Progress in Objective Detection of Depression and Online Monitoring of Patients Based on Physiological Complexity

There is a well-established connection between depression and increased physiological complexity that can be derived from a person's electroencephalogram (EEG), but also from other electrophysiological signals. Our research showed that it is possible to differentiate between healthy controls and patients with MDD (based on resting-state EEG), as well as to detect the difference between the states of episode and remission in patients during the treatment (which is important information for the therapist). Based on complexity measures, it is also possible to predict the responders to certain therapy, be it a medication or a type of electromagnetic stimulation. Various combinations of different features (both conventional and fractal & nonlinear) extracted from EEG recorded from depressive patients with machine learning models (which can yield highly accurate classification in comparison with healthy controls) are possible. The last decade of research showed how the combination of electrophysiological biomarkers and machine learning can bring useful additional information to the clinicians working with patients diagnosed with depression, but also what we need to change and further test in our applications and methodology. This paper is a brief commentary and cross-section analysis of what proved to be successful, what is possible, and what needs to be changed, to advance Telehealth & IoT detection as well as monitoring of outpatients with depression.The advent of artificial intelligence (AI) and machine learning (ML) in particular, in medicine, holds many promises. Although the acceptance of any innovation in medicine is chronically slow, psychiatry showed to be especially conservative in this regard (Shorter and Fink, 2010) A recent publication demonstrated that ML's purely reliance on patient's medical history, medication, epidemiological data, and scales/questionaries data (Coley et al., 2021) are simply not capable of providing practically useful results. We also explored the possibilities of this methodology in forecasting mania in bipolar depression (Llamoca et al., 2018, 2019, 2021 and 2021a). Whelan, Garavan, Gillan, and their colleagues, explained in their publications before 2017, why computational psychiatry projects, even when relying on neuroimaging data are flawed (Gillan and Whelan, 2017; Gilland and Daw, 2016), arguing that neuroscientists who are mainly practicing it, are lacking the understanding of basic postulates of ML. The consequence is overly optimistic (and misleading) results, that are not leading to clinically useful applications (see also Ng, 1997and Tibshirani et al., 1996.We argue here that the central thing that can lead to the resolution of this frustration, is an addition of electrophysiological signals analysis, and appropriate characterization of it which yields highly accurate results in detection and prediction of any ML model used (Čukić et al., 2020a). We showed that in this way, it is possible to discern between episode and remission The key concept to understand is the so-called physiological complexity (or complex variability in physiology), an analytical approach to electrophysiological signal analysis stemming from electrical engineering, statistical physics and complex systems dynamics A more familiar name for this approach is fractal and nonlinear analysis (Goldberger et al., , 2001). Despite the fact that many medical professionals are labeling this approach 'novel' it is not novel by any standard; seminal work by Mandelbrot from '70s, Pincus, Hausdorf, Peng, and Goldeberger from 80's and 90's last century made that possible. They all buit on early mathematics work of Cantor (Cantor's set, 1893), Peano (1890), Sierpinski (Triangle,1907), Koch (Snowflake, 1909), Lucia (Lucia's set, 1917) and others who could not generalize their findings before the advent of modern computers.If a researcher in any medical field wants to explore the effect of a certain factor, the most probable way to do it is to calculate the means, standard deviations, p values, and other measures coming from frequentist statistics that dominates the field. Irregularity statistics, like any entropy-based measure, for example, quantify the changes in physiological systems in a much more accurate and practically meaningful way (Pincus, 2003). These two approaches (standard or conventional vs nonlinear) are simply measuring different information contained in the data. But knowing that human physiology is not linear and has many fractal dependencies in its control mechanisms, a better approach to analyzing signals from such a complex system dynamics would be nonlinear analysis (Klonowski, 2002 and. Wouldn't it be logical to apply analytics that is better suited for nonstationary, nonlinear, and noisy signals, than to just focus on how smeared are the data around the means?From the existing literature, coming mainly from engineering and technical background, it is clear that fractal and nonlinear analysis is much better suited for this task (Klonowski,2002;Goldberger et al., 2002).Just to mention some of the facts important in research of depression (and mood disorders in general), to improve the understanding of the above-mentioned research. Cortico-vagal control (CVC) is connected to heart rate variability (HRV), which showed to be a robust marker of depression, anxiety, and several other psychiatric conditions (Rottenberg, 2007;Willner et al., 2013;van der Kolk, 2014). Cortico-vagal control (as well as many structural and functional physiological phenomena) is proven to have fractal nature (Rottenberg 2007;. Heart dynamics also has a fractal structure (Peng, Hausdorf & Goldberger, 2002). There is much research evidence on the connection between ANS and heart dynamics in depression, obtained by using nonlinear analysis (Kemp et al., 2010(Kemp et al., , 2014Jelinek et al., 2011;Kop et al., 2010). There is also evidence that nonlinear measures are much more effective in detecting this relationship with a much larger effect size in publications in the last two decades (Čukić et al., 2021 in review).What we know now, from the analytical perspective and possible application in clinical practice, is that by the mere addition of that layer of (nonlinear) analysis, possible in real-time, one can: detect depression (Čukić et al., 2020), detect the subtypes of depressionmelancholic vs non-melancholic (Kemp et al., 2014), detect comorbidities (Kemp et al., 2011-12?), discern episode and remission phase (Čukić et al., 2019), detect cardiovascular risks early (Khodayari-Rhostamabad et al., 2013), differentiate between unipolar and bipolar depression (Chang et al., 2015) and even detect existent suicidal ideation (Khandoker et al., 2017).With todays' technology that made possible Telehealth &IoT (portable monitoring devices, with medical-grade signal quality), as a reliable way of remote monitoring of outpatients, we can support clinicians with objective additional information that might largely improve the effectiveness of therapy for depression. It might be close to previously envisaged personal medicine, increasing the ability of every clinician to better navigate some diagnostic decisions. Revisiting some not-so-well-known mathematical concepts that can thrive with cloud technology, would pay off in improved psychiatric diagnostics and treatment.Although the citation is coming from the economics, it effectively applies to the adoption of these innovations in psychiatry: The difficulty lies not in the absence of new ideas, but an escape from the old ones (Keynes).