Depression is a significant risk factor for other serious health conditions, such as heart failure, dementia, and diabetes. In this study, a quantitative method was developed to detect depressive states in individuals using electrocardiogram (ECG) and photoplethysmogram (PPG) signals. Data were obtained from 59 people affiliated with the high-specialized medical center of Bajio T1, which consists of medical professionals, administrative personnel, and service workers. Data were analyzed using the Beck Depression Inventory (BDI-II) to discern potential false positives. The statistical analyses performed elucidated distinctive features with variable behavior in response to diverse physical stimuli, which were adeptly processed through a machine learning classification framework. The method achieved an accuracy rate of up to 92% in the identification of depressive states, substantiating the potential of biophysical data in increasing the diagnostic process of depression. The results suggest that this method is innovative and has significant potential. With additional refinements, this approach could be utilized as a screening tool in psychiatry, incorporated into everyday devices for preventive diagnostics, and potentially lead to alarm systems for individuals with suicidal thoughts.

The creation of a system for depression detection is proposed, based on the acquisition and processing of ECG and PPG signals, followed by the development of an algorithm for pattern classification and detection. The main goal is to achieve an accurate classification of an individual into a depression or non-depression group, ultimately achieving the correct detection of the mentioned problem. This was accomplished through the analysis of a set of physiological variables taken from PPG and ECG signals, using an experimental protocol called script-driven imagery adapted to the current paradigm, applied to individuals from the High Specialized Medical Center of Bajio T1 in Leon, Mexico. The variables to be used include heart rate variability, pulse transit time, heart rate, respiration signal, physiological coherence of each, and the global coherence index, as well as other frequency-related variables selected based on a literature review. Subsequently, a Python program was developed for processing and obtaining the values of the mentioned variables for later use in a machine learning code. A multi machine learning model test was carried just to find out that the binary classification algorithm that yielded the best performance was a Random Forest, with a sensitivity and accuracy of 76% for the validation group, although higher percentages were achieved with smaller groups of individuals. A review of the performance of the best features in the algorithm was also conducted to identify which variables can have a greater impact when attempting to detect a depressive state in an individual.