help people?

We provide data from microscope slides or pathology samples that can benefit research, clinical trials and patients. For example, we generate quantitative pathology data to help pharmaceutical companies develop therapeutic drugs. We use it for clinical trials to increase precision and stratify patient groups. We’re also in the process of building pathology data applications to help pathologists diagnose disease in a way that will ultimately benefit patients.

How are your services different from other similar companies in the market?

We are fairly unique in that we have a scientific team in the lab doing pathology and a computational team of data scientists and software engineers who are developing our AI-based platform. Our ImageDx platform includes models to generate very quantitative data and diagnostic outputs that can be applied to many different diseases. The products that we are working on are unique and differentiate us, but the main driver is the quantitative pathology data that we generate.

How did you marry artificial intelligence with pathology?

We’ve been using traditional machine learning to identify and quantify cells from images for a while, but in the last few years, AI has advanced significantly. It’s impressive to see how well it works in pathology images. We’ve made the natural evolution from more traditional machine learning into AI. Compute power is now more readily available, which means that we can generate data from one patient slide in minutes, rather than the days or weeks it used to take. This sea change in computational speed means that the data we generate is more meaningful and relevant to routine pathology workflows.

You are planning to use cloud-based technology to deliver accurate diagnosis and to address medical needs worldwide. How does that work?

There’s a huge shortage of pathologists worldwide. Even in the US, where we have very highly qualified pathologists, we’re heading for a retirement cliff, and less pathologists are coming through residency to maintain their numbers. This is particularly evident in rural areas where there’s a real shortage of expertise. Having a cloud-based approach will help address some of those problems.

I’m excited by the potential for AI in a cloud-based platform to bring advanced pathology expertise to anywhere with internet access. Hospitals or pathology labs throughout the world could upload an image from a microscope slide into the cloud, and that image can be analyzed to generate advanced diagnostics. Countries with limited resources often have the ability to generate the most basic kind of microscope slide, but they sometimes lack the ability to do the more advanced diagnostics. The possibility to do so is going to revolutionize pathology and be impactful for healthcare globally. This should also benefit patients in the US by helping to lower the cost of healthcare.

How is AI impacting pathology?

The application of AI in pathology is a very new thing. We’ve been developing this for a while and we’re launching the first products in the clinic for patients in 2019. We are also building more enhanced pathology models by integrating other data sources. We’re finding that we can use AI to detect aspects of cancer that are not obvious just by looking down a microscope. For example, we’re detecting small changes in the texture of the nucleus of cells or small cellular changes that you wouldn’t necessarily notice by eye, but can be predictive or prognostic of disease. I think this is going to be really impactful for personalized medicine.

(Editor’s note: To read the full interview with Claire, click here.)

-- Sean Ferrel, Managed Solution