Automated detection of panic disorder based on multimodal physiological signals using machine learning

Abstract

We tested the feasibility of automated discrimination of patients with panic disorder (PD) from healthy controls (HCs) based on multimodal physiological responses using machine learning. Electrocardiogram (ECG), electrodermal activity (EDA), respiration (RESP), and peripheral temperature (PT) of the participants were measured during three experimental phases: rest, stress, and recovery. Eleven physiological features were extracted from each phase and used as input data. Logistic regression (LoR), k‐nearest neighbor (KNN), support vector machine (SVM), random forest (RF), and multilayer perceptron (MLP) algorithms were implemented with nested cross‐validation. Linear regression analysis showed that ECG and PT features obtained in the stress and recovery phases were significant predictors of PD. We achieved the highest accuracy (75.61%) with MLP using all 33 features. With the exception of MLP, applying the significant predictors led to a higher accuracy than using 24 ECG features. These results suggest that combining multimodal physiological signals measured during various states of autonomic arousal has the potential to differentiate patients with PD from HCs.

A Comprehensive Survey of ArtificialIntelligence in Precision Healthcare:Shedding Light on Interpretability

“…It is an efficient way to understand the activations of the brain network [41]. Common types of neuroimaging methods used to track human brain activities are, fMRI which uses BOLD signals for imaging, which detects the difference in a magnetic moment between oxygenated hemoglobin and deoxygenated hemoglobin gives high temporal resolution that is ability to capture changes of brain activation with time [39] ,CT scan is a collection of X-ray images that have been converted into cross-sectional images of the brain and PET scan uses glucose in the bloodstream and acts as a radioactive tracer and EEG Electrical activities of brain are captured using electrodes fixed on the scalp [42,38]. Data can be captured by any of neuroimaging techniques with the help of audio stimuli, visual stimuli or by giving sentences to read.…”

Exploring Emotion Analysis Using Artificial Intelligence, Geospatial Information Systems, and Extended Reality for Urban Services

“…Jang et al [98] demonstrated the potential of multimodal emotion detection and recognition using physiological signals in medical applications. Their study analyzed the ECG, EDA, respiration (RESP), and peripheral temperature (PT) data during rest, stress, and recovery using machine learning methods such as logistic regression (LoR), KNN, SVM, RF, and MLP to detect patients with panic disorder (PD).…”

“…Their study analyzed the ECG, EDA, respiration (RESP), and peripheral temperature (PT) data during rest, stress, and recovery using machine learning methods such as logistic regression (LoR), KNN, SVM, RF, and MLP to detect patients with panic disorder (PD). This automated system had the potential to enhance the diagnostic accuracy and minimize human error, leading to improved patient outcomes and lower healthcare costs [98]. Additionally, by leveraging multiple modalities of emotional expression, AI systems could potentially achieve more accurate and contextsensitive emotion recognition across different applications and cultures-such as medical and customer service.…”