ADHD-AID: Aiding Tool for Detecting Children’s Attention Deficit Hyperactivity Disorder via EEG-Based Multi-Resolution Analysis and Feature Selection

Biomimetics, Journal Year: 2024, Volume and Issue: 9(3), P. 188 - 188

Published: March 20, 2024

The severe effects of attention deficit hyperactivity disorder (ADHD) among adolescents can be prevented by timely identification and prompt therapeutic intervention. Traditional diagnostic techniques are complicated time-consuming because they subjective-based assessments. Machine learning (ML) automate this process prevent the limitations manual evaluation. However, most ML-based models extract few features from a single domain. Furthermore, studies have not examined effective electrode placement on skull, which affects process, while others employed feature selection approaches to reduce space dimension consequently complexity training models. This study presents an tool for automatically identifying ADHD entitled "ADHD-AID". present uses several multi-resolution analysis including variational mode decomposition, discrete wavelet transform, empirical decomposition. ADHD-AID extracts thirty time time-frequency domains identify ADHD, nonlinear features, band-power entropy-based statistical features. also looks at best EEG detecting ADHD. Additionally, it into location combinations that significant impact accuracy. variety methods choose those greatest influence diagnosis reducing classification's time. results show has provided scores accuracy, sensitivity, specificity, F1-score, Mathew correlation coefficients 0.991, 0.989, 0.992, 0.982, respectively, in with 10-fold cross-validation. Also, area under curve reached 0.9958. ADHD-AID's significantly higher than all earlier detection adolescents. These notable trustworthy findings support use such automated as means assistance doctors youngsters.

Language: Английский

---

A machine learning based depression screening framework using temporal domain features of the electroencephalography signals

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(3), P. e0299127 - e0299127

Published: March 27, 2024

Depression is a serious mental health disorder affecting millions of individuals worldwide. Timely and precise recognition depression vital for appropriate mediation effective treatment. Electroencephalography (EEG) has surfaced as promising tool inspecting the neural correlates therefore, potential to contribute diagnosis effectively. This study presents an EEG-based depressive detection mechanism using publicly available EEG dataset called Multi-modal Open Dataset Mental-disorder Analysis (MODMA). uses data acquired from 55 participants 3 electrodes in resting-state condition. Twelve temporal domain features are extracted by creating non-overlapping window 10 seconds, which presented novel feature selection mechanism. The algorithm selects optimum chunk attributes with highest discriminative power classify disorders patients healthy controls. selected classified three different classification algorithms i.e., Best- First (BF) Tree, k-nearest neighbor (KNN), AdaBoost. accuracy 96.36% achieved BF-Tree vector length 12. proposed scheme outperforms existing state-of-the-art schemes terms number used recording, length, accuracy. framework could be psychiatric settings, providing valuable support psychiatrists.

Language: Английский

---

A BCI System Based on Motor Imagery for Assisting People with Motor Deficiencies in the Limbs

Brain Sciences, Journal Year: 2020, Volume and Issue: 10(11), P. 864 - 864

Published: Nov. 17, 2020

Motor deficiencies constitute a significant problem affecting millions of people worldwide. Such suffer from debility in daily functioning, which may lead to decreased and incoherence routines deteriorate their quality life (QoL). Thus, there is an essential need for assistive systems help those achieve actions enhance overall QoL. This study proposes novel brain-computer interface (BCI) system assisting with limb motor disabilities performing activities by using brain signals control devices. The extraction useful features vital efficient BCI system. Therefore, the proposed consists hybrid feature set that feeds into three machine-learning (ML) classifiers classify Imagery (MI) tasks. selection (FS) practical, real-time, low computation cost. We investigate different combinations channels select combination has highest impact on performance. results indicate achieved accuracies support vector machine (SVM) classifier are 93.46% 86.0% competition III-IVa dataset autocalibration recurrent adaptation dataset, respectively. These datasets used test performance BCI. Also, we verify effectiveness comparing its recent studies. show accurate efficient. Future work can apply individuals assist them capability improve Moreover, forthcoming examine system's controlling devices such as wheelchairs or artificial limbs.

Language: Английский

---

CoMB-Deep: Composite Deep Learning-Based Pipeline for Classifying Childhood Medulloblastoma and Its Classes

Frontiers in Neuroinformatics, Journal Year: 2021, Volume and Issue: 15

Published: May 28, 2021

Childhood medulloblastoma (MB) is a threatening malignant tumor affecting children all over the globe. It believed to be foremost common pediatric brain causing death. Early and accurate classification of childhood MB its classes are great importance help doctors choose suitable treatment observation plan, avoid progression, lower death rates. The current gold standard for diagnosing histopathology biopsy samples. However, manual analysis such images complicated, costly, time-consuming, highly dependent on expertise skills pathologists, which might cause inaccurate results. This study aims introduce reliable computer-assisted pipeline called CoMB-Deep automatically classify with high accuracy from histopathological images. key challenge lack datasets, especially four categories (defined by WHO) inadequate related studies. All relevant works were based either deep learning (DL) or textural feature extractions. Also, studies employed distinct features accomplish procedure. Besides, most them only extracted spatial features. Nevertheless, blends advantages extraction techniques DL approaches. consists composite techniques. Initially, it extracts 10 convolutional neural networks (CNNs). then performs fusion step using discrete wavelet transform (DWT), texture method capable reducing dimension fused Next, explores best combination features, enhancing performance process two search strategies. Afterward, employs selection sets selected in previous step. A bi-directional long-short term memory (Bi-LSTM) network; DL-based approach that utilized phase. maintains categories: binary category distinguishing between abnormal normal cases multi-class identify subclasses MB. results both prove reliable. also indicate strategies have enhanced Bi-LSTM compared individual verify competitiveness, this comparison confirmed robustness outperformance. Hence, can pathologists perform diagnoses, reduce misdiagnosis risks could occur diagnosis, accelerate procedure, decrease diagnosis costs.

Language: Английский

---

GASTRO-CADx: a three stages framework for diagnosing gastrointestinal diseases

PeerJ Computer Science, Journal Year: 2021, Volume and Issue: 7, P. e423 - e423

Published: March 10, 2021

Gastrointestinal (GI) diseases are common illnesses that affect the GI tract. Diagnosing these is quite expensive, complicated, and challenging. A computer-aided diagnosis (CADx) system based on deep learning (DL) techniques could considerably lower examination cost processes increase speed quality of diagnosis. Therefore, this article proposes a CADx called Gastro-CADx to classify several using DL techniques. involves three progressive stages. Initially, four different CNNs used as feature extractors extract spatial features. Most related work approaches extracted features only. However, in following phase Gastro-CADx, first stage applied discrete wavelet transform (DWT) cosine (DCT). DCT DWT temporal-frequency spatial-frequency Additionally, reduction procedure performed stage. Finally, third combinations fused concatenated manner inspect effect combination output results select best-fused set. Two datasets referred Dataset I II utilized evaluate performance Gastro-CADx. Results indicated has achieved an accuracy 97.3% 99.7% for respectively. The were compared with recent works. comparison showed proposed approach capable classifying higher other work. Thus, it can be reduce medical complications, death-rates, addition treatment. It also help gastroenterologists producing more accurate while lowering inspection time.

Language: Английский

---