About
Dr. Rekha G R is an Assistant Professor in the School of Engineering at RV University. She holds an MCA from University of Mysore and a Ph.D. from JSS Science and Technology University, Mysuru, and has 4.5 years of industry experience. Her academic and professional interests lie at the intersection of Artificial Intelligence, Machine Learning, Computer Vision, and real-world applications of emerging technologies.
Her research interests include Pattern Recognition, Artificial Intelligence in Healthcare, Image Processing, Deep Learning, Computer Vision, and Machine Learning. Her doctoral research focused on developing deep learning-based approaches for the analysis and digitization of unstructured handwritten medical prescriptions, addressing challenges in layout detection, segmentation, instance segmentation, and handwritten text recognition.
She has published research articles in international journals and conference proceedings, including publications with PUBETA, BonView, and Springer. Her recent research work includes instance segmentation, treatment regimen segmentation, and content extraction from handwritten medical prescriptions using advanced deep learning and YOLO-based approaches. She has also presented research at international conferences in the areas of Computer Vision, Robotics, Computational Intelligence, and Data Communication.
She actively engages in continuous professional development through workshops and faculty development programmes in areas such as biometrics, scholarly publishing, LaTeX, intellectual property, and classical machine learning. She is a member of the Institute for Educational Research and Publication (IFERP) and the Soft Computing Research Society.
As an educator, she is passionate about experiential and application-oriented learning. She aims to inspire students to explore emerging technologies, develop innovative solutions to real-world problems, and cultivate strong analytical and problem-solving skills through teaching, research, mentoring, and industry-oriented learning.
SAHI-BAR: An Instance Segmentation Model for Medical Prescriptions Status: Published, 2026
This research presents SAHI-BAR, an instance segmentation approach developed for extracting and identifying important components from handwritten medical prescriptions. The work addresses challenges arising from complex layouts, variations in handwriting, and closely positioned prescription elements. The approach incorporates Slicing Aided Hyper Inference (SAHI) to improve detection of smaller and spatially distributed prescription components. The proposed framework supports structured extraction of medication-related information and contributes towards automated digitization of handwritten prescriptions.
Link: https://etasr.com/index.php/ETASR/article/view/16418
Treatment Regimen Segmentation from Handwritten Medical Prescriptions Using Advanced Neural Network Status: Published. 2026
This work focuses on treatment regimen segmentation in handwritten medical prescriptions using advanced neural network techniques. The study addresses the challenge of identifying and separating treatment-related information while preserving spatial and semantic relationships between prescription components. The proposed approach applies deep learning-based image segmentation techniques to handwritten prescription documents and facilitates subsequent recognition and structured extraction of medication information.
Link: https://doi.org/10.47852/bonviewAIA62027648
Categorization and Content Extraction in Medical Prescription Using YOLOv8 Status: Published – Springer 2026
This research investigates the application of YOLOv8 for categorizing and extracting relevant information from handwritten medical prescriptions. The study explores object detection and deep learning techniques for identifying different components within prescription documents. By treating prescription elements as distinct visual entities, the approach facilitates structured extraction of medication-related information and contributes towards automated prescription digitization.
Link: https://doi.org/10.1007/978-981-96-9554-6_33
A Deep Association-Guided Segmentation of Handwritten Medical Prescriptions Status: In Press – CVR 2026
This research presents a deep learning-based approach for segmentation of handwritten medical prescriptions, focusing on identifying and separating relevant prescription components from complex and unstructured document layouts. The work explores contextual and spatial associations to improve segmentation and support subsequent recognition and structured extraction of medication-related information. The paper is associated with the 6th International Conference on Computer Vision and Robotics (CVR 2026) and is currently in press.
Detection of Layouts in Unstructured Handwritten Medical Prescription using Deep Learning Approach Status: Accepted – November 2026
This work investigates deep learning-based layout detection for unstructured handwritten medical prescriptions. The study focuses on identifying different structural regions within prescription documents to establish a reliable foundation for subsequent segmentation and information extraction. The research contributes to an end-to-end framework for automated prescription analysis and demonstrates the role of deep learning in understanding complex document layouts.
Conference: Second International Conference on Computational Intelligence and Data Communication, 26–28 November 2026.
Few-Shot Learning for Binary Classification in Red Chilli Powder Status: Published 2026
This research explores the application of few-shot learning for binary classification of red chilli powder images. The study investigates how deep learning models can perform image classification when only a limited number of labelled samples are available. The work demonstrates the potential of few-shot learning for specialized image classification problems where collecting and annotating large datasets can be challenging.
Conference: 2nd Doctoral Research Conference in Engineering, Sciences and Management (DRC 2026), JSS Science and Technology University, Mysuru, India.
Artificial Intelligence & Machine Learning: Development and application of intelligent computational techniques for solving real-world problems.
Computer Vision: Research in image understanding, object detection, segmentation, and visual information extraction.
Handwritten Document Analysis: Deep learning-based analysis of complex and unstructured handwritten documents.
Medical Prescription Digitization: Automated extraction and recognition of medicine-related information from handwritten prescriptions.
Pattern Recognition: Identification and classification of meaningful patterns from complex visual and document data.
Deep Learning and Few-Shot Learning: Application of CNNs, object detection architectures, segmentation networks, and neural network-based approaches.
AI in Healthcare: Exploring Artificial Intelligence for intelligent healthcare document processing and information management.
Image Processing: Preprocessing, enhancement, segmentation, and analysis of challenging handwritten and document images.
Emerging AI Applications: Exploring practical applications of AI that bridge academic research and real-world technological needs.
Research Interests
Pattern Recognition, AI in Healthcare, Artificial Intelligence, Machine Learning, Image Processing, Deep Learning, Few-shot Learning, Computer Vision
Ph.D. in Computer Science
Completed Ph.D. from JSS Science and Technology University, Mysuru, with research focused on deep learning and computer vision techniques for handwritten medical prescription analysis and digitization
Research Publications
Published research in international journals, conference proceedings, and Springer publications covering Artificial Intelligence, Computer Vision, Deep Learning, and handwritten medical prescription analysis.
International Conference Presentations
Presented research at international conferences in Computer Vision, Robotics, Computational Intelligence, and Data Communication, contributing to research dissemination in AI-driven document analysis.
Continuous Professional Development
Completed workshops and faculty development programmes in biometrics, scholarly publications, LaTeX, intellectual property, and classical machine learning through recognized academic and professional institutions.
Assistant Professor
School of Engineering
-
Mysuru