On April 25, 2023, myTomorrows, a platform designed to help patients obtain access to experimental medicines, recently launched a new tool called TrialSearch AI, an artificial intelligence-powered resource for physicians that will allow them to quickly and efficiently identify expanded access programs (EAPs) or clinical trials that their patients may be eligible to join.
Physicians using TrialSearch AI will need to take into account the entirety of a patient’s medical record. In order to match a patient to a potential trial, they have to take their record and compare it to the trial’s eligibility criteria, which are often quite specific. This tool will allow them to complete this challenging task more quickly. The tool will cross-reference a summary of a patient’s medical data with criteria drawn from publicly available registries and present a physician with a list of possible EAPs or clinical trials that fit that patient’s unique profile.
However, the physician will still need to do manual checks and certify that pre-approval is appropriate for the patient. Nevertheless, TrialSearch AI is expected to save critical time for physicians and, by extension, patients themselves. The tool will tap into myTomorrow’s search engine, which is optimized for people in search of pre-approval options and draws from public registry data.
Efficiency is key to fulfilling the mission of myTomorrows, and new technologies like AI can play an important role to improving experiences for users while centering the needs of physicians and their patients. The tool could also ultimately be valuable for improving enrollment for clinical trials as a whole.
TrialSearch AI will utilize a form of AI called Large Language Models–in this case, the GPT models from Azure OpenAI. The model will take the patient medical summary submitted by the physician and search against all available trials that are listed in public registries. The medical summary and the public registry info are the critical sources of data in this process, not info sourced from the AI model itself. For more info on how AI will be used in this tool, click here.
For more information about TrialSearch AI, click here.
Physicians can sign up for the beta version here.
