Using retinal electrophysiology toward precision psychiatry

Abstract

Precision medicine in psychiatry is based on the identification of homogeneous subgroups of patients with the help of biosignatures-sets of biomarkers-in order to enhance diagnosis, stratification of patients, prognosis, evaluation, and prediction of treatment response. Within the broad domain of biomarker discovery, we propose retinal electrophysiology as a tool for identification of biosignatures. The retina is a window to the brain and provides an indirect access to brain functioning in psychiatric disorders. The retina is organized in layers of specialized neurons which share similar functional properties with brain neurons. The functioning of these neurons can be evaluated by electrophysiological techniques named electroretinogram (ERG). Since the study of retinal functioning gives a unique opportunity to have an indirect access to brain neurons, retinal dysfunctions observed in psychiatric disorders inform on brain abnormalities. Up to now, retinal dysfunctions observed in psychiatric disorders provide indicators for diagnosis, identification of subgroups of patients, prognosis, evaluation, and prediction of treatment response. The use of signal processing and machine learning applied on ERG data enhances retinal markers extraction, thus providing robust, reproducible, and reliable retinal electrophysiological markers to identify biosignatures in precision psychiatry. We propose that retinal electrophysiology may be considered as a new approach in the domain of electrophysiology and could now be added to the routine evaluations in psychiatric disorders. Retinal electrophysiology may provide, in combination with other approaches and techniques, sets of biomarkers to produce biosignatures in mental health.

Keywords: Biomarkers; electrophysiology; precision medicine; retina.

Figure 1.

Retinal electrophysiology added in sets of biomarkers to identify biosignatures in mental health. Retinal electrophysiology as measured with ERG and collected as routine measures for patients. Signal processing and machine learning techniques applied on ERG signal. ERG integrated in sets of biomarkers in combination with other approaches and domains to identify biosignatures in mental health. Abbreviations: ECG, electrocardiography; EEG, electroencephalography; ERG, electroretinogram.