
The science behind DeepPsy
At DeepPsy, we aim to provide solutions for patients suffering from neuropsychiatric disorders at the highest level of available scientific evidence. While there exist several markers for diagnostic approaches or prediction of treatment outcomes for several disorders based on electrophysiological time series such as the electroencepohalogram (EEG) and electrocardiogram (ECG), none of these findings has made its way into the clinical routine.
At DeepPSy, we are totally convinced that this is not necessarily due to insufficient markers or results. The transformation from conservative diagnostic processes and predictive approaches for determining the best possible treatment in psychiatry has just started. It takes some time until the traditional physician and therapist based decision models will be overcome in favor for a stratified management approach. Based on objective physiological patterns, a better and more reliable way to find the best treatments in psychiatry is poossible, just as it is the case in many other medical fields. One could not imagine to seek help from a surgon when being severly ill andthen relying on the personal opinion of the therapist withoput further diagnostics such as blood samples, X-ray, ultrasound, computertomography or magnetic resonance imaging. However, in psychiatry the patients are deemed to trust the personal experience of their therapist without any diagnostic approaches that could inform the further decisions in the management of the disorder.
This is where DeepPsy starts. Based on the personal >15 years of clinical and research experience, DeepPsy provides the lates anbd most reliable reserach results at your hand to improve the way how we treat mental disorders. Below is a list with some of the most important findings in the neurophysiologial research field that guides the algorithms used at DeepPsy. Special attention is paid to the inclusion of replicated findings. This means, DeepPSy will weight the different results of studies that have been publsihed in peer-reviewed scientiic journals. Findings that not only have been reported from a single laboratory but have been confirmed by a different research group and possibly with a different dataset count the most for our algorithms. This guarantees that the algorithms used by DeepPsy to make predictions on the outcome of different treatment approaches are as reliable as possible at that stage. However, we are totally aware that neurophysiologial research in mental disorders is a fast moving field and more biomarkers and more prediction models will appear and possibly find the way into the DeepPsy algorithms. Further, DeepPsy is on its way to even overcome this man-chosen biomarker approach and intends to apply Deep Learning models to the data to achieve higher accuracies in the predictive models.
1. Biomarkers derived from the electroencephalogram (EEG)
2. Biomarkers dereived from the elecgtrocardiogram (ECG)