Depression Recognition Using Daily Wearable-Derived Physiological Data

Sensors, Journal Year: 2025, Volume and Issue: 25(2), P. 567 - 567

Published: Jan. 19, 2025

The objective identification of depression using physiological data has emerged as a significant research focus within the field psychiatry. advancement wearable measurement devices opened new avenues for individuals with in everyday-life contexts. Compared to other methods, wearables offer potential continuous, unobtrusive monitoring, which can capture subtle changes indicative depressive states. present study leverages multimodal wristband collect from fifty-eight participants clinically diagnosed during their normal daytime activities over six hours. Data collected include pulse wave, skin conductance, and triaxial acceleration. For comparison, we also utilized matched healthy controls publicly available dataset, same equivalent durations. Our aim was identify through analysis measurements derived daily life scenarios. We extracted static features such mean, variance, skewness, kurtosis indicators like heart rate, acceleration, well autoregressive coefficients these signals reflecting temporal dynamics. Utilizing Random Forest algorithm, distinguished non-depressive varying classification accuracies on aggregated 6 h, 2 30 min, 5 min segments, 90.0%, 84.7%, 80.1%, 76.0%, respectively. results demonstrate feasibility wearable-derived recognition. achieved suggest that this approach could be integrated into clinical settings early detection monitoring symptoms. Future work will explore methods personalized interventions real-time offering promising avenue enhancing mental health care integration technology.

Language: Английский

---

A combination of deep learning models and type-2 fuzzy for EEG motor imagery classification through spatiotemporal-frequency features

Journal of Medical Engineering & Technology, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 14

Published: Feb. 14, 2025

Developing a robust and effective technique is crucial for interpreting user's brainwave signals accurately in the realm of biomedical signal processing. The variability uncertainty present EEG patterns over time, compounded by noise, pose notable challenges, particularly mental tasks like motor imagery. Introducing fuzzy components can enhance system's ability to withstand noisy environments. emergence deep learning has significantly impacted artificial intelligence data analysis, prompting extensive exploration into assessing understanding brain signals. This work introduces hybrid series architecture called FCLNET, which combines Compact-CNN extract frequency spatial features alongside LSTM network temporal feature extraction. activation functions CNN were implemented using type-2 tackle uncertainties. Hyperparameters FCLNET model are tuned Bayesian optimisation algorithm. efficacy this approach assessed through BCI Competition IV-2a database IV-1 database. By incorporating employing tuning, proposed indicates good classification accuracy compared literature. Outcomes showcase exceptional achievements model, suggesting that integrating units other classifiers could lead advancements imagery-based systems.

Language: Английский

---

Transformer-Driven Affective State Recognition from Wearable Physiological Data in Everyday Contexts

Sensors, Journal Year: 2025, Volume and Issue: 25(3), P. 761 - 761

Published: Jan. 27, 2025

The rapid advancement in wearable physiological measurement technology recent years has brought affective computing closer to everyday life scenarios. Recognizing states daily contexts holds significant potential for applications human–computer interaction and psychiatry. Addressing the challenge of long-term, multi-modal data settings, this study introduces a Transformer-based algorithm state recognition, designed fully exploit temporal characteristics signals interrelationships between different modalities. Utilizing DAPPER dataset, which comprises continuous 5-day wrist-worn recordings heart rate, skin conductance, tri-axial acceleration from 88 subjects, our model achieved an average binary classification accuracy 71.5% self-reported positive or negative sampled at random moments during collection, 60.29% 61.55% five-class based on valence arousal scores. results demonstrate feasibility applying recognition contexts.

Language: Английский

---

Speech-based emotion recognition using a hybrid RNN-CNN network

Signal Image and Video Processing, Journal Year: 2024, Volume and Issue: 19(2)

Published: Dec. 12, 2024

Language: Английский

---

EEG classification of resting state and arithmetic cognitive workload using functional connectivity of different frequency bands and machine learning techniques

Smart Science, Journal Year: 2024, Volume and Issue: unknown, P. 1 - 12

Published: Aug. 18, 2024

This study aimed to perform a comparative of the functional connectivity different frequency bands for identification resting and arithmetic cognitive workload EEG using machine learning techniques. Functional was calculated from preprocessed EEGs both rest task states in 5 sub-bands: alpha (8–13 Hz), theta (4–8 delta (1–4 gamma (30–45 beta (13–30 Hz). done through Weighted Phase Lag Index (WPLI). After that, PCA applied feature vectors decrease dimensionality space. Eventually, normalized chosen features were used as input learning-based classification models, performance assessed leave-one-subject cross-validation (LOSOCV) algorithm. Experimental results showed that on basis delta, theta, alpha, beta, 90.27%, 77.78%, 62.50%, 76.39%, respectively. The obtained models technique are successfully detect mental rest- task-EEG. In summary, is potent tool comprehending neural has significant applications various fields.

Language: Английский

---

Automated recognition of mental cognitive workload through nonlinear EEG analysis

Web Intelligence, Journal Year: 2024, Volume and Issue: unknown, P. 1 - 18

Published: Sept. 30, 2024

Nowadays, with the remarkable advancements in detection instruments and artificial intelligence, there has been extensive utilization of human mental state monitoring various domains. Few studies have explored how nonlinear analysis methods can detect cognitive workload despite complex nature EEG signals signal processing techniques. In addition, fuzziness conditions makes need to use fuzzy engineering tools tangible this field. Therefore, investigation aimed develop a decision support algorithm improve previous efforts for classification task resting through machine learning algorithms. Various features were calculated from all 19 channels: Hurst exponent, Lempel–Ziv complexity, detrended fluctuation analysis, Higuchi fractal dimension, Katz permutation entropy, singular value decomposition Petrosian sample Lyapunov exponent. During step, newly developed EPC-FC (Expert per Class Fuzzy Classifier) is introduced, utilizing an ensemble framework specialized sub-classifiers identifying particular condition. By training negative correlation (NCL) approach, designed be exceptionally adaptable. Additionally, separation within each class provides versatility clarity system’s design. The proposed approach based on systems analyses was applied data recognition, which excellent accuracy 98.50% F1-score 98.56% much higher than findings Also, obtained results indicate that classifier maintains consistently high exceeding 90% across levels SNRs. proved potential states brain, consistent data. Other approaches should considered future current results.

Language: Английский

---