KADYSNET: A NOVEL APPROACH TO DYSLEXIA PREDICTION IN CHILDREN: COMBINING HANDWRITTEN TEXT RECOGNITION WITH A HYBRID CNN-BILSTM-CTC MODEL AND PERSONALIZED LEARNING STRATEGIES

Authors

Shailesh Prabhakar Patil, Department of Computer Engineering, Smt. Kashibai Navale College of Engineering, Savitribai Phule Pune University, Pune, 411041, IndiaFollow
Ravindra Sadashivrao Apare, Artificial Intelligence and Data Science Department, Pune Vidyarthi Griha's College of Engineering, Technology and Management, Savitribai Phule Pune University, Pune, 411009, India
Ravindra Honaji Borhade, Department of Computer Engineering, Smt. Kashibai Navale College of Engineering, Savitribai Phule Pune University, Pune, 411041, India
Parikshit Narendra Mahalle, Department of Artificial Intelligence and Data Science, Vishwakarma Institute of Technology, Pune, 411037, India.

Abstract

This study introduces a novel approach for early dyslexia detection in children through automated handwriting analysis, integrating a hybrid CNN-BiLSTM-CTC architecture with personalized learning strategies. Our method combines a custom CNN-BiLSTM-CTC model with tailored educational interventions to support dyslexic learners. We analyzed children's handwritten text images in English, collected from specially designed tests involving word rewriting, sentence reconstruction, and paragraph composition, particularly challenging tasks for dyslexic individuals. Notably, our CNN-BiLSTM-CTC model achieved the best result with an accuracy of 97.67%, outperforming other architectures. Compared to pre-trained models like EfficientNetB7, DenseNet121, and MobileNetV2, our custom CNN-BiLSTM-CTC model demonstrated superior performance. Key contributions include the development of a novel CNN-BiLSTM-CTC architecture for handwriting analysis and the integration of personalized learning strategies to enhance educational outcomes. By facilitating earlier and more accurate detection, this approach can significantly improve educational support for children at risk of dyslexia. Future research will focus on expanding the dataset and conducting longitudinal studies to assess the long-term impact. Customization and CNN-BiLSTM hybridization together enhance performance by capturing subtle handwriting variations and integrating spatial with sequential learning.

Keywords

CNN; Bi-LSTM; LD; HTR; CTC

Publication Date

2025

Received Date

27/05/2025

Revised Date

12/09/2025

Accepted Date

13/09/2025

Recommended Citation

Patil, Shailesh Prabhakar; Apare, Ravindra Sadashivrao; Borhade, Ravindra Honaji; and Mahalle, Parikshit Narendra (2025) "KADYSNET: A NOVEL APPROACH TO DYSLEXIA PREDICTION IN CHILDREN: COMBINING HANDWRITTEN TEXT RECOGNITION WITH A HYBRID CNN-BILSTM-CTC MODEL AND PERSONALIZED LEARNING STRATEGIES," ASEAN Journal on Science and Technology for Development: Vol. 43: No. 1, Article 13.
DOI: https://doi.org/10.61931/2224-9028.1639
Available at: https://ajstd.ubd.edu.bn/journal/vol43/iss1/13