1
|
Parsa M, Rad HY, Vaezi H, Hossein-Zadeh GA, Setarehdan SK, Rostami R, Rostami H, Vahabie AH. EEG-based classification of individuals with neuropsychiatric disorders using deep neural networks: A systematic review of current status and future directions. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE 2023; 240:107683. [PMID: 37406421 DOI: 10.1016/j.cmpb.2023.107683] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 06/05/2022] [Revised: 05/23/2023] [Accepted: 06/18/2023] [Indexed: 07/07/2023]
Abstract
The use of deep neural networks for electroencephalogram (EEG) classification has rapidly progressed and gained popularity in recent years, but automatic feature extraction from EEG signals remains a challenging task. The classification of neuropsychiatric disorders demands the extraction of neuro-markers for use in automated EEG classification. Numerous advanced deep learning algorithms can be used for this purpose. In this article, we present a comprehensive review of the main factors and parameters that affect the performance of deep neural networks in classifying different neuropsychiatric disorders using EEG signals. We also analyze the EEG features used for improving classification performance. Our analysis includes 82 scientific journal papers that applied deep neural networks for subject-wise classification based on EEG signals. We extracted information on the EEG dataset and types of disorders, deep neural network structures, performance, and hyperparameters. The results show that most studies have focused on clinical classification, achieving an average accuracy of 91.83 ± 7.34, with convolutional neural networks (CNNs) being the most frequently used network architecture and resting-state EEG signals being the most commonly used data type. Additionally, the review reveals that depression (N = 18), Alzheimer's (N = 11), and schizophrenia (N = 11) were studied more frequently than other types of neuropsychiatric disorders. Our review provides insight into the performance of deep neural networks in EEG classification and highlights the importance of EEG feature extraction in improving classification accuracy. By identifying the main factors and parameters that affect deep neural network performance in EEG classification, our review can guide future research in this area. We hope that our findings will encourage further exploration of deep learning methods for EEG classification and contribute to the development of more accurate and effective methods for diagnosing and monitoring neuropsychiatric disorders using EEG signals.
Collapse
Affiliation(s)
- Mohsen Parsa
- School of Electrical and Computer Engineering, College of Engineering, University of Tehran, North Kargar St., P.O. Box 14395/515, Tehran, Iran; School of Cognitive Sciences, Institute for Research in Fundamental Sciences, Artesh Highway, P.O. Box 19568-36484, Tehran, Iran
| | - Habib Yousefi Rad
- School of Electrical and Computer Engineering, College of Engineering, University of Tehran, North Kargar St., P.O. Box 14395/515, Tehran, Iran; School of Cognitive Sciences, Institute for Research in Fundamental Sciences, Artesh Highway, P.O. Box 19568-36484, Tehran, Iran
| | - Hadi Vaezi
- School of Electrical and Computer Engineering, College of Engineering, University of Tehran, North Kargar St., P.O. Box 14395/515, Tehran, Iran; School of Cognitive Sciences, Institute for Research in Fundamental Sciences, Artesh Highway, P.O. Box 19568-36484, Tehran, Iran
| | - Gholam-Ali Hossein-Zadeh
- Control and Intelligent Processing Center of Excellence, Faculty of Electrical and Computer Engineering, University of Tehran, North Kargar St., P.O. Box 14395/515, Tehran, Iran
| | - Seyed Kamaledin Setarehdan
- Control and Intelligent Processing Center of Excellence, Faculty of Electrical and Computer Engineering, University of Tehran, North Kargar St., P.O. Box 14395/515, Tehran, Iran
| | - Reza Rostami
- Faculty of Psychology and Education, University of Tehran, Jalal-Al-e-Ahmed, P.O. Box 14155-6456, Tehran, Iran
| | - Hana Rostami
- ACNC, Atieh Clinical Neuroscience Center, Valiasr St., P.O. Box 19697-13663, Tehran, Iran
| | - Abdol-Hossein Vahabie
- Faculty of Psychology and Education, University of Tehran, Jalal-Al-e-Ahmed, P.O. Box 14155-6456, Tehran, Iran; Cognitive Systems Laboratory, Control and Intelligent Processing Center of Excellence (CIPCE), School of Electrical and Computer Engineering, College of Engineering, University of Tehran, P.O. Box 14395/515, Tehran, Iran; Pasargad Institute for Advanced Innovative Solutions (PIAIS), Tehran, Iran.
| |
Collapse
|
2
|
Patricios JS, Schneider GM, van Ierssel J, Purcell LK, Davis GA, Echemendia RJ, Fremont P, Fuller GW, Herring SA, Harmon KG, Holte K, Loosemore M, Makdissi M, McCrea M, Meehan WP, O'Halloran P, Premji Z, Putukian M, Shill IJ, Turner M, Vaandering K, Webborn N, Yeates KO, Schneider KJ. Beyond acute concussion assessment to office management: a systematic review informing the development of a Sport Concussion Office Assessment Tool (SCOAT6) for adults and children. Br J Sports Med 2023; 57:737-748. [PMID: 37316204 DOI: 10.1136/bjsports-2023-106897] [Citation(s) in RCA: 4] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Accepted: 04/26/2023] [Indexed: 06/16/2023]
Abstract
OBJECTIVES To systematically review the scientific literature regarding the assessment of sport-related concussion (SRC) in the subacute phase (3-30 days) and provide recommendations for developing a Sport Concussion Office Assessment Tool (SCOAT6). DATA SOURCES MEDLINE, Embase, PsycINFO, Cochrane CENTRAL, CINAHL, SPORTDiscus and Web of Science searched from 2001 to 2022. Data extracted included study design, population, definition of SRC diagnosis, outcome measure(s) and results. ELIGIBILITY CRITERIA (1) Original research, cohort studies, case-control studies, diagnostic accuracy and case series with samples >10; (2) SRC; (3) screening/technology that assessed SRC in the subacute period and (4) low risk of bias (ROB). ROB was performed using adapted Scottish Intercollegiate Guidelines Network criteria. Quality of evidence was evaluated using the Strength of Recommendation Taxonomy classification. RESULTS Of 9913 studies screened, 127 met inclusion, assessing 12 overlapping domains. Results were summarised narratively. Studies of acceptable (81) or high (2) quality were used to inform the SCOAT6, finding sufficient evidence for including the assessment of autonomic function, dual gait, vestibular ocular motor screening (VOMS) and mental health screening. CONCLUSION Current SRC tools have limited utility beyond 72 hours. Incorporation of a multimodal clinical assessment in the subacute phase of SRC may include symptom evaluation, orthostatic hypotension screen, verbal neurocognitive tests, cervical spine evaluation, neurological screen, Modified Balance Error Scoring System, single/dual task tandem gait, modified VOMS and provocative exercise tests. Screens for sleep disturbance, anxiety and depression are recommended. Studies to evaluate the psychometric properties, clinical feasibility in different environments and time frames are needed. PROSPERO REGISTRATION NUMBER CRD42020154787.
Collapse
Affiliation(s)
- Jon S Patricios
- Wits Sport and Health (WiSH), School of Clinical Medicine, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg-Braamfontein, South Africa
| | - Geoff M Schneider
- Department of Radiology, University of Calgary, Calgary, Alberta, Canada
| | | | - Laura K Purcell
- Department of Pediatrics, McMaster University, Hamilton, Ontario, Canada
| | - Gavin A Davis
- Murdoch Children's Research Institute, Parkville, Victoria, Australia
| | - Ruben J Echemendia
- Psychology, University of Missouri, Kansas City, Missouri, USA
- University Orthopedics Concussion Care Clinic, State College Area School District, State College, Pennsylvania, USA
| | - Pierre Fremont
- Rehabilitation, Laval University, Quebec, Quebec, Canada
| | - Gordon Ward Fuller
- School of Health and Related Research, University of Sheffield, Sheffield, UK
| | - Stanley A Herring
- Departments of Rehabilitation Medicine, Orthopaedics and Sports Medicine and Neurological Surgery, University of Washington, Seattle, Washington, USA
| | | | | | - Mike Loosemore
- Institute for Sport Exercise and Health, University Collage Hospital London, London, UK
| | - Michael Makdissi
- Florey Institute of Neuroscience and Mental Health - Austin Campus, Heidelberg, Victoria, Australia
- La Trobe Sport and Exercise Medicine Research Centre, Melbourne, Victoria, Australia
| | - Michael McCrea
- Neurosurgery, Medical College of Wisconsin, Milwaukee, Wisconsin, USA
| | - William P Meehan
- Sports Medicine, Children's Hospital Boston, Boston, Massachusetts, USA
- Emergency Medicine, Children's Hospital Boston, Boston, Massachusetts, USA
| | - Patrick O'Halloran
- Neurotrauma and Ophthalmology Research Group, University of Birmingham, Birmingham, UK
- Health Education England West Midlands, Edgbaston, UK
| | - Zahra Premji
- Libraries, University of Victoria, Victoria, British Columbia, Canada
| | | | - Isla Jordan Shill
- Sport Injury Prevention Research Centre, University of Calgary, Calgary, Alberta, Canada
| | - Michael Turner
- International Concussion and Head Injury Research Foundation, London, UK
- University College London, London, UK
| | - Kenzie Vaandering
- University of Calgary Faculty of Kinesiology, Calgary, Alberta, Canada
| | - Nick Webborn
- Medical Committee, International Paralympic Committee, Bonn, Germany
- School of Sport, Exercise and Health Sciences, Loughborough University, Loughborough, UK
| | - Keith Owen Yeates
- Department of Psychology, University of Calgary, Calgary, Alberta, Canada
- Hotchkiss Brain Institute, University of Calgary, Calgary, Alberta, Canada
- Alberta Children's Hospital Research Institute, University of Calgary, Calgary, Alberta, Canada
| | - Kathryn J Schneider
- Sport Injury Prevention Research Centre, Faculty of Kinesiology, University of Calgary, Calgary, Alberta, Canada
| |
Collapse
|
3
|
DeMatteo CA, Jakubowski J, Randall S, Stazyk K, Lin CY, Yakubov R. School performance in youth after a concussion. Front Sports Act Living 2022; 4:1008551. [PMID: 36619354 PMCID: PMC9813779 DOI: 10.3389/fspor.2022.1008551] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/31/2022] [Accepted: 11/16/2022] [Indexed: 12/24/2022] Open
Abstract
Objective This study aimed to identify school problems and levels of cognitive activity in youths aged 5-18 years with a concussion during the recovery stages of return to school (RTS). Study Design In a prospective cohort, participants completed in-person assessments at three time points: First Visit Post-injury, Symptom Resolution Visit, and Follow-Up Visit. These time points varied based on the participants' recovery progress. The post-concussion symptom scale (PCSS) and a cognitive activity scale were completed every 2 days until symptom resolution was achieved. Participants and their parents completed a school questionnaire detailing how their concussion had impacted their school learning/performance and their level of concern about their injury as well as the Immediate Post-Concussion Assessment and Cognitive Testing (ImPACT). Results Sixty-three percent (N = 44/70) of participants returned to school by the First Visit Post-injury (average 7.7 days following injury), and of these, 50% (N = 22) were experiencing school problems. Sixty-five participants (out of 70) returned to school at the Follow-Up Visit, and of these, 18% reported school problems. There was a significant difference in the school problems reported by parents and youth. At the First Visit Post-injury, the youth reported more problems (p = 0.02), and the In-Person Symptom Resolution Visit with parents reported more problems (p = 0.01). The cognitive activity score increased, while the PCSS score decreased from RTS Stage 1 to Stage 5. Conclusions This study identified that 50% of youth experienced school problems at the First Visit Post-injury, whereas only 18% reported school problems at the Follow-Up Visit. There is a significant difference in the perception of school problems reported by youth and their parents at different stages of recovery. The amount and complexity of cognitive activity increased with decreasing symptoms and increasing RTS stage. Findings can guide youth with a concussion and their parents in supporting a cautious return to school with accommodations. Healthcare providers and researchers can use this knowledge to better support youth in their return to school and understand the importance of gathering information from youth and their parents to gain the best insight into recovery.
Collapse
Affiliation(s)
- Carol A. DeMatteo
- School of Rehabilitation Sciences, McMaster University, Hamilton, ON, Canada,CanChild Centre for Childhood Disability Research, McMaster University, Hamilton, ON, Canada,Correspondence: Carol A. DeMatteo
| | - Josephine Jakubowski
- CanChild Centre for Childhood Disability Research, McMaster University, Hamilton, ON, Canada
| | - Sarah Randall
- School of Rehabilitation Sciences, McMaster University, Hamilton, ON, Canada,CanChild Centre for Childhood Disability Research, McMaster University, Hamilton, ON, Canada
| | - Kathy Stazyk
- School of Rehabilitation Sciences, McMaster University, Hamilton, ON, Canada
| | - Chia-Yu Lin
- CanChild Centre for Childhood Disability Research, McMaster University, Hamilton, ON, Canada
| | - Rebecca Yakubov
- CanChild Centre for Childhood Disability Research, McMaster University, Hamilton, ON, Canada
| |
Collapse
|
4
|
Jin D, Sergeeva E, Weng WH, Chauhan G, Szolovits P. Explainable deep learning in healthcare: A methodological survey from an attribution view. WIREs Mech Dis 2022; 14:e1548. [PMID: 35037736 DOI: 10.1002/wsbm.1548] [Citation(s) in RCA: 4] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/09/2021] [Revised: 11/06/2021] [Accepted: 11/09/2021] [Indexed: 11/08/2022]
Abstract
The increasing availability of large collections of electronic health record (EHR) data and unprecedented technical advances in deep learning (DL) have sparked a surge of research interest in developing DL based clinical decision support systems for diagnosis, prognosis, and treatment. Despite the recognition of the value of deep learning in healthcare, impediments to further adoption in real healthcare settings remain due to the black-box nature of DL. Therefore, there is an emerging need for interpretable DL, which allows end users to evaluate the model decision making to know whether to accept or reject predictions and recommendations before an action is taken. In this review, we focus on the interpretability of the DL models in healthcare. We start by introducing the methods for interpretability in depth and comprehensively as a methodological reference for future researchers or clinical practitioners in this field. Besides the methods' details, we also include a discussion of advantages and disadvantages of these methods and which scenarios each of them is suitable for, so that interested readers can know how to compare and choose among them for use. Moreover, we discuss how these methods, originally developed for solving general-domain problems, have been adapted and applied to healthcare problems and how they can help physicians better understand these data-driven technologies. Overall, we hope this survey can help researchers and practitioners in both artificial intelligence and clinical fields understand what methods we have for enhancing the interpretability of their DL models and choose the optimal one accordingly. This article is categorized under: Cancer > Computational Models.
Collapse
Affiliation(s)
- Di Jin
- Computer Science and Artificial Intelligence Laboratory, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA
| | - Elena Sergeeva
- Computer Science and Artificial Intelligence Laboratory, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA
| | - Wei-Hung Weng
- Computer Science and Artificial Intelligence Laboratory, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA
| | - Geeticka Chauhan
- Computer Science and Artificial Intelligence Laboratory, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA
| | - Peter Szolovits
- Computer Science and Artificial Intelligence Laboratory, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA
| |
Collapse
|
5
|
Thanjavur K, Hristopulos DT, Babul A, Yi KM, Virji-Babul N. Deep Learning Recurrent Neural Network for Concussion Classification in Adolescents Using Raw Electroencephalography Signals: Toward a Minimal Number of Sensors. Front Hum Neurosci 2021; 15:734501. [PMID: 34899212 PMCID: PMC8654150 DOI: 10.3389/fnhum.2021.734501] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/01/2021] [Accepted: 11/01/2021] [Indexed: 11/13/2022] Open
Abstract
Artificial neural networks (ANNs) are showing increasing promise as decision support tools in medicine and particularly in neuroscience and neuroimaging. Recently, there has been increasing work on using neural networks to classify individuals with concussion using electroencephalography (EEG) data. However, to date the need for research grade equipment has limited the applications to clinical environments. We recently developed a deep learning long short-term memory (LSTM) based recurrent neural network to classify concussion using raw, resting state data using 64 EEG channels and achieved high accuracy in classifying concussion. Here, we report on our efforts to develop a clinically practical system using a minimal subset of EEG sensors. EEG data from 23 athletes who had suffered a sport-related concussion and 35 non-concussed, control athletes were used for this study. We tested and ranked each of the original 64 channels based on its contribution toward the concussion classification performed by the original LSTM network. The top scoring channels were used to train and test a network with the same architecture as the previously trained network. We found that with only six of the top scoring channels the classifier identified concussions with an accuracy of 94%. These results show that it is possible to classify concussion using raw, resting state data from a small number of EEG sensors, constituting a first step toward developing portable, easy to use EEG systems that can be used in a clinical setting.
Collapse
Affiliation(s)
- Karun Thanjavur
- Department of Physics and Astronomy, University of Victoria, Victoria, BC, Canada
| | | | - Arif Babul
- Department of Physics and Astronomy, University of Victoria, Victoria, BC, Canada
| | - Kwang Moo Yi
- Department of Computer Science, University of British Columbia, Vancouver, BC, Canada
| | - Naznin Virji-Babul
- Djavad Mowafaghian Centre for Brain Health, University of British Columbia, Vancouver, BC, Canada.,Department of Physical Therapy, Faculty of Medicine, University of British Columbia, Vancouver, BC, Canada
| |
Collapse
|
6
|
Carrick FR, Pagnacco G, Azzolino SF, Hunfalvay M, Oggero E, Frizzell T, Smith CJ, Pawlowski G, Campbell NKJ, Fickling SD, Lakhani B, D'Arcy RCN. Brain Vital Signs in Elite Ice Hockey: Towards Characterizing Objective and Specific Neurophysiological Reference Values for Concussion Management. Front Neurosci 2021; 15:670563. [PMID: 34434084 PMCID: PMC8382572 DOI: 10.3389/fnins.2021.670563] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/21/2021] [Accepted: 07/09/2021] [Indexed: 12/02/2022] Open
Abstract
Background: Prior concussion studies have shown that objective neurophysiological measures are sensitive to detecting concussive and subconcussive impairments in youth ice-hockey. These studies monitored brain vital signs at rink-side using a within-subjects design to demonstrate significant changes from pre-season baseline scans. However, practical clinical implementation must overcome inherent challenges related to any dependence on a baseline. This requires establishing the start of normative reference data sets. Methods: The current study collected specific reference data for N = 58 elite, youth, male ice-hockey players and compared these with a general reference dataset from N = 135 of males and females across the lifespan. The elite hockey players were recruited to a select training camp through CAA Hockey, a management agency for players drafted to leagues such as the National Hockey League (NHL). The statistical analysis included a test-retest comparison to establish reliability, and a multivariate analysis of covariance to evaluate differences in brain vital signs between groups with age as a covariate. Findings: Test-retest assessments for brain vital signs evoked potentials showed moderate-to-good reliability (Cronbach’s Alpha > 0.7, Intraclass correlation coefficient > 0.5) in five out of six measures. The multivariate analysis of covariance showed no overall effect for group (p = 0.105), and a significant effect of age as a covariate was observed (p < 0.001). Adjusting for the effect of age, a significant difference was observed in the measure of N100 latency (p = 0.022) between elite hockey players and the heterogeneous control group. Interpretation: The findings support the concept that normative physiological data can be used in brain vital signs evaluation in athletes, and should additionally be stratified for age, skill level, and experience. These can be combined with general norms and/or individual baseline assessments where appropriate and/or possible. The current results allow for brain vital sign evaluation independent of baseline assessment, therefore enabling objective neurophysiological evaluation of concussion management and cognitive performance optimization in ice-hockey.
Collapse
Affiliation(s)
- Frederick R Carrick
- University of Central Florida College of Medicine, Orlando, FL, United States.,MGH Institute of Health Professions, Boston, MA, United States.,Centre for Mental Health Research, University of Cambridge, Cambridge, United Kingdom.,Centre for Mental Health Research in Association with University of Cambridge, Cambridge, United Kingdom
| | - Guido Pagnacco
- Centre for Mental Health Research in Association with University of Cambridge, Cambridge, United Kingdom.,Department of Electrical and Computer Engineering, University of Wyoming, Laramie, WY, United States
| | - Sergio F Azzolino
- Centre for Mental Health Research in Association with University of Cambridge, Cambridge, United Kingdom
| | - Melissa Hunfalvay
- Centre for Mental Health Research in Association with University of Cambridge, Cambridge, United Kingdom
| | - Elena Oggero
- Centre for Mental Health Research in Association with University of Cambridge, Cambridge, United Kingdom.,Department of Electrical and Computer Engineering, University of Wyoming, Laramie, WY, United States
| | - Tory Frizzell
- BrainNET, Health and Technology District, Vancouver, BC, Canada
| | | | - Gabriela Pawlowski
- BrainNET, Health and Technology District, Vancouver, BC, Canada.,Centre for Neurology Studies, HealthTech Connex, Vancouver, BC, Canada
| | - Natasha K J Campbell
- BrainNET, Health and Technology District, Vancouver, BC, Canada.,Centre for Neurology Studies, HealthTech Connex, Vancouver, BC, Canada
| | - Shaun D Fickling
- BrainNET, Health and Technology District, Vancouver, BC, Canada.,Centre for Neurology Studies, HealthTech Connex, Vancouver, BC, Canada
| | - Bimal Lakhani
- Centre for Neurology Studies, HealthTech Connex, Vancouver, BC, Canada
| | - Ryan C N D'Arcy
- BrainNET, Health and Technology District, Vancouver, BC, Canada.,Centre for Neurology Studies, HealthTech Connex, Vancouver, BC, Canada.,DM Centre for Brain Health, Department of Radiology, University of British Columbia, Vancouver, BC, Canada
| |
Collapse
|
7
|
Ozyegen O, Ilic I, Cevik M. Evaluation of interpretability methods for multivariate time series forecasting. APPL INTELL 2021; 52:4727-4743. [PMID: 34764613 PMCID: PMC8315500 DOI: 10.1007/s10489-021-02662-2] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Accepted: 07/05/2021] [Indexed: 11/30/2022]
Abstract
Being able to interpret a model's predictions is a crucial task in many machine learning applications. Specifically, local interpretability is important in determining why a model makes particular predictions. Despite the recent focus on interpretable Artificial Intelligence (AI), there have been few studies on local interpretability methods for time series forecasting, while existing approaches mainly focus on time series classification tasks. In this study, we propose two novel evaluation metrics for time series forecasting: Area Over the Perturbation Curve for Regression and Ablation Percentage Threshold. These two metrics can measure the local fidelity of local explanation methods. We extend the theoretical foundation to collect experimental results on four popular datasets. Both metrics enable a comprehensive comparison of numerous local explanation methods, and an intuitive approach to interpret model predictions. Lastly, we provide heuristical reasoning for this analysis through an extensive numerical study.
Collapse
Affiliation(s)
- Ozan Ozyegen
- Data Science Lab, Ryerson University, Toronto, Canada
| | - Igor Ilic
- Data Science Lab, Ryerson University, Toronto, Canada
| | - Mucahit Cevik
- Data Science Lab, Ryerson University, Toronto, Canada
| |
Collapse
|
8
|
Thanjavur K, Babul A, Foran B, Bielecki M, Gilchrist A, Hristopulos DT, Brucar LR, Virji-Babul N. Recurrent neural network-based acute concussion classifier using raw resting state EEG data. Sci Rep 2021; 11:12353. [PMID: 34117309 PMCID: PMC8196170 DOI: 10.1038/s41598-021-91614-4] [Citation(s) in RCA: 9] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/08/2020] [Accepted: 05/24/2021] [Indexed: 02/05/2023] Open
Abstract
Concussion is a global health concern. Despite its high prevalence, a sound understanding of the mechanisms underlying this type of diffuse brain injury remains elusive. It is, however, well established that concussions cause significant functional deficits; that children and youths are disproportionately affected and have longer recovery time than adults; and that individuals suffering from a concussion are more prone to experience additional concussions, with each successive injury increasing the risk of long term neurological and mental health complications. Currently, the most significant challenge in concussion management is the lack of objective, clinically- accepted, brain-based approaches for determining whether an athlete has suffered a concussion. Here, we report on our efforts to address this challenge. Specifically, we introduce a deep learning long short-term memory (LSTM)-based recurrent neural network that is able to distinguish between non-concussed and acute post-concussed adolescent athletes using only short (i.e. 90 s long) samples of resting state EEG data as input. The athletes were neither required to perform a specific task nor expected to respond to a stimulus during data collection. The acquired EEG data were neither filtered, cleaned of artefacts, nor subjected to explicit feature extraction. The LSTM network was trained and validated using data from 27 male, adolescent athletes with sports related concussion, benchmarked against 35 non-concussed adolescent athletes. During rigorous testing, the classifier consistently identified concussions with an accuracy of > 90% and achieved an ensemble median Area Under the Receiver Operating Characteristic Curve (ROC/AUC) equal to 0.971. This is the first instance of a high-performing classifier that relies only on easy-to-acquire resting state, raw EEG data. Our concussion classifier represents a promising first step towards the development of an easy-to-use, objective, brain-based, automatic classification of concussion at an individual level.
Collapse
Affiliation(s)
- Karun Thanjavur
- Department of Physics and Astronomy, University of Victoria, Victoria, BC, V8P 5C2, Canada.
| | - Arif Babul
- Department of Physics and Astronomy, University of Victoria, Victoria, BC, V8P 5C2, Canada
| | - Brandon Foran
- Department of Computer Science, Middlesex College, Western University, London, ON, N6A 5B7, Canada
| | - Maya Bielecki
- Department of Computer Science, Middlesex College, Western University, London, ON, N6A 5B7, Canada
| | - Adam Gilchrist
- Department of Computer Science, Middlesex College, Western University, London, ON, N6A 5B7, Canada
| | - Dionissios T Hristopulos
- School of Electrical and Computer Engineering, Technical University of Crete, 73100, Chania, Greece
| | - Leyla R Brucar
- Djavad Mowafaghian Centre for Brain Health, University of British Columbia, Vancouver, BC, V6T 1Z3, Canada
| | - Naznin Virji-Babul
- Djavad Mowafaghian Centre for Brain Health, University of British Columbia, Vancouver, BC, V6T 1Z3, Canada
- Department of Physical Therapy, Faculty of Medicine, University of British Columbia, Vancouver, BC, V6T 1Z3, Canada
| |
Collapse
|
9
|
Sargolzaei S. Can Deep Learning Hit a Moving Target? A Scoping Review of Its Role to Study Neurological Disorders in Children. Front Comput Neurosci 2021; 15:670489. [PMID: 34025380 PMCID: PMC8131543 DOI: 10.3389/fncom.2021.670489] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/21/2021] [Accepted: 04/09/2021] [Indexed: 12/12/2022] Open
Abstract
Neurological disorders dramatically impact patients of any age population, their families, and societies. Pediatrics are among vulnerable age populations who differently experience the devastating consequences of neurological conditions, such as attention-deficit hyperactivity disorders (ADHD), autism spectrum disorders (ASD), cerebral palsy, concussion, and epilepsy. System-level understanding of these neurological disorders, particularly from the brain networks' dynamic perspective, has led to the significant trend of recent scientific investigations. While a dramatic maturation in the network science application domain is evident, leading to a better understanding of neurological disorders, such rapid utilization for studying pediatric neurological disorders falls behind that of the adult population. Aside from the specific technological needs and constraints in studying neurological disorders in children, the concept of development introduces uncertainty and further complexity topping the existing neurologically driven processes caused by disorders. To unravel these complexities, indebted to the availability of high-dimensional data and computing capabilities, approaches based on machine learning have rapidly emerged a new trend to understand pathways better, accurately diagnose, and better manage the disorders. Deep learning has recently gained an ever-increasing role in the era of health and medical investigations. Thanks to its relatively more minor dependency on feature exploration and engineering, deep learning may overcome the challenges mentioned earlier in studying neurological disorders in children. The current scoping review aims to explore challenges concerning pediatric brain development studies under the constraints of neurological disorders and offer an insight into the potential role of deep learning methodology on such a task with varying and uncertain nature. Along with pinpointing recent advancements, possible research directions are highlighted where deep learning approaches can assist in computationally targeting neurological disorder-related processes and translating them into windows of opportunities for interventions in diagnosis, treatment, and management of neurological disorders in children.
Collapse
Affiliation(s)
- Saman Sargolzaei
- Department of Engineering, College of Engineering and Natural Sciences, University of Tennessee at Martin, Martin, TN, United States
| |
Collapse
|
10
|
Elevated and Slowed EEG Oscillations in Patients with Post-Concussive Syndrome and Chronic Pain Following a Motor Vehicle Collision. Brain Sci 2021; 11:brainsci11050537. [PMID: 33923286 PMCID: PMC8145977 DOI: 10.3390/brainsci11050537] [Citation(s) in RCA: 6] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/14/2021] [Revised: 04/16/2021] [Accepted: 04/21/2021] [Indexed: 11/17/2022] Open
Abstract
(1) Background: Mild traumatic brain injury produces significant changes in neurotransmission including brain oscillations. We investigated potential quantitative electroencephalography biomarkers in 57 patients with post-concussive syndrome and chronic pain following motor vehicle collision, and 54 healthy nearly age- and sex-matched controls. (2) Methods: Electroencephalography processing was completed in MATLAB, statistical modeling in SPSS, and machine learning modeling in Rapid Miner. Group differences were calculated using current-source density estimation, yielding whole-brain topographical distributions of absolute power, relative power and phase-locking functional connectivity. Groups were compared using independent sample Mann–Whitney U tests. Effect sizes and Pearson correlations were also computed. Machine learning analysis leveraged a post hoc supervised learning support vector non-probabilistic binary linear kernel classification to generate predictive models from the derived EEG signatures. (3) Results: Patients displayed significantly elevated and slowed power compared to controls: delta (p = 0.000000, r = 0.6) and theta power (p < 0.0001, r = 0.4), and relative delta power (p < 0.00001) and decreased relative alpha power (p < 0.001). Absolute delta and theta power together yielded the strongest machine learning classification accuracy (87.6%). Changes in absolute power were moderately correlated with duration and persistence of symptoms in the slow wave frequency spectrum (<15 Hz). (4) Conclusions: Distributed increases in slow wave oscillatory power are concurrent with post-concussive syndrome and chronic pain.
Collapse
|
11
|
Fickling SD, Smith AM, Stuart MJ, Dodick DW, Farrell K, Pender SC, D'Arcy RCN. Subconcussive brain vital signs changes predict head-impact exposure in ice hockey players. Brain Commun 2021; 3:fcab019. [PMID: 33855296 PMCID: PMC8023684 DOI: 10.1093/braincomms/fcab019] [Citation(s) in RCA: 9] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/13/2020] [Revised: 12/04/2020] [Accepted: 03/08/2021] [Indexed: 01/09/2023] Open
Abstract
The brain vital signs framework is a portable, objective, neurophysiological evaluation of brain function at point-of-care. We investigated brain vital signs at pre- and post-season for age 14 or under (Bantam) and age 16-20 (Junior-A) male ice hockey players to (i) further investigate previously published brain vital sign results showing subconcussive cognitive deficits and (ii) validate these findings through comparison with head-impact data obtained from instrumented accelerometers. With a longitudinal study design, 23 male ice hockey players in Bantam (n = 13; age 13.63 ± 0.62) and Tier II Junior-A (n = 10; age 18.62 ± 0.86) divisions were assessed at pre- and post-season. None were diagnosed with a concussion during the season. Cognitive evoked potential measures of Auditory sensation (N100), Basic attention (P300) and Cognitive processing (N400) were analysed as changes in peak amplitudes and latencies (six standard scores total). A regression analysis examined the relationship between brain vital signs and the number of head impacts received during the study season. Significant pre/post differences in brain vital signs were detected for both groups. Bantam and Junior-A players also differed in number of head impacts (Bantam: 32.92 ± 17.68; Junior-A: 195.00 ± 61.08; P < 0.001). Importantly, the regression model demonstrated a significant linear relationship between changes in brain vital signs and total head impacts received (R = 0.799, P = 0.007), with clear differences between the Bantam and Junior-A groups. In the absence of a clinically diagnosed concussion, the brain vital sign changes appear to have demonstrated the compounding effects of repetitive subconcussive impacts. The findings underscored the importance of an objective physiological measure of brain function along the spectrum of concussive impacts.
Collapse
Affiliation(s)
- Shaun D Fickling
- Faculty of Science and Applied Sciences, Simon Fraser University, Metro Vancouver, BC V5A1S6, Canada.,Center for Neurology Studies, HealthTech Connex, Metro Vancouver, BC V3V0C6, Canada.,BrainNET, Health and Technology District, Surrey, BC V3V0C6, Canada
| | - Aynsley M Smith
- Department of Physical Medicine and Rehabilitation, Sports Medicine Center, Mayo Clinic, Rochester, MN 55905, USA.,Department of Orthopedic Surgery, Sports Medicine Center, Mayo Clinic, Rochester, MN 55905, USA
| | - Michael J Stuart
- Department of Orthopedic Surgery, Sports Medicine Center, Mayo Clinic, Rochester, MN 55905, USA
| | - David W Dodick
- Department of Neurology, Mayo Clinic, Phoenix, AZ 85259, USA
| | - Kyle Farrell
- Creighton University School of Medicine, Omaha, Nebraska 68178, USA
| | - Sara C Pender
- School of Medicine, University College Dublin, Dublin D04 V1W8, Ireland
| | - Ryan C N D'Arcy
- Faculty of Science and Applied Sciences, Simon Fraser University, Metro Vancouver, BC V5A1S6, Canada.,Center for Neurology Studies, HealthTech Connex, Metro Vancouver, BC V3V0C6, Canada.,BrainNET, Health and Technology District, Surrey, BC V3V0C6, Canada.,DM Centre for Brain Health, Radiology, University of British Columbia, Metro Vancouver, BC V6T1Z4, Canada
| |
Collapse
|
12
|
Ferris LM, Kontos AP, Eagle SR, Elbin RJ, Collins MW, Mucha A, Clugston JR, Port NL. Predictive Accuracy of the Sport Concussion Assessment Tool 3 and Vestibular/Ocular-Motor Screening, Individually and In Combination: A National Collegiate Athletic Association-Department of Defense Concussion Assessment, Research and Education Consortium Analysis. Am J Sports Med 2021; 49:1040-1048. [PMID: 33600216 DOI: 10.1177/0363546520988098] [Citation(s) in RCA: 16] [Impact Index Per Article: 5.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 01/31/2023]
Abstract
BACKGROUND Vestibular and ocular symptoms in sport-related concussions are common. The Vestibular/Ocular-Motor Screening (VOMS) tool is a rapid, free, pen-and-paper tool that directly assesses these symptoms and shows consistent utility in concussion identification, prognosis, and management. However, a VOMS validation study in the acute concussion period of a large sample is lacking. PURPOSE To examine VOMS validity among collegiate student-athletes, concussed and nonconcussed, from the multisite National Collegiate Athletic Association-Department of Defense Concussion Assessment, Research and Education (CARE) Consortium. A secondary aim was to utilize multidimensional machine learning pattern classifiers to deduce the additive power of the VOMS in relation to components of the Sport Concussion Assessment Tool 3 (SCAT3). STUDY DESIGN Cohort study (diagnosis); Level of evidence, 3. METHODS Preseason and acute concussion assessments were analyzed for 419 student-athletes. Variables in the analysis included the VOMS, Balance Error Scoring System, Standardized Assessment of Concussion, and SCAT3 symptom evaluation score. Descriptive statistics were calculated for all tools, including Kolmogorov-Smirnov significance and Cohen d effect size. Correlations between tools were analyzed with Spearman r, and predictive accuracy was evaluated through an Ada Boosted Tree machine learning model's generated receiver operating characteristic curves. RESULTS Total VOMS scores and SCAT3 symptom scores demonstrated significant increases in the acute concussion time frame (Cohen d = 1.23 and 1.06; P < .0001), whereas the Balance Error Scoring System lacked clinical significance (Cohen d = 0.17). Incorporation of VOMS into the full SCAT3 significantly boosted overall diagnostic ability by 4.4% to an area under the curve of 0.848 (P < .0001) and produced a 9% improvement in test sensitivity over the existing SCAT3 battery. CONCLUSION The results from this study highlight the relevance of the vestibular and oculomotor systems to concussion and the utility of the VOMS tool. Given the 3.8 million sports-related and 45,121 military-related concussions per year, the addition of VOMS to the SCAT3 is poised to identify up to an additional 304,000 athletes and 3610 servicemembers annually who are concussed, thereby improving concussion assessment and diagnostic rates. Health care providers should consider the addition of VOMS to their concussion assessment toolkits, as its use can positively affect assessment and management of concussions, which may ultimately improve outcomes for this complex and common injury.
Collapse
Affiliation(s)
- Lyndsey M Ferris
- Indiana University School of Optometry, Bloomington, Indiana, USA
| | | | - Shawn R Eagle
- University of Pittsburgh, Pittsburgh, Pennsylvania, USA
| | - R J Elbin
- University of Arkansas, Fayatteville, Arkansas, USA
| | | | - Anne Mucha
- UPMC Centers for Rehab Services, Pittsburgh, Pennsylvania, USA
| | | | - Nicholas L Port
- Indiana University School of Optometry, Bloomington, Indiana, USA
| |
Collapse
|
13
|
Vivaldi N, Caiola M, Solarana K, Ye M. Evaluating Performance of EEG Data-Driven Machine Learning for Traumatic Brain Injury Classification. IEEE Trans Biomed Eng 2021; 68:3205-3216. [PMID: 33635785 PMCID: PMC9513823 DOI: 10.1109/tbme.2021.3062502] [Citation(s) in RCA: 11] [Impact Index Per Article: 3.7] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/10/2022]
Abstract
Objectives: Big data analytics can potentially benefit the assessment and management of complex neurological conditions by extracting information that is difficult to identify manually. In this study, we evaluated the performance of commonly used supervised machine learning algorithms in the classification of patients with traumatic brain injury (TBI) history from those with stroke history and/or normal EEG. Methods: Support vector machine (SVM) and K-nearest neighbors (KNN) models were generated with a diverse feature set from Temple EEG Corpus for both two-class classification of patients with TBI history from normal subjects and three-class classification of TBI, stroke and normal subjects. Results: For two-class classification, an accuracy of 0.94 was achieved in 10-fold cross validation (CV), and 0.76 in independent validation (IV). For three-class classification, 0.85 and 0.71 accuracy were reached in CV and IV respectively. Overall, linear discriminant analysis (LDA) feature selection and SVM models consistently performed well in both CV and IV and for both two-class and three-class classification. Compared to normal control, both TBI and stroke patients showed an overall reduction in coherence and relative PSD in delta frequency, and an increase in higher frequency (alpha, mu, beta and gamma) power. But stroke patients showed a greater degree of change and had additional global decrease in theta power. Conclusions: Our study suggests that EEG data-driven machine learning can be a useful tool for TBI classification. Significance: Our study provides preliminary evidence that EEG ML algorithm can potentially provide specificity to separate different neurological conditions.
Collapse
|
14
|
Gong S, Xing K, Cichocki A, Li J. Deep Learning in EEG: Advance of the Last Ten-Year Critical Period. IEEE Trans Cogn Dev Syst 2021. [DOI: 10.1109/tcds.2021.3079712] [Citation(s) in RCA: 11] [Impact Index Per Article: 3.7] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/08/2022]
|
15
|
Boshra R, Ruiter KI, Dhindsa K, Sonnadara R, Reilly JP, Connolly JF. On the time-course of functional connectivity: theory of a dynamic progression of concussion effects. Brain Commun 2020; 2:fcaa063. [PMID: 32954320 PMCID: PMC7491441 DOI: 10.1093/braincomms/fcaa063] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/18/2019] [Revised: 04/15/2020] [Accepted: 04/24/2020] [Indexed: 12/27/2022] Open
Abstract
The current literature presents a discordant view of mild traumatic brain injury and its effects on the human brain. This dissonance has often been attributed to heterogeneities in study populations, aetiology, acuteness, experimental paradigms and/or testing modalities. To investigate the progression of mild traumatic brain injury in the human brain, the present study employed data from 93 subjects (48 healthy controls) representing both acute and chronic stages of mild traumatic brain injury. The effects of concussion across different stages of injury were measured using two metrics of functional connectivity in segments of electroencephalography time-locked to an active oddball task. Coherence and weighted phase-lag index were calculated separately for individual frequency bands (delta, theta, alpha and beta) to measure the functional connectivity between six electrode clusters distributed from frontal to parietal regions across both hemispheres. Results show an increase in functional connectivity in the acute stage after mild traumatic brain injury, contrasted with significantly reduced functional connectivity in chronic stages of injury. This finding indicates a non-linear time-dependent effect of injury. To understand this pattern of changing functional connectivity in relation to prior evidence, we propose a new model of the time-course of the effects of mild traumatic brain injury on the brain that brings together research from multiple neuroimaging modalities and unifies the various lines of evidence that at first appear to be in conflict.
Collapse
Affiliation(s)
- Rober Boshra
- ARiEAL Research Centre, McMaster University, Hamilton, ON L8S 4K1, Canada.,School of Biomedical Engineering, McMaster University, Hamilton, ON L8S 4K1, Canada.,Vector Institute, Toronto, ON M5G 1M1, Canada
| | - Kyle I Ruiter
- ARiEAL Research Centre, McMaster University, Hamilton, ON L8S 4K1, Canada.,Linguistics and Languages, McMaster University, Hamilton, ON L8S 4K1, Canada
| | - Kiret Dhindsa
- ARiEAL Research Centre, McMaster University, Hamilton, ON L8S 4K1, Canada.,Vector Institute, Toronto, ON M5G 1M1, Canada.,Department of Surgery, McMaster University, Hamilton, ON L8S 4K1, Canada
| | - Ranil Sonnadara
- ARiEAL Research Centre, McMaster University, Hamilton, ON L8S 4K1, Canada.,Vector Institute, Toronto, ON M5G 1M1, Canada.,Department of Surgery, McMaster University, Hamilton, ON L8S 4K1, Canada.,Department of Psychology, Neuroscience & Behaviour, McMaster University, Hamilton, ON L8S 4K1, Canada
| | - James P Reilly
- ARiEAL Research Centre, McMaster University, Hamilton, ON L8S 4K1, Canada.,School of Biomedical Engineering, McMaster University, Hamilton, ON L8S 4K1, Canada.,Vector Institute, Toronto, ON M5G 1M1, Canada.,Department of Electrical & Computer Engineering, McMaster University, Hamilton, ON L8S 4K1, Canada
| | - John F Connolly
- ARiEAL Research Centre, McMaster University, Hamilton, ON L8S 4K1, Canada.,School of Biomedical Engineering, McMaster University, Hamilton, ON L8S 4K1, Canada.,Vector Institute, Toronto, ON M5G 1M1, Canada.,Linguistics and Languages, McMaster University, Hamilton, ON L8S 4K1, Canada.,Department of Psychology, Neuroscience & Behaviour, McMaster University, Hamilton, ON L8S 4K1, Canada
| |
Collapse
|