Supported diagnosis of adhd from eeg signals based on hidden markov models and probability product kernels

dc.contributorOrozco Gutiérrez, Alvaro Angel
dc.creatorMaya Piedrahita, María Camila
dc.date2022-04-19T20:57:04Z
dc.date2022-04-19T20:57:04Z
dc.date2021
dc.date.accessioned2022-09-23T21:32:12Z
dc.date.available2022-09-23T21:32:12Z
dc.identifierUniversidad Tecnológica de Pereira
dc.identifierRepositorio institucional Universidad Tecnológica de Pereira
dc.identifierhttps://repositorio.utp.edu.co/home
dc.identifierhttps://hdl.handle.net/11059/14024
dc.identifier.urihttp://repositorioslatinoamericanos.uchile.cl/handle/2250/3530896
dc.descriptionAttention deficit hyperactivity disorder (ADHD), most often present in childhood, may persist in adult life, hampering personal development. However, ADHD diagnosis is a real challenge since it highly depends on the clinical observation of the patient, the parental and scholar information, and the specialist expertise. Despite demanded objective diagnosis aids from biosignals, the physiological biomarkers lack robustness and significance under the non-stationary and non-linear electroencephalographic dynamics. Therefore, this work presents a supported diagnosis methodology for ADHD from the dynamic characterization of EEG based on hidden Markov models (HMM) and probability product kernels (PPK). Based on the symptom of impulsivity, the proposed approach trains an HMM for each subject from EEG signals in failed inhibition tasks. In the first instance, PPK measures the similarity between subjects through the inner product between their trained HMMs. Then, given the computational costs, fast computation of PPK for HMM facilitates parameter tuning of kernel similarity. Finally, the Kernel Principal Component Analysis (KPCA) projects the PPK to a lower dimensional space, allowing the interpretability of the results. Thus, a support vector machine supports the diagnosis of ADHD as a classification task using PPK as the inner product operator. The methodology compared classification results on EEG signals with all channels, channels of interest (COI), and analysis in the Theta, Alpha, and Beta frequency bands. The results show an accuracy rate of 97.0% in the Beta band in COI, which supports the assumption that this frequency rhythm may be correlated to differences between ADHD and controls regarding attentional allocation during the execution of the cognitive task.
dc.descriptionEl trastorno por déficit de atención e hiperactividad (TDAH), que suele presentarse en la infancia, puede persistir en la vida adulta, obstaculizando el desarrollo personal. Sin embargo, el diagnóstico del TDAH es un verdadero reto, ya que depende en gran medida de la observación clínica del paciente, de la información de los padres y de los estudiosos, y de la experiencia de los especialistas. A pesar de la demanda de ayudas para el diagnóstico objetivo a partir de bioseñales, los biomarcadores fisiológicos carecen de robustez y significación bajo la dinámica electroencefalográfica no estacionaria y no lineal. Por lo tanto, este trabajo presenta una metodología de diagnóstico apoyada para el TDAH a partir de la caracterización dinámica del EEG basada en modelos ocultos de Markov (HMM) y productos de kernel de probabilidad (PPK). Basándose en el síntoma de impulsividad, el enfoque propuesto entrena un HMM para cada sujeto a partir de las señales del EEG en tareas de inhibición fallidas. En primer lugar, el PPK mide la similitud entre los sujetos a través del producto interno entre sus HMMs entrenados. Luego, dados los costes computacionales, el cálculo rápido de PPK para los HMM facilita el ajuste de los parámetros de similitud del kernel. Por último, el Análisis de Componentes Principales del Kernel (KPCA) proyecta el PPK a un espacio de menor dimensión, lo que permite la interpretabilidad de los resultados. Así, una máquina de vectores de apoyo apoya el diagnóstico del TDAH como una tarea de clasificación utilizando el PPK como operador de producto interno. La metodología comparó los resultados de clasificación en señales de EEG con todos los canales, canales de interés (COI), y análisis en las bandas de frecuencia Theta, Alpha, y Beta. Los resultados muestran una tasa de precisión del 97,0% en la banda Beta en COI, lo que apoya la suposición de que este ritmo de frecuencia puede estar correlacionado con las diferencias entre el TDAH y los controles en cuanto a la asignación atencional durante la ejecución de la tarea cognitiva.
dc.descriptionMaestría
dc.descriptionMagíster en Ingeniería Eléctrica
dc.descriptionContents 1 List of Symbols and Abbreviations 5 1.1 Symbols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.2 Abbrevations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 2 Introduction 7 2.1 Problem statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 2.2 Justification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 2.3 State of the art . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 2.4 Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 2.4.1 General objective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 2.4.2 Specific objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 3 Develop a multichannel time series classification methodology taking into account signal dynamics 13 3.1 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 3.1.1 Similarity between time series . . . . . . . . . . . . . . . . . . . . . . 13 3.2 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 3.2.1 EEG Dataset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 3.2.2 HMM training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 3.2.3 Parameter tuning and Classification . . . . . . . . . . . . . . . . . . . 17 3.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 3.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 4 Develop a time series classification methodology that takes into account spectral information and reduces the computational cost of training. 21 4.1 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 4.1.1 Fast computation of PPK for HMM . . . . . . . . . . . . . . . . . . . 22 4.2 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 4.2.1 Synthetic Database . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 4.2.2 Training and Parameter tuning and classification . . . . . . . . . . . 24 4.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 1 CONTENTS 4.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 5 Develop a methodology for visualizing stochastic representations to facilitate the interpretability of inference machines 32 5.1 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 5.1.1 Model interpretability . . . . . . . . . . . . . . . . . . . . . . . . . . 32 5.1.2 Low-dimensional HMM visualization . . . . . . . . . . . . . . . . . . 33 5.1.3 Low-dimensional state visualization . . . . . . . . . . . . . . . . . . 34 5.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 5.3 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 6 Conclusions 40
dc.format51 páginas
dc.formatapplication/pdf
dc.formatapplication/pdf
dc.languageeng
dc.publisherUniversidad Tecnológica de Pereira
dc.publisherFacultad de Ingenierías
dc.publisherPereira
dc.publisherMaestría en Ingeniería Eléctrica
dc.relationS. Khoshnoud, M. A. Nazari, and M. Shamsi, “Functional brain dynamic analysis of ADHD and control children using nonlinear dynamical features of EEG signals,” Journal of integrative neuroscience, 2018.
dc.relationV. del Barrio, “Diagnostic and Statistical Manual of Mental Disorders,” in Encyclopedia of Applied Psychology, Three-Volume Set, 2004.
dc.relationS. M. Snyder, T. A. Rugino, M. Hornig, and M. A. Stein, “Integration of an EEG biomarker with a clinician ’ s ADHD evaluation,” vol. 330, pp. 1–17, 2015.
dc.relationM. Ahmadlou and H. Adeli, “Wavelet-synchronization methodology: a new approach for EEG-based diagnosis of ADHD,” Clinical EEG and Neuroscience, vol. 41, no. 1, pp. 1–10, 2010
dc.relationL. Marquardt, H. Eichele, A. J. Lundervold, J. Haavik, and T. Eichele, “Event-related potential (ERP) correlates of performance monitoring in adults with attention-deficit hyperactivity disorder (ADHD),” Frontiers in psychology, vol. 9, p. 485, 2018
dc.relationJ. F. Saad, M. R. Kohn, S. Clarke, J. Lagopoulos, and D. F. Hermens, “Is the Theta/Beta EEG Marker for ADHD Inherently Flawed?” Journal of Attention Disorders, vol. 22, no. 9, pp. 815–826, 2018.
dc.relationM. J. Groom, G. Scerif, P. F. Liddle, M. J. Batty, E. B. Liddle, K. L. Roberts, J. D. Cahill, M. Liotti, and C. Hollis, “Effects of motivation and medication on electrophysiological markers of response inhibition in children with attention-deficit/hyperactivity disor der,” Biological Psychiatry, vol. 67, no. 7, pp. 624–631, 2010.
dc.relationS. Whitmont, R. Meares, E. Gordon, I. Lazzaro, and S. Clarke, “The Modulation of Late Component Event Related Potentials by Pre-Stimulus EEG Theta Activity in ADHD,” International Journal of Neuroscience, vol. 107, no. 3-4, pp. 247–264, 2008.
dc.relationJ. Tornero Lucas, “Machine Learning: Modelos Ocultos de Markov (HMM) y Redes Neuronales Artificiales (ANN),” p. 53, 2017. [Online]. Available: http://diposit.ub.edu/dspace/bitstream/2445/122446/2/memoria.pd
dc.relationB.-H. Juang and L. R. Rabiner, “A probabilistic distance measure for hidden markov models,” AT&T technical journal, vol. 64, no. 2, pp. 391–408, 1985
dc.relationC. Lu, J. M. Schwier, R. M. Craven, L. Yu, R. R. Brooks, and C. Griffin, “A normalized statistical metric space for hidden Markov models,” in IEEE Transactions on Cybernet ics, 2013
dc.relationM. Falkhausen, H. Reininger, and D. Wolf, “Calculation of distance measures between hidden Markov models.” Simulation, 1995.
dc.relationJ. Zeng, J. Duan, and C. Wu, “A new distance measure for hidden Markov models,” Expert Systems with Applications, 2010.
dc.relationE. Epaillard and N. Bouguila, “Data-free metrics for dirichlet and generalized dirichlet mixture-based hmms – a practical study,” Pattern Recognition, vol. 85, pp. 207 – 219, 2019. [Online]. Available: http://www.sciencedirect.com/science/ article/pii/S003132031830311X
dc.relationP. Ruiz, C. Philbrick, and P. Sauer, “Modeling approaches for computational cost re duction in stochastic unit commitment formulations,” Power Systems, IEEE Transac tions on, vol. 25, pp. 588–589, 02 2010.
dc.relationK. Sadatnezhad, R. Boostani, and A. Ghanizadeh, “Classification of BMD and ADHD patients using their EEG signals,” Expert Systems with Applications, vol. 38, no. 3, pp. 1956–1963, 2011.
dc.relationB. Abibullaev and J. An, “Decision support algorithm for diagnosis of ADHD using electroencephalograms,” Journal of medical systems, vol. 36, no. 4, pp. 2675–2688, 2012.
dc.relationS. Itani, M. Rossignol, F. Lecron, and P. Fortemps, “Towards interpretable machine learning models for diagnosis aid: a case study on attention deficit/hyperactivity disorder,” PloS one, vol. 14, no. 4, p. e0215720, 2019.
dc.relationB. E. Bejnordi, M. Veta, P. J. Van Diest, B. Van Ginneken, N. Karssemeijer, G. Litjens, J. A. Van Der Laak, M. Hermsen, Q. F. Manson, M. Balkenhol et al., “Diagnostic assessment of deep learning algorithms for detection of lymph node metastases in women with breast cancer,” Jama, vol. 318, no. 22, pp. 2199–2210, 2017
dc.relationM. A. Ahmad, C. Eckert, and A. Teredesai, “Interpretable machine learning in health care,” in Proceedings of the 2018 ACM international conference on bioinformatics, com putational biology, and health informatics, 2018, pp. 559–560.
dc.relationM. Fredriksen, A. A. Dahl, E. W. Martinsen, O. Klungsoyr, S. V. Faraone, and D. E. Peleikis, “Childhood and persistent ADHD symptoms associated with educational fail ure and long-term occupational disability in adult ADHD.” Attention deficit and hyper activity disorders, vol. 6, no. 2, pp. 87–99, jun 2014.
dc.relationJ. Biederman, S. V. Faraone, T. J. Spencer, E. Mick, M. C. Monuteaux, and M. Aleardi, “Functional impairments in adults with self-reports of diagnosed ADHD: A controlled study of 1001 adults in the community.” The Journal of clinical psychiatry, vol. 67, no. 4, pp. 524–540, apr 2006
dc.relationM. Brod, B. Pohlman, R. Lasser, and P. Hodgkins, “Comparison of the burden of illness for adults with ADHD across seven countries: a qualitative study.” Health and quality of life outcomes, vol. 10, p. 47, may 2012.
dc.relationA. Cussen, E. Sciberras, O. C. Ukoumunne, and D. Efron, “Relationship between symptoms of attention-deficit/hyperactivity disorder and family functioning: a community-based study.” European journal of pediatrics, vol. 171, no. 2, pp. 271–280, feb 2012
dc.relationK. L. Humphreys and S. S. Lee, “Risk Taking and Sensitivity to Punishment in Chil dren with ADHD, ODD, ADHD+ODD, and Controls.” Journal of psychopathology and behavioral assessment, vol. 33, no. 3, pp. 299–307, 2011
dc.relationC. Gomez-Restrepo, C. Escudero, D. Matallana, L. Gonz ´ alez, and V. Rodriguez, “En- ´ cuesta Nacional de Salud Mental,” MInsalud. Colciencias, vol. 1, 2015
dc.relationC. Hidalgo-Lopez, A. M. G ´ omez- ´ Alzate, J. Garc ´ ´ıa-Valencia, and J. D. Palacio-Ortiz, “Riesgo de trastorno por d´eficit de atencion e hiperactividad y otros trastornos ´ psiquiatricos de los hermanos de pacientes con TDAH,” ´ Revista Colombiana de Psiquiatr´ıa, vol. 48, no. 1, pp. 44–49, 2019
dc.relationS. Mnookin, “Out of The Shadows: Making Mental Health a Global Development Priority,” Ecosystems and Human Well-being: A Framework for Assessment, 2016.
dc.relationCONSEJO NACIONAL DE POL´ITICA ECONOMICA Y SOCIAL CON- ´ PES, “ESTRATEGIA PARA LA PROMOCION DE LA SALUD MENTAL ´ EN COLOMBIA,” Bogota,Colombia, p. 70, 2020. [Online]. Available: ´ https://colaboracion.dnp.gov.co/CDT/Conpes/Econ{\unhbox\voidb@x\bgroup\ let\unhbox\voidb@x\setbox\@tempboxa\hbox{o\global\mathchardef\accent@ spacefactor\spacefactor}\let\begingroup\endgroup\relax\let\ignorespaces\relax\ accent19o\egroup\spacefactor\accent@spacefactor}micos/3992.pdf
dc.relationN. T. Mart´ınez, C. J. R. Rodr´ıguez, C. de Santacruz, N. B. Bautista, J. Collazos, and C. Gomez–Restrepo, “Problemas mentales, trastornos del afecto y de ansiedad en ´ la poblacion desplazada por la violencia en Colombia, resultados de la Encuesta Na- ´ cional de Salud Mental 2015,” Revista colombiana de psiquiatr´ıa, vol. 45, pp. 113–118, 2016.
dc.relationJ. M. Berg, R. D. Latzman, N. G. Bliwise, and S. O. Lilienfeld, “Parsing the hetero geneity of impulsivity: A meta-analytic review of the behavioral implications of the UPPS for psychopathology,” Psychological assessment, 2015.
dc.relationF. Di Michele, L. Prichep, E. R. John, and R. J. Chabot, “The neurophysiology of attention-deficit/hyperactivity disorder,” International Journal of Psychophysiology, vol. 58, no. 1, pp. 81–93, 2005.
dc.relationAmerican Psychiatric Association, Diagnostic and Statistical Manual of Mental Disor ders (5th Edition), 2013.
dc.relationW. Hospital and W. Hospital, “THE MODULATION OF LATE COMPONENT EVENT RELATED POTENTIALS BY PRE-STIMULUS EEG THETA ACTIVITY IN ADHD,” vol. 107, 2001.
dc.relationS. V. Faraone and C. Bonvicini, “Biomarkers in the Diagnosis of ADHD – Promising Directions,” 2014.
dc.relationR. J. Barry, A. R. Clarke, and S. J. Johnstone, “A review of electrophysiology in attention-deficit / hyperactivity disorder : I . Qualitative and quantitative electroen cephalography,” vol. 114, pp. 171–183, 2003
dc.relationJ. R. Wiersema, J. J. V. D. Meere, and H. Roeyers, “ERP correlates of impaired error monitoring in children with ADHD,” pp. 1417–1430, 2005.
dc.relationG. Deshpande, P. Wang, D. Rangaprakash, and B. Wilamowski, “Fully connected cas cade artificial neural network architecture for attention deficit hyperactivity disorder classification from functional magnetic resonance imaging data,” IEEE transactions on cybernetics, vol. 45, no. 12, pp. 2668–2679, 2015.
dc.relationA. Mueller, G. Candrian, V. A. Grane, J. D. Kropotov, V. A. Ponomarev, and G.-M. Baschera, “Discriminating between ADHD adults and controls using independent ERP components and a support vector machine: a validation study,” Nonlinear biomedical physics, vol. 5, no. 1, pp. 1–18, 2011.
dc.relationD. Fair, J. T. Nigg, S. Iyer, D. Bathula, K. L. Mills, N. U. F. Dosenbach, B. L. Schlag gar, M. Mennes, D. Gutman, and S. Bangaru, “Distinct neural signatures detected for ADHD subtypes after controlling for micro-movements in resting state functional connectivity MRI data,” Frontiers in systems neuroscience, vol. 6, p. 80, 2013.
dc.relationA. Allahverdy, A. M. Nasrabadi, and M. R. Mohammadi, “Detecting ADHD children us ing symbolic dynamic of nonlinear features of EEG,” in 2011 19th Iranian Conference on Electrical Engineering. IEEE, 2011, pp. 1–4.
dc.relationA. C. Kelly, A. Scheres, E. S. SONUGA-BARKE, and F. X. CASTELLANOS, “11 func tional neuroimaging of reward and motivational pathways in adhd,” Handbook of attention deficit hyperactivity disorder, p. 209, 2007.
dc.relationL. Chen and H. Man, “Fast schemes for computing similarities between gaussian hmms and their applications in texture image classification,” EURASIP Journal on Advances in Signal Processing, vol. 2005, no. 13, p. 164742, 2005.
dc.relationJ. Holmes, S. E. Gathercole, M. Place, T. P. Alloway, J. G. Elliott, and K. A. Hilton, “The diagnostic utility of executive function assessments in the identification of adhd in children,” Child and Adolescent Mental Health, vol. 15, no. 1, pp. 37–43, 2010.
dc.relationR. A. Barkley, “Neuropsychological testing is not useful in the diagnosis of adhd: Stop it (or prove it)!” The ADHD Report, vol. 27, no. 2, pp. 1–8, 2019.
dc.relationN. D. Volkow, G.-J. Wang, J. H. Newcorn, S. H. Kollins, T. L. Wigal, F. Telang, J. S. Fowler, R. Z. Goldstein, N. Klein, J. Logan et al., “Motivation deficit in adhd is as sociated with dysfunction of the dopamine reward pathway,” Molecular psychiatry, vol. 16, no. 11, pp. 1147–1154, 2011.
dc.relationE. B. Liddle, C. Hollis, M. J. Batty, M. J. Groom, J. J. Totman, M. Liotti, G. Scerif, and P. F. Liddle, “Task-related default mode network modulation and inhibitory control in adhd: Effects of motivation and methylphenidate,” Journal of Child Psychology and Psychiatry, vol. 52, no. 7, pp. 761–771, 2011.
dc.relationI. Marx, T. Hacker, X. Yu, S. Cortese, and E. Sonuga-Barke, “Adhd and the choice of small immediate over larger delayed rewards: a comparative meta-analysis of per formance on simple choice-delay and temporal discounting paradigms,” Journal of attention disorders, p. 1087054718772138, 2018.
dc.relationM. Arns, C. K. Conners, and H. C. Kraemer, “A decade of eeg theta/beta ratio research in adhd: a meta-analysis,” Journal of attention disorders, vol. 17, no. 5, pp. 374–383, 2013.
dc.relationJ. Van Doren, H. Heinrich, M. Bezold, N. Reuter, O. Kratz, S. Horndasch, M. Berking, T. Ros, H. Gevensleben, G. H. Moll et al., “Theta/beta neurofeedback in children with adhd: feasibility of a short-term setting and plasticity effects,” International Journal of Psychophysiology, vol. 112, pp. 80–88, 2017.
dc.rightsManifiesto (Manifestamos) en este documento la voluntad de autorizar a la Biblioteca Jorge Roa Martínez de la Universidad Tecnológica de Pereira la publicación en el Repositorio institucional (http://biblioteca.utp.edu.co), la versión electrónica de la OBRA titulada: ________________________________________________________________________________________________ ________________________________________________________________________________________________ ________________________________________________________________________________________________ La Universidad Tecnológica de Pereira, entidad académica sin ánimo de lucro, queda por lo tanto facultada para ejercer plenamente la autorización anteriormente descrita en su actividad ordinaria de investigación, docencia y publicación. La autorización otorgada se ajusta a lo que establece la Ley 23 de 1982. Con todo, en mi (nuestra) condición de autor (es) me (nos) reservo (reservamos) los derechos morales de la OBRA antes citada con arreglo al artículo 30 de
dc.rightsinfo:eu-repo/semantics/openAccess
dc.rightshttp://purl.org/coar/access_right/c_abf2
dc.rightsAtribución-NoComercial-SinDerivadas 4.0 Internacional (CC BY-NC-ND 4.0)
dc.rightshttps://creativecommons.org/licenses/by-nc-nd/4.0/
dc.subject620 - Ingeniería y operaciones afines::629 - Otras ramas de la ingeniería
dc.subjectElectroencephalography
dc.subjectMedical diagnosis
dc.subjectSystem kernels
dc.subjectProbability product kernel
dc.subjectHidden markov model
dc.subjectADHD
dc.titleSupported diagnosis of adhd from eeg signals based on hidden markov models and probability product kernels
dc.typeTrabajo de grado - Maestría
dc.typehttp://purl.org/coar/resource_type/c_bdcc
dc.typeText
dc.typeinfo:eu-repo/semantics/masterThesis


Este ítem pertenece a la siguiente institución