Transactions on Machine Intelligence

Transactions on Machine Intelligence

Leveraging Bagging Ensemble Architectures for the Automated Diagnosis of Dyslexia via Visual Task Paradigm Analysis

Document Type : Original Article

Authors
M. Zaree 1
B. Arefi-Rad 2
M. Mohebbi 1
R. Rostami 3
1 Department of Biomedical Engineering, K. N. Toosi University of Technology, Tehran, Iran
2 Department of Electrical Engineering, K. N. Toosi University of Technology, Tehran, Iran
3 Department of Psychology, University of Tehran, Tehran, Iran
10.22034/tmi.2025.244851
Abstract
Dyslexia is a neurobiological learning disability that fundamentally impairs a child’s literacy acquisition, specifically manifesting as persistent deficits in reading and writing. Absent a timely diagnosis, the disorder precipitates profound psychological distress and academic marginalization for both the pediatric patients and their families. Furthermore, delayed intervention often results in cumulative achievement gaps that become increasingly difficult to bridge by secondary education. Consequently, early screening and clinical intervention are paramount to preserving student self-esteem and optimizing long-term academic trajectories. This study proposes an automated diagnostic framework utilizing a Bagging (Bootstrap Aggregating) ensemble learning approach to classify dyslexia in children. The methodology involves the rigorous preprocessing of electroencephalogram (EEG) signals recorded across a 19-channel montage. Feature extraction focused on the morphometry of Event-Related Potentials (ERPs), specifically quantifying the amplitude and latency of key components. To address the "curse of dimensionality" inherent in the high-dimensional feature set, Principal Component Analysis (PCA) was implemented for optimal feature reduction. To ensure the generalizability of the model and mitigate the risk of overfitting, a K-fold cross-validation strategy was employed during the training phase. Finally, the Bagging classifier was deployed to distinguish between dyslexic and neurotypical subjects. The proposed ensemble framework demonstrated robust performance, yielding an average classification accuracy of 90.6%. Notably, the model achieved a sensitivity rate of 100%, ensuring no dyslexic cases were omitted, and a specificity of 81.2%, reflecting its capability to accurately identify neurotypical controls.
Keywords
Dyslexia
Electroencephalography (EEG)
Ensemble Learning
Machine Learning
Event-Related Potential (ERP)
[1]      Gagliardi, L., Rusconi, F., Bellù, R., Zanini, R., & Italian Neonatal Network. (2014). Association of maternal hypertension and chorioamnionitis with preterm outcomes. Pediatrics, 134(1), e154–e161. https://doi.org/10.1542/peds.2013-3898.
[2]      Zahia, S., Garcia-Zapirain, B., Saralegui, I., & Fernandez-Ruanova, B. (2020). Dyslexia detection using 3D convolutional neural networks and functional magnetic resonance imaging. Computer Methods and Programs in Biomedicine, 197, 105726. https://doi.org/10.1016/j.cmpb.2020.105726
[3]      Rauschenberger, M., Rello, L., Baeza-Yates, R., & Bigham, J. P. (2018, April). Towards language independent detection of dyslexia with a web-based game. In Proceedings of the 15th International Web for All Conference (pp. 1–10). https://doi.org/10.1145/3192714.3192816
[4]      Perera, H., Shiratuddin, M. F., & Wong, K. W. (2016). A review of electroencephalogram-based analysis and classification frameworks for dyslexia. In Neural Information Processing: 23rd International Conference, ICONIP 2016, Proceedings, Part IV (pp. 626–635). Springer. https://doi.org/10.1007/978-3-319-46681-1_74
[5]      Kohli, M., & Prasad, T. V. (2010, June). Identifying dyslexic students by using artificial neural networks. In Proceedings of the World Congress on Engineering (Vol. 1, pp. 1–4). WCE.
[6]      Martins, V. F., Lima, T., Sampaio, P. N., & de Paiva, M. (2016, November). Mobile application to support dyslexia diagnostic and reading practice. In 2016 IEEE/ACS 13th International Conference of Computer Systems and Applications (AICCSA) (pp. 1–6). IEEE. https://doi.org/10.1109/AICCSA.2016.7945710
[7]      Perera, H., Shiratuddin, M. F., Wong, K. W., & Fullarton, K. (2017, June). EEG signal analysis of real-word reading and nonsense-word reading between adults with dyslexia and without dyslexia. In 2017 IEEE 30th International Symposium on Computer-Based Medical Systems (CBMS) (pp. 73–78). IEEE. https://doi.org/10.1109/CBMS.2017.108
[8]      Frid, A., & Breznitz, Z. (2012, November). An SVM based algorithm for analysis and discrimination of dyslexic readers from regular readers using ERPs. In 2012 IEEE 27th Convention of Electrical and Electronics Engineers in Israel (pp. 1–4). IEEE. https://doi.org/10.1109/EEEI.2012.6377068
[9]      Cui, Z., Xia, Z., Su, M., Shu, H., & Gong, G. (2016). Disrupted white matter connectivity underlying developmental dyslexia: A machine learning approach. Human Brain Mapping, 37(4), 1443–1458. https://doi.org/10.1002/hbm.23112
[10]   Meissner, N. A. (2018). A single-subject evaluation of facilitated communication in the completion of school-assigned homework [Master's thesis, University of Wisconsin-Eau Claire]. Minds@UW.
[11]   Heshmatollah, A., Dommershuijsen, L. J., Fani, L., Koudstaal, P. J., Ikram, M. A., & Ikram, M. K. (2021). Long-term trajectories of decline in cognition and daily functioning before and after stroke. Journal of Neurology, Neurosurgery & Psychiatry, 92(11), 1158–1163. https://doi.org/10.1136/jnnp-2021-326043
[12]   Kropotov, J. D., & Ponomarev, V. A. (2009). Decomposing N2 NOGO wave of event-related potentials into independent components. Neuroreport, 20(18), 1592–1596. https://doi.org/10.1097/WNR.0b013e3283309cbd
[13]   Daffertshofer, A., Lamoth, C. J., Meijer, O. G., & Beek, P. J. (2004). PCA in studying coordination and variability: A tutorial. Clinical Biomechanics, 19(4), 415–428. https://doi.org/10.1016/j.clinbiomech.2004.01.005
[14]   Breiman, L. (1996). Bagging predictors. Machine Learning, 24(2), 123–140. https://doi.org/10.1007/BF00058655
[15]   Altman, N., & Krzywinski, M. (2017). Ensemble methods: Bagging and random forests. Nature Methods, 14(10), 933–935. https://doi.org/10.1038/nmeth.4438
[16]   AIML. (2022). What is bagging? https://aiml.com/what-is-bagging/.
Transactions on Machine Intelligence
Volume 8, Issue 2
Spring 2025
Pages 120-127

  • Receive Date 23 January 2025
  • Revise Date 24 April 2025
  • Accept Date 08 June 2025