Epileptic Seizures Detection Using Deep Learning Techniques: A Review

International Journal of Environmental Research and Public Health, Год журнала: 2021, Номер 18(11), С. 5780 - 5780

Опубликована: Май 27, 2021

A variety of screening approaches have been proposed to diagnose epileptic seizures, using electroencephalography (EEG) and magnetic resonance imaging (MRI) modalities. Artificial intelligence encompasses a areas, one its branches is deep learning (DL). Before the rise DL, conventional machine algorithms involving feature extraction were performed. This limited their performance ability those handcrafting features. However, in features classification are entirely automated. The advent these techniques many areas medicine, such as diagnosis has made significant advances. In this study, comprehensive overview works focused on automated seizure detection DL neuroimaging modalities presented. Various methods seizures automatically EEG MRI described. addition, rehabilitation systems developed for analyzed, summary provided. tools include cloud computing hardware required implementation algorithms. important challenges accurate with discussed. advantages limitations employing DL-based Finally, most promising models possible future delineated.

Язык: Английский

---

Deep learning for neuroimaging-based diagnosis and rehabilitation of Autism Spectrum Disorder: A review

Computers in Biology and Medicine, Год журнала: 2021, Номер 139, С. 104949 - 104949

Опубликована: Окт. 29, 2021

Язык: Английский

---

Application of artificial intelligence in wearable devices: Opportunities and challenges

Computer Methods and Programs in Biomedicine, Год журнала: 2021, Номер 213, С. 106541 - 106541

Опубликована: Ноя. 17, 2021

Язык: Английский

---

A comprehensive survey on design and application of autoencoder in deep learning

Applied Soft Computing, Год журнала: 2023, Номер 138, С. 110176 - 110176

Опубликована: Март 8, 2023

Язык: Английский

---

Time series forecasting of new cases and new deaths rate for COVID-19 using deep learning methods

Results in Physics, Год журнала: 2021, Номер 27, С. 104495 - 104495

Опубликована: Июнь 26, 2021

The first known case of Coronavirus disease 2019 (COVID-19) was identified in December 2019. It has spread worldwide, leading to an ongoing pandemic, imposed restrictions and costs many countries. Predicting the number new cases deaths during this period can be a useful step predicting facilities required future. purpose study is predict rate one, three seven-day ahead next 100 days. motivation for every n days (instead just day) investigation possibility computational cost reduction still achieving reasonable performance. Such scenario may encountered real-time forecasting time series. Six different deep learning methods are examined on data adopted from WHO website. Three LSTM, Convolutional GRU. bidirectional extension then considered each method forecast Australia Iran This novel as it carries out comprehensive evaluation aforementioned their extensions perform prediction COVID-19 death To best our knowledge, that Bi-GRU Bi-Conv-LSTM models used presented form graphs Friedman statistical test. results show have lower errors than other models. A several error metrics compare all models, finally, superiority determined. research could organisations working against determining long-term plans.

Язык: Английский

---

Densely attention mechanism based network for COVID-19 detection in chest X-rays

Scientific Reports, Год журнала: 2023, Номер 13(1)

Опубликована: Янв. 6, 2023

Automatic COVID-19 detection using chest X-ray (CXR) can play a vital part in large-scale screening and epidemic control. However, the radiographic features of CXR have different composite appearances, for instance, diffuse reticular-nodular opacities widespread ground-glass opacities. This makes automatic recognition imaging challenging task. To overcome this issue, we propose densely attention mechanism-based network (DAM-Net) CXR. DAM-Net adaptively extracts spatial from infected regions with various appearances scales. Our proposed is composed dense layers, channel adaptive downsampling layer, label smoothing regularization loss function. Dense layers extract approach builds up weights major feature channels suppresses redundant representations. We use cross-entropy function based on to limit effect interclass similarity upon The trained tested largest publicly available dataset, i.e., COVIDx, consisting 17,342 CXRs. Experimental results demonstrate that obtains state-of-the-art classification an accuracy 97.22%, sensitivity 96.87%, specificity 99.12%, precision 95.54%.

Язык: Английский

---

Ensemble of Autoencoders for Anomaly Detection in Biomedical Data: A Narrative Review

IEEE Access, Год журнала: 2024, Номер 12, С. 17273 - 17289

Опубликована: Янв. 1, 2024

In the context of biomedical data, an anomaly could refer to a rare or new type disease, aberration from normal behavior, unexpected observation requiring immediate attention. The detection anomalies in data has direct impact on health and safety individuals. However, anomalous events are rare, diverse, infrequent. Often, collection may involve significant loss human life healthcare costs. Therefore, traditional supervised machine deep learning algorithms not be directly applicable such problems. Biomedical often collected form images, electronic records, time series. Typically, autoencoder (AE) its corresponding variant is trained identified as deviation these based reconstruction error other metrics. An Ensemble AEs (EoAEs) can serve robust approach for by combining diverse accurate views data. EoAE provide superior single encoder; however, performance depend various factors, including diversity created accuracy individual AEs, combination their outcomes. Herein, we perform comprehensive narrative literature review use EoAEs when using different types Such ensemble provides promising offering potential improvement leveraging strengths AEs. several challenges remain, need specification determination optimal number ensemble. By addressing challenges, researchers enhance effectiveness Furthermore, through this review, highlight significance evaluating comparing with that establishing agreed-upon evaluation metrics investigating normalization techniques scores. We conclude presenting open questions field future research.

Язык: Английский

---

Epileptic Seizures Detection in EEG Signals Using Fusion Handcrafted and Deep Learning Features

Sensors, Год журнала: 2021, Номер 21(22), С. 7710 - 7710

Опубликована: Ноя. 19, 2021

Epilepsy is a brain disorder disease that affects people's quality of life. Electroencephalography (EEG) signals are used to diagnose epileptic seizures. This paper provides computer-aided diagnosis system (CADS) for the automatic seizures in EEG signals. The proposed method consists three steps, including preprocessing, feature extraction, and classification. In order perform simulations, Bonn Freiburg datasets used. Firstly, we band-pass filter with 0.5-40 Hz cut-off frequency removal artifacts datasets. Tunable-Q Wavelet Transform (TQWT) signal decomposition. second step, various linear nonlinear features extracted from TQWT sub-bands. this statistical, frequency, based on fractal dimensions (FDs) entropy theories. classification different approaches conventional machine learning (ML) deep (DL) discussed. CNN-RNN-based DL number layers applied. have been fed input CNN-RNN model, satisfactory results reported. K-fold cross-validation k = 10 employed demonstrate effectiveness procedure. revealed achieved an accuracy 99.71% 99.13%, respectively.

Язык: Английский

---

Automatic diagnosis of schizophrenia and attention deficit hyperactivity disorder in rs-fMRI modality using convolutional autoencoder model and interval type-2 fuzzy regression

Cognitive Neurodynamics, Год журнала: 2022, Номер 17(6), С. 1501 - 1523

Опубликована: Ноя. 12, 2022

Язык: Английский

---

Machine learning models for prediction of co-occurrence of diabetes and cardiovascular diseases: a retrospective cohort study

Journal of Diabetes & Metabolic Disorders, Год журнала: 2022, Номер 21(1), С. 251 - 261

Опубликована: Янв. 12, 2022

Abstract Background Diabetic mellitus (DM) and cardiovascular diseases (CVD) cause significant healthcare burden globally often co-exists. Current approaches fail to identify many people with co-occurrence of DM CVD, leading delay in seeking, increased complications morbidity. In this paper, we aimed develop evaluate a two-stage machine learning (ML) model predict the CVD. Methods We used diabetes screening research initiative (DiScRi) dataset containing >200 variables from >2000 participants. first stage, two ML models (logistic regression Evimp functions) implemented multivariate adaptive splines infer common risk factors for CVD applied correlation matrix reduce redundancy. second classification algorithm our model. evaluated prediction using accuracy, sensitivity specificity as performance metrics. Results Common was family history diseases, gender, deep breathing heart rate change, lying standing blood pressure HbA1c, HDL TC\HDL ratio. The predictive showed that participants HbA1c >6.45 ratio > 5.5 were at developing both (97.9% probability). contrast, ≤ more likely have only (84.5% probability) those ≤5.45 >1.45 be healthy (82.4%. Further, <1.45 (100% accuracy detect is 94.09%, 93.5%, 95.8%. Conclusions Our can significantly high attending program. This might help early detection patients who could benefit preventive treatment future burden.

Язык: Английский

---