Diagnostic value of structural, functional and effective connectivity in bipolar disorder

Acta Psychiatrica Scandinavica, Journal Year: 2024, Volume and Issue: 151(3), P. 192 - 209

Published: Aug. 13, 2024

The aim of this systematic review is to assess the functional magnetic resonance imaging (fMRI) studies bipolar disorder (BD) patients that characterize differences in terms structural, functional, and effective connectivity between with BD, other psychiatric disorders healthy controls as possible biomarkers for diagnosing using neuroimaging. Following Preferred Reporting Items Systematic Reviews Meta-Analyses (PRISMA), guidelines a search recent (since 2015) original on was conducted PUBMED SCOPUS. A total 60 were included review: 20 structural connectivity, 33 only 7 focused complied inclusion exclusion criteria. Despite great heterogeneity findings, there are several trends emerge. In studies, main abnormalities frontal gyrus, anterior, well posterior cingulate cortex emotion reward-related networks. Cerebellum (vermis) cerebrum found be most common finding BD. Moreover, prefrontal amygdala part rich-club hubs often reported disrupted. findings based alterations salience network, default mode network executive control network. Although more larger sample sizes needed ascertain altered brain diagnostic biomarker, perspective method could used single marker diagnosis future, process adoption accelerated by approaches such semiunsupervised machine learning.

Language: Английский

---

Deep multimodal representations and classification of first-episode psychosis via live face processing

Frontiers in Psychiatry, Journal Year: 2025, Volume and Issue: 16

Published: Feb. 26, 2025

Schizophrenia is a severe psychiatric disorder associated with wide range of cognitive and neurophysiological dysfunctions long-term social difficulties. Early detection expected to reduce the burden disease by initiating early treatment. In this paper, we test hypothesis that integration multiple simultaneous acquisitions neuroimaging, behavioral, clinical information will be better for prediction psychosis than unimodal recordings. We propose novel framework investigate neural underpinnings symptoms (that can develop into age) using multimodal behavioral recordings including functional near-infrared spectroscopy (fNIRS) electroencephalography (EEG), facial features. Our data acquisition paradigm based on live face-toface interaction in order study correlates cognition first-episode (FEP). deep representation learning framework, Neural-PRISM, joint compressed representations combining as well These learned are subsequently used describe, classify, predict severity patients, measured Positive Negative Syndrome Scale (PANSS) Global Assessment Functioning (GAF) scores evaluate impact symptomatology. found incorporating from fNIRS EEG along enhances classification between typical controls FEP individuals (significant improvements 10 − 20%). Additionally, our results suggest geometric topological features such curvatures path signatures embedded trajectories brain activity enable discriminatory characteristics psychosis.

Language: Английский

---

Differences in fractional amplitude of low-frequency fluctuations (fALFF) and cognitive function between untreated major depressive disorder and schizophrenia with depressive mood patients

BMC Psychiatry, Journal Year: 2024, Volume and Issue: 24(1)

Published: April 24, 2024

Abstract Background Distinguishing untreated major depressive disorder without medication (MDD) from schizophrenia with depressed mood (SZDM) poses a clinical challenge. This study aims to investigate differences in fractional amplitude of low-frequency fluctuations (fALFF) and cognition MDD SZDM patients. Methods The included 42 cases, 30 patients, 46 healthy controls (HC). Cognitive assessment utilized the Repeatable Battery for Assessment Neuropsychological Status (RBANS). Resting-state functional magnetic resonance imaging (rs-fMRI) scans were conducted, data processed using fALFF slow-4 slow-5 bands. Results Significant changes observed four brain regions across MDD, SZDM, HC groups both fALFF. Compared group showed increased right gyrus rectus (RGR). Relative HC, exhibited decreased left (LGR) putamen. Changes RGR LGR negatively correlated RBANS scores. No significant correlations found between remaining (slow-4 bands) scores or groups. Conclusions Alterations may serve as potential biomarkers distinguishing providing preliminary insights into neural mechanisms cognitive function schizophrenia.

Language: Английский

---

A Comparative Analysis of Early Fusion Architectures for Multimodal Gas Detection Using Machine Learning Models

Instrumentation Mesure Métrologie, Journal Year: 2024, Volume and Issue: 23(4), P. 297 - 306

Published: Aug. 23, 2024

Single-sensor gas detection models often lack robustness and accuracy, hindering safety security.To enhance the accurate classification performance data from seven sensors along with thermal camera images has been used in this study, to train model.The dataset focuses on four classes: Smoke, Perfume, No Gas Mixture of Smoke Perfume.Data various sources capture different perspectives that trained model, hence, early fusion technique was adopted combine extracted features, for an improved feature space.The sensor undergoes preprocessing normalize remove noise.VGG16 model extract image features.The fused then acted as input machine learning Among tested (SVM, Random Forest Classifier, KNN), achieved best validation accuracy 96.41%, outperforming SVM (94.22%) KNN (94.53%).This approach demonstrates effectiveness multi-sensor enhanced high potentially improving response times reducing false alarms.

Language: Английский

---

Applications of Multi-objective, Multi-label, and Multi-class Classifications

Springer tracts in nature-inspired computing, Journal Year: 2024, Volume and Issue: unknown, P. 135 - 164

Published: Jan. 1, 2024

Language: Английский

---