Functional Near‐Infrared Spectroscopy‐Based Computer‐Aided Diagnosis of Major Depressive Disorder Using Convolutional Neural Network with a New Channel Embedding Layer Considering Inter‐Hemispheric Asymmetry in Prefrontal Hemodynamic Responses DOI Creative Commons
Kyeonggu Lee, Jinuk Kwon, M. Chun

и другие.

Depression and Anxiety, Год журнала: 2024, Номер 2024(1)

Опубликована: Янв. 1, 2024

Functional near-infrared spectroscopy (fNIRS) is being extensively explored as a potential primary screening tool for major depressive disorder (MDD) because of its portability, cost-effectiveness, and low susceptibility to motion artifacts. However, the fNIRS-based computer-aided diagnosis (CAD) MDD using deep learning methods has rarely been studied. In this study, we propose novel framework based on convolutional neural network (CNN) CAD with high accuracy. The fNIRS data participants-48 patients 68 healthy controls (HCs)-were obtained while they performed Stroop task. hemodynamic responses calculated from preprocessed were used inputs proposed CNN model an ensemble architecture, comprising three 1D depth-wise layers specifically designed reflect interhemispheric asymmetry in between HCs, which known be distinct characteristic previous studies. performance was evaluated leave-one-subject-out cross-validation strategy compared those conventional machine models. exhibited accuracy, sensitivity, specificity 84.48%, 83.33%, 85.29%, respectively. accuracies algorithms-shrinkage linear discriminator analysis, regularized support vector machine, EEGNet, ShallowConvNet-were 73.28%, 74.14%, 62.93%, 62.07%, conclusion, can differentiate HCs more accurately than models, demonstrating applicability systems.

Язык: Английский

Application of functional near-infrared spectroscopy and machine learning to predict treatment response after six months in major depressive disorder DOI Creative Commons
Cyrus S. H. Ho, Jin-Yuan Wang, Gabrielle Wann Nii Tay

и другие.

Translational Psychiatry, Год журнала: 2025, Номер 15(1)

Опубликована: Янв. 11, 2025

Depression treatment responses vary widely among individuals. Identifying objective biomarkers with predictive accuracy for therapeutic outcomes can enhance efficiency and avoid ineffective therapies. This study investigates whether functional near-infrared spectroscopy (fNIRS) clinical assessment information predict response in major depressive disorder (MDD) through machine-learning techniques. Seventy patients MDD were included this 6-month longitudinal study, the primary outcome measured by changes Hamilton Rating Scale (HAM-D) scores. fNIRS strictly evaluated using nested cross-validation to responders non-responders based on models, including support vector machine, random forest, XGBoost, discriminant analysis, Naïve Bayes, transformers. The task change of total haemoglobin (HbT), defined as difference between pre-task post-task average HbT concentrations, dorsolateral prefrontal cortex (dlPFC) is significantly correlated (p < 0.005). Leveraging a Bayes model, inner performance (bAcc = 70% [SD 4], AUC 0.77 0.04]) outer results 73% 3], 0.02]) yielded predicting solely data. bimodal model combining data showed inferior 68%, 0.70) compared fNIRS-only model. Collectively, holds potential scalable neuroimaging modality MDD.

Язык: Английский

Процитировано

1
Deep learning approach to predict developmental outcomes of non-suicidal self-injury: An ERP study DOI Creative Commons

Fei Yin,

Feng Si, Wenlong Jiang

и другие.

Research Square (Research Square), Год журнала: 2025, Номер unknown

Опубликована: Янв. 10, 2025

Abstract Background Identifying predictors of developmental outcomes in non-suicidal self-injury (NSSI) is crucial and goes beyond tracking its progression. EEG technology notable for consistent objective neurophysiological recordings NSSI detection. Using ERP components deep learning models predicting these still underexplored. Methods Twenty-six the remission group (RG), twenty-nine aggravation (AG), twenty-seven healthy (HG) completed affective Stroop task with EEG. N2 P3 component differences were analyzed across groups, EEGNet model was used to assess outcomes. Result A significant interaction observed between emotion on (F (2, 79) = 16.934, p < 0.001, η2 0.300). Under neutral stimuli, smallest HG, larger RG, largest AG, while negative HG smaller than RG AG. effect noted 79) = 7.607, η2 = 0.161), exhibiting compared The under stimuli achieved highest classification accuracy (94.31%). Conclusion findings indicate that linked cognitive processing deficits, including impaired control resource allocation stimuli. Additionally, amplitudes shown reliably predict NSSI.

Язык: Английский

Процитировано

0
A free association semantic task for fNIRS-based perinatal depression assessment DOI Creative Commons
Danni Chen,

Xuanjin Yang,

Yuanyuan Liang

и другие.

Frontiers in Neurology, Год журнала: 2025, Номер 15

Опубликована: Янв. 15, 2025

Perinatal depression (PD) is a highly prevalent psychological disorder that has detrimental effect on infant and maternal physical mental health, but effective objective assessment of PD still insufficient. In recent years, the functional near-infrared spectroscopy (fNIRS) been acknowledged as an non-invasive tool for clinical depression. This study proposed free association semantic task (FAST) paradigm fNIRS-based PD. To better address emotion characteristics PD, participants are required to generate dynamic concept chain based positive, negative or neutral seed words, while 48-channel fNIRS recordings over frontal bilateral temporal regions. Results from twenty-two late-pregnant women revealed that, oxyhemoglobin (oxy-Hb) changes during FAST with positive words region were correlated severity, which was different correlation patterns in word classical verbal fluency test (VFT). Furthermore, distinct also observed manifested channels corresponding right dorsolateral prefrontal cortex (DLPFC) inferior gyrus (IFG), respectively. Moreover, regression analyses showed can well explain severity Our findings suggest promising approach assessment.

Язык: Английский

Процитировано

0
Accurate identification of anxiety and depression based on the dlPFC in an emotional autobiographical memory task: A machine learning approach DOI
Guixiang Wang, Yusen Huang, Yan Zhang

и другие.

Biomedical Signal Processing and Control, Год журнала: 2025, Номер 104, С. 107503 - 107503

Опубликована: Янв. 18, 2025

Язык: Английский

Процитировано

0
Identification of Subthreshold Depression Based on fNIRS–VFT Functional Connectivity: A Machine Learning Approach DOI Creative Commons
Lin Li,

Jingxuan Liu,

Yifan Zheng

и другие.

Depression and Anxiety, Год журнала: 2025, Номер 2025(1)

Опубликована: Янв. 1, 2025

Background: Subthreshold depression (SD) is regarded as a prodromal stage and substantial risk factor for major depressive disorder (MDD). The timely identification of SD critical clinical significance. This study aimed to develop machine learning (ML) classification model the individuals with using functional near‐infrared spectroscopic imaging (fNIRS) verbal fluency task (VFT). Methods: recruited total 70 participants matched 73 healthy controls (HCs) differentiate between two groups based on connectivity (FC) features during fNIRS–VFT, an interpretable random forest (RF) model. Results: RF demonstrated area under curve (AUC) 0.77, accuracy (ACC) 75.86%, sensitivity 75.00%, specificity 76.00% F1 score 0.75 identifying SD. highest‐ranked FC features, in terms importance, were identified Channel (CH) 26 (the right frontal eye fields (FEFs)) CH 30 FEF), 3 left premotor supplementary motor cortex (PMC‐and‐SMA)) 42 PMC‐and‐SMA), well FEF) 32 primary somatosensory (PSC)). Conclusion: has capacity effectively classify efficacy abnormal particularly FEF, bilateral PSC PMC‐and‐SMA. findings this have provided foundation large‐scale screening populations, offering promising opportunities early diagnosis prevention MDD.

Язык: Английский

Процитировано

0
Robust semi-supervised extraction of information using functional near-infrared spectroscopy for diagnosing depression DOI

Shi Qiao,

Jitao Zhong, Lu Zhang

и другие.

Biomedical Signal Processing and Control, Год журнала: 2025, Номер 105, С. 107571 - 107571

Опубликована: Фев. 8, 2025

Язык: Английский

Процитировано

0
Cortical activation patterns in generalized anxiety and major depressive disorders measured by multi-channel near-infrared spectroscopy DOI

Anfeirea Jialin,

Zhang Hongguang, Xiaohui Wang

и другие.

Journal of Affective Disorders, Год журнала: 2025, Номер unknown

Опубликована: Март 1, 2025

Язык: Английский

Процитировано

0
Data collection, enhancement, and classification of functional near-infrared spectroscopy motor execution and imagery DOI

Baiwei Sun,

Zhang Xiu, Xin Zhang

и другие.

Review of Scientific Instruments, Год журнала: 2025, Номер 96(3)

Опубликована: Март 1, 2025

Recognition and execution of motor imagery play a key role in brain–computer interface (BCI) are prerequisites for converting thoughts into executable instructions. However, to date, data acquired through commonly used electroencephalography (EEG) methods very sensitive motion interference, which will affect the accuracy classification. The emerging functional near-infrared spectroscopy (fNIRS) technique, while overcoming drawbacks EEG’s susceptibility interference difficulty detecting signals, has less publicly available data. In this paper, we designed experiment based on wearable fNIRS device acquire brain signals proposed modified Kolmogorov–Arnold network (named SE-KAN) recognizing corresponding task. Due small number subjects experiment, Wasserstein generative adversarial was enhance processing. For recognition task, SE-KAN method achieved 96.36 ± 2.43% single-subject 84.72 3.27% cross-subject accuracy. It is believed that dataset paper help development BCI.

Язык: Английский

Процитировано

0
Identification of Potential Biomarkers for Major Depressive Disorder: Based on Integrated Bioinformatics and Clinical Validation DOI
Xiaogang Zhong, Yue Chen, Weiyi Chen

и другие.

Molecular Neurobiology, Год журнала: 2024, Номер 61(12), С. 10355 - 10364

Опубликована: Май 9, 2024

Язык: Английский

Процитировано

2
AI in Neurodegeneration Prediction DOI

Neelima Priyanka Nutulapati,

Naresh Babu Karunakaran,

V. Banupriya

и другие.

Advances in medical technologies and clinical practice book series, Год журнала: 2024, Номер unknown, С. 114 - 130

Опубликована: Июнь 28, 2024

This chapter explores the capability of artificial intelligence (AI) in predicting development neurodegenerative sicknesses, particular focusing on Alzheimer's ailment. The goal is to recognize cutting-edge nation AI studies this area and identify rising superior procedures. Through conducting a complete literature evaluation reading existing research, authors spotlight strengths barriers use for neurodegeneration prediction. Similarly, they discuss role huge information, system mastering, deep mastering strategies developing accurate reliable prediction models. These findings endorse that has capacity seriously enhance early diagnosis disease progression. We conclude with ability future instructions demanding situations unexpectedly increasing vicinity

Язык: Английский

Процитировано

1