Unlearn AI creates "digital twins" of individual patients using generative AI. These digital twins are simulated clinical records that provide a comprehensive model of potential health outcomes under different scenarios. Healthcare professionals can personalize treatment plans, monitor patient progress, and make informed decisions to improve patient outcomes.
Unlearn.AI's digital twins of patients can be used to simulate the effects of different treatments on a patient's individual biology by using real-world data. This allows HCPs to personalize treatment plans for each patient and monitor patients' progress during treatment, adjusting treatment plans as needed, which can help to improve patient outcomes.
As a concrete example in hematology, consider a patient diagnosed with chronic lymphocytic leukemia (CLL). The clinician could use Unlearn's digital twin technology to simulate the patient's potential disease progression under different treatment strategies, such as chemotherapy or targeted therapy. By comparing these simulated outcomes, the clinician could make a more informed decision on the best course of action for the patient. Additionally, the digital twin could be used in a clinical trial context, allowing researchers to better understand the potential effects of new treatments for CLL without the need for larger control groups.
In terms of supporting clinician professional education and better patient follow-up, these digital twins can be used to improve clinical research and decision-making. Unlearn develops novel methods for using the predicted health outcomes from patients’ digital twins to determine the best treatments for each patient. These applications help to streamline processes, making them more efficient and personalized.
A study published by Cornell University evaluated the performance of the Digital Twin platform against a held-out test dataset and showed how Digital Twins simultaneously capture the progression of a number of key endpoints in clinical trials across a broad spectrum of disease severity, including MCI and mild-to-moderate AD.
"Modeling Disease Progression in Mild Cognitive Impairment and Alzheimer's Disease with Digital Twins". December 2020.
https://arxiv.org/abs/2012.13455
Note: The generative AI used by Unlearn is based on deep learning models for multivariate time-series data, with these models built and deployed to forecast potential health outcomes for individual patients. This approach is not about creating an exact replication of a patient from an external cohort, but rather a computational model that offers a probabilistic forecast of their specific health outcomes.
