Volume 22, Issue 2 (9-2025)
JSDP 2025, 22(2): 97-108 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Goshvarpour A. Classifying Various Stages of Typing Learning through EEG Rhodonea Curve Asymmetry Indices: A Focus on the Optimal Number of Petals and Brain Channels. JSDP 2025; 22 (2) : 6
URL: http://jsdp.rcisp.ac.ir/article-1-1376-en.html
URL: http://jsdp.rcisp.ac.ir/article-1-1376-en.html
Department of Biomedical Engineering, Imam Reza International University, Mashhad, Razavi Khorasan, Iran
Abstract: (34 Views)
This research aimed to provide a novel algorithm for analyzing biological data. Specifically, the proposed method was investigated for characterizing electroencephalography (EEG) data related to skill acquisition. EEG data from 10 participants (6 women and 4 men) were collected across 9 brain channels (F3, Fz, F4, C3, Cz, C4, P3, Poz, and P4) available on IEEE DataPort, while participants learned twelve typing lessons on a computer, using the Colemak keyboard, with each lesson repeated five times. EEG recordings were analyzed during the fourth, eighth, and eleventh lessons. For the first time, a Rhodonea curve-based method was introduced for signal analysis, characterized by a flower-like structure with an adjustable number of petals. This study evaluated models with one to ten petals. Three new indicators were developed based on the asymmetry of the Rhodonea curve and utilized with a support vector machine (SVM) to differentiate the stages of learning. The impact of cerebral channels and the optimal number of petals were assessed by evaluating the classification results for each channel and petal count separately. The binary classification results for distinguishing 15 sessions (5 repetitions x 3 lessons) ranged from 79.3 % to 93.3% utilizing one vs. all approach. The highest accuracy was observed in the F3, Fz, C3, C4, and POz channels with four petals. For the classification of the three sessions, the highest accuracy corresponded to the eleventh session (92%), followed by the fourth session (90%) and the eighth session (72.6%). The results indicate that during typing practice, specific regions of the frontal, parietal, and occipital lobes are activated. Furthermore, brain dynamics are more distinct during the completion of the learning process (11th session) and in the early stages of learning (4th session). The strong performance of the Rhodonea curve indices highlights its potential for EEG signal analysis.
Article number: 6
Keywords: Electroencephalography, Learning, Rhodonea curve, Signal processing, Classification, Asymmetry.
Type of Study: Research |
Subject:
Paper
Received: 2023/04/24 | Accepted: 2025/03/8 | Published: 2025/09/13 | ePublished: 2025/09/13
Received: 2023/04/24 | Accepted: 2025/03/8 | Published: 2025/09/13 | ePublished: 2025/09/13
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
