Digital reconstructions in clinical trials. New method for selection of patients based on RWD via Danny Platform
- company News
- 1 month ago
Sqilline took part in the Annual Oncology Conference MORE 2021. This year, the three main topics were: pancreatic cancer, national cancer plan, and digitization in oncology.
During the panel on "Digital reconstructions in Bulgarian clinical oncology," we had the pleasure of demonstrating, live our Cohorts application with precisely its methodology for selection of patients in clinical trials and specifically for the category of observational studies.
Observational studies (both prospective and retrospective) are essentially made to increase clinical knowledge. They are explanatory, they observe, they give responses based on the real-world data collected, and they conclude based on all that information. In order to produce a high-quality observational study and provide valid estimates of exposure/treatment, the patient data included should be very carefully picked out.
Mihail Zekov, Product Architect for Sqilline Healthcare Solution was the speaker at the Conference. He successfully demonstrated how technologies and machine-learning algorithms can optimize the time and effort in terms of clinical trials, medical publications, and statistical analyses.
Sqilline developed an algorithm that automatically searches by a set of keywords and from free text in all patient documentation - epicrisis, oncology protocols, laboratory, tests. In just minutes, our technology can process all medical records of a specific hospital going years back. All selected patients' records are then visualized in Cohorts. Through quick and easy access, the profile of each patient is reviewed by the investigating physician. Once a specific patient is selected for examination, the specific clinical trial criteria are effectively visualized with all medical evidence - diagnosis, prescribed therapies, genetic markers, TNM stage, habits, and others. For the respective study, the physician needs to further mark each record as eligible, a possible match, or not eligible.
The solution marks the beginning of a new, fundamental method to quickly analyze high dimensionality and complex health data to provide results and empower both physicians as well as sponsoring companies to accelerate the current process of patient selection and manage each study analysis more effectively.
The digital processing of large datasets and the use of advanced AI and ML could enable health applications to be much more scalable these days and Sqilline has already taken the steps in this direction.