Memorial Sloan Kettering partnered Feb. 17 with Amazon Web Services to use AI, high-performance computing and cloud technologies to boost cancer care and research. Becker’s caught up with Yashodhara Dash, MD, PhD, vice president of technology development and commercialization at the cancer center, to learn more about the work.
Question: How did this collaboration come together, and what are your hopes for it?
Yashodhara Dash: We have a lot of research and clinical pursuits, and more and more we are becoming aware of the need to incorporate AI and cloud computing into the work that we do. We’ve been working with AWS already on a more focused basis, and we believe they were the perfect partner for us to combine our deep oncology knowledge and our data assets with their cloud infrastructure, AI tools and expertise.
Q: What are some specific use cases you plan to get out of this collaboration?
YD: We have three major areas of focus. We want to partner with them to unlock our access to our data. So we have a lot of really valuable clinical data, which we would like to curate and make accessible to our researchers.
The second major area of focus is enhancing MSK innovation. Our scientists and clinicians are always engaged in innovative research, and quite often, they will come up with discoveries. We want to work with Amazon to expand and advance those faster. So we have specific programs, including our innovation hub, which we want to be able to leverage AWS support for. We’re also starting a new AI-focused technology development fund and a new venture incubator.
And then, we want to leverage AWS expertise and training resources to train our own scientists and clinicians in the language of AI, so that our scientists become fluent not only in the language of biology, but also technology.
Q: Why was AWS the best partner for this type of work?
YD: We’ve been collaborating with them on a few different projects already, and it seemed to us that they really understand the type of work we do, the complexity of the work we do.
Their technology is somewhat platform agnostic, and so we believe we will be able to use tools from other sources, other companies, and integrate them with the AWS platform. We also have some in-house AI efforts, and we think that integration will be easier because of the agnostic nature of their platform.
Q: What uses do you think will come out of this collaboration for AI and large language models specifically?
YD: Some of the use cases might include, for example, better tracking how a patient’s cancer changes over time, being able to combine clinical genomic imaging data and outcomes data. So we want to be able to uncover new insights, personalized treatments, improve patient care, all from being able to access our data.
Q: Industrywide, what gaps still exist when it comes to AI, data and precision medicine?
YD: We are very interested in drug discovery and development. So our scientists working in the labs, they’re studying mechanisms of cancer, and sometimes they come up with new targets that can be interrogated to create new drugs. We believe there is a gap between those biological insights and being able to predict what kinds of molecules can be used to interrogate those insights.
Beyond target discovery, if you go into the refinement of specific drug molecules, for example, chemical molecules that you want to make into a drug to be able to treat patients, how can you tweak those molecules?
