What is the clinical evidence from this study?

Study design: Observational. Population: Depression (n=140). Intervention: fNIRS-based machine learning model (SVM with temporal and correlation fusion features). Primary outcome: Classification accuracy for distinguishing mild from severe depression (AUC 0.963).

May 4, 2023Open Access

Functional Near-Infrared Spectroscopy-based Diagnosis Support System for Distinguishing between Mild and Severe Depression Using Machine Learning Approaches

ZHZhiyong HuangNational University of Singapore MLMan LiuChongqing University of Posts and Telecommunications HYHui YangNanjing University of Chinese Medicine

Key Result

A machine learning model using temporal and correlation fusion features from fNIRS data achieved 92.8% accuracy and an AUC of 0.963 in distinguishing between mild and severe depression.

Study Design

Type

Observational (n=140)

Multicenter

Structured PICO

Can fNIRS-based machine learning models accurately distinguish between mild and severe depression?

Population

140 patients with mild or severe depression, free of other neurological or cardiovascular disorders, underwent fNIRS during a verbal fluency task to develop a diagnostic prediction model.

Exposure

Functional near-infrared spectroscopy (fNIRS)-based prediction models using a fusion of temporal and correlation features extracted via CEEMDAN-WPT during a verbal fluency task

Outcome

Classification accuracy in distinguishing severe from mild depression

An fNIRS-based machine learning model using temporal and correlation features can effectively distinguish between mild and severe depression with high accuracy.

Main Result

Effect estimate: AUC 0.963

Limitations

Small sample size of 140 subjects may have limited the precision of the models.
Diagnostic data were obtained from a single psychiatric research center, limiting generalizability.
Patients with moderate depression were not included in the study due to their small sample size.

Abstract

Abstract Due to the fast-paced lifestyle, individuals may experience varying degrees of depression attributed to stress, academic pursuits, and other causative factors. Hemodynamics and functional connectivity (correlation between channels) of the prefrontal lobe have been identified as crucial factors in assessing the severity of depression. As a non-invasive technique for monitoring cerebral blood flow, functional near-infrared spectroscopy (fNIRS) shows promising potential as a tool for objective auxiliary diagnosis of depression.This study aimed to develop prediction models for distinguishing patients with severe depression from those with mild depression based on the dataset collected by fNIRS.We collected the fNIRS data from 140 subjects, and used a complete ensemble empirical mode decomposition with adaptive noise-wavelet threshold combined denoising method (CEEMDAN-WPT) to remove the jitter and artefact noise generated during the verbal fluency task (VFT). The temporal and correlation features of 18 channels in the prefrontal lobe of the subjects were extracted as predictors. We screened out the optimal temporal features (TF) or correlation features (CF) using the RFECV, and investigated their role in distinguishing severe and mild depression, respectively. The fusion of TF and CF, as the input of the prediction model, yielded higher classification accuracy than using TF or CF alone as the prediction factor. Among the prediction models, the SVM-based predictive model performed well in nested cross-validation, with an accuracy rate of 92.8%.The proposed model effectively distinguishes mild depression from severe depression, provides an objective diagnostic method for mental health workers, and is significant in treating patients with depression.

اسأل الذكاء الاصطناعي

Bookmark

View Full Paper