Skip to main content

Predictix ™ - a user-friendly procedure that analyses genomic, clinical and demographic data to generate a personalized report on the efficacy of antidepressants and their side effects

Periodic Reporting for period 1 - Predictix (Predictix ™ - a user-friendly procedure that analyses genomic, clinical and demographic data to generate a personalized report on the efficacy of antidepressants and their side effects)

Reporting period: 2020-09-01 to 2021-02-28

Major depressive disorder (MDD) is a common psychiatric disorder that is associated with a high burden on patients and their families. The World Health Organization estimates that, for every euro spent on digital solutions for mental health, countries can reap a return of four euros in improved health and productivity. Unfortunately, clinical trials indicate that current best-practice methods of determining the optimal treatment for a specific MDD patient lack efficiency. This inefficiency is plausibly caused by the polygenic nature and the phenotypic heterogeneity of MDD. Recent technological advancements lead to the accumulation of data through electronic health records, next-generation sequencing technologies, and sensory devices. This has paved the way to a new era of brain research; now we can start using Machine Learning (ML) as an advanced approach to understanding MDD.
Applying machine learning methods to the accumulating data derived from Next Generation Sequencing (NGS) technologies, Electronic Health Records (EHR), and sensory devices can transform the way psychiatric disorders are treated.
The PREDICTIX Antidepressant tool was developed out of this growing need for personalized treatment selection for patients diagnosed as suffering from depression. We aspired to create a prediction tool that relies on combinatorial data of clinical, demographic, and genetic information of each patient, in accordance with the applicable literature and guidelines.
We optimized the algorithms at the core of our PREDICTIX Antidepressant tool taking into consideration most up to date established clinical guidelines and medications’ information, antidepressant metabolism and applying refined Machine Learning algorithms to take into account genetic, clinical, and demographic features.
The output report, consisting of the likely effectiveness of 11 new-generation antidepressants, will be used to inform clinical decisions in a Clinical Trial for which we finalized the design and we are now seeking regulatory approval.
One challenge in designing any prediction algorithm is in selecting the right combination of features that will predict a well-defined clinical outcome. The PREDICTIX algorithm and pipeline perfectly tackles this challenge and act as a decision-supporting tool for clinicians, generating precise, efficient, quick, and cost-effective treatment for depression.
PREDICTIX workflow