Dr Enrique Velazquez Villarreal - City of Hope, Duarte, USA
This study was about to create for the first time an artificial intelligence software to integrate the clinical, the genomic and the social determinants of health in order to better understand our patients and for the first time coin this as an integration of three levels or three layers of knowledge to provide and basically predict the healthcare to provide this comprehensive care as one of our main missions in City of Hope.
What was the study design?
This study was designed from scratch. We basically generated for the first time a cognitive model which is based on a Python-level script where we tried to find the best way to integrate data by having a conversational activity with our software. Basically this tool is a conversational tool that was designed to basically have these conversations with a software that can basically analyse the data for us, integrate the data for us, and provide recommendations based on the analysis, deep analysis, at the bioinformatics level.
What were the results of this study?
The result of this study is that we provide and basically we developed for the first time an artificial intelligent conversational tool that helps us not just to analyse data but also to interpret data and integrate data at different levels by using databases, public databases, that everybody can basically download and understand. One of our main points is that this model will be free for everybody, it’s right now at the publication part, where people can basically check this crystal model of AI, check it and familiarise with this, be confident about how it works and, of course, let us know questions since we built the model for the community, the scientific community, and we are so happy at this point to share the first model as a conversational tool to integrate clinical, genomic and social determinants of health data basically from a public source.
What do you think is the significance of these results and what is next for this study?
The significance of these results is that basically we compared different journals, different studies, that had exactly the same results that our AI agent, AI conversational tool agent. One of the impressive parts here is that the AI conversational tool agent that we developed brings results, very accurate results, in minutes just in comparison with usually projects that take days and sometimes months to generate the same type of results.
One of the key components in this generation of very accurate with a short time the results is basically based on the embedded bioinformatics and oncology knowledge from the software that interacts 24/7 in order to understand better how to integrate the data and compare with different studies a more accurate, a more precise way to show results and analyse the data.
