PROGRAMME OVERVIEW
3rd International Conference on Advances in Science, Engineering & Technology(ICASET-2025)
PROGRAMME OVERVIEW
Day 1 - 22nd March 2025
Registration Desk
National Anthem of India
Welcome Speech by Moderator
Welcome Speech by Special Guest of Honor
Speech by Keynote Speaker
Prof. Dr. Subhas Chandra Mukhopadhyay
Professor of Mechanical / Electronics Engineering
Macquarie University,
Australia.
Speech by Session Speaker
Assoc Prof Dr. Tiruveedula Gopi Krishna
Adama Science and Technology University,
Ethiopia.
Speech by Keynote Speaker
Assoc Prof . Dr. Nurnadiah Zamri
Senior Lecturer
Faculty of Informatics and Computing,
University of Sultan Zainal Abidin,
Terengganu, Malaysia.
Photographic Session
Refreshment Break
Speech by Session Speaker
Asst Prof Dr. Madhurima Dasgupta
School of Humanities, Management and Social Sciences,
The Neotia University,
India.
Technical Session I
Lunch break
Speech by Session Speaker
Prof. Dr Karuppasamy Periyasamy
Department of Electronics & Communication Engineering
Adithya Institute of Technology,
Coimbatore,India.
Technical Session II
Refreshment Break
Valedictory
Vote of Thanks
Day 2 - 23rd March 2024
Welcome Speech by Moderator
Welcome Speech by
Speech by Exclusive Event Speaker
Speech by Keynote Speaker
Technical Session I A
Technical Session I B
Speech by Session Speaker
Prof. Dr.Tanuja Satish Dhope (Shendkar)
Department of Electronics and Communication Engineering,
Bharati Vidhyapeeth (Deemed to be University) College of Engineering,
Pune, India.
Lunch Break
Speech by Session Speaker
Mr. Hardik Ruparel
Nutanix | Founder- Stealth Project
United States.
Speech by Session Speaker
Prof. Dr Amit Kumar Marwah
Head of the Department of Mechanical Engineering
Acropolis Institute of Technology and Research,
Indore (MP), India.
Technical Session II A
Technical Session II B
Speech by Session Speaker
Assoc Prof. Dr. Ku Nurul Fazira Ku Azir
Faculty of Electronic Engineering & Technology
Head Centre of Excellence for Advanced Computing
Universiti Malaysia Perlis,
Malaysia.
Technical Session III A
Technical Session III B
Valedictory
Vote of Thanks
Technical Session
Paper Title :Small Object Detection in Autonomous Vehicles: Evaluating YOLOv10 with a Custom Dataset Against YOLOv8
Name : Archana chintapatla
Technical session : Computer Science & Artificial Intelligence
Abstract :
This research aims to enhance object detection accuracy in autonomous vehicle systems by employing the YOLOv10 model. Building upon the foundations of the YOLOv8 model presented in the base paper, this investigation seeks to surpass the detection performance achieved by YOLOv8, particularly for small objects in complex driving environments. The YOLOv10 model is trained on a custom dataset obtained from Roboflow, with adjustments implemented to optimize its performance and accuracy. Experimental results demonstrate a significant increase in mean Average Precision and overall detection accuracy compared to the baseline YOLOv8 model. These findings suggest that YOLOv10 provides a more effective solution for the real-time detection of small and distant objects in autonomous driving applications.
Keywords:
YOLOv10, object detection, autonomous vehicles, small object detection, accuracy improvement, deep learning, real-time detection, dataset optimization, machine learning.
Paper Title :Rice Leaf Classification Using CNN and GRU
Name : Ansik Aryan Samal
Technical session : Computer Science & Artificial Intelligence
Abstract :
To overcome this limitation, the paper proposes a hybrid model that leverages the strengths of CNNs for spatial feature extraction and GRUs for modeling sequential dependencies In the proposed approach, the CNN component first processes the images to extract relevant features. These features are then fed into the GRU component, which models the sequential dependencies between the features. This integration allows the model to benefit from both the spatial analysis capabilities of CNNs and the sequential modeling capabilities of GRUs potentially improving classification performance in scenarios where temporal patterns are important. To evaluate the effectiveness of this hybrid model the authors studied and tested it on a list of rice leaves used to classify and predict leaf diseases. The evaluation process consists of several key steps: - Feature Extraction with CNNs - Sequential Modeling with GRUs - Model Training and Testing - Performance Evaluation - Parameter Analysis - Visualization
Keywords:
CNN, GRU, neural networks, maxpooling, dense layer, dropout layer, rice leaf diseases, Classification.
Paper Title :Precision-Driven Real-Time Pose Estimation for Therapeutic Interventions: Advanced Heatmap Regression, Reference Video Alignment, and Real-Time Corrective Feedback
Name : Sarvesh Kumar
Technical session : Computer Science & Artificial Intelligence
Abstract :
Accurate movement and posture are essential for effective physical therapy, as improper form can hinder recovery and worsen injuries. This project introduces a real-time human pose estimation system specifically designed for physical therapy, providing precise feedback on body alignment. Utilizing a mod- ified YOLOv8 architecture with custom heatmap regression, the system monitors key joints—particularly the wrist, elbow, and shoulder—vital for upper-body rehabilitation. Initially trained on a combined MPII and COCO 2017 dataset, the model was fine-tuned on a custom dataset of 6,000 images derived from 1,250 video frames under varied lighting conditions, with a 380% augmentation rate to improve robustness across scenarios. Achieving a detection accuracy of 91.61%, the system surpasses widely used models like OpenPose and MediaPipe, which deliver accuracies of 85% and 88%, respectively. With an average frame rate of 27.94 FPS and latency of 19.24 milliseconds per frame, the system provides instant feedback, enabling users to adjust posture in real time. Personalized guidance is offered by calculating the distance between live and reference keypoints, maintaining a mean keypoint detection error under 5 pixels. This real- time corrective feature enhances rehabilitation by empowering users to self-adjust and allowing healthcare providers to track progress effectively. By focusing on physical therapy-specific movements, this system represents a significant advancement in integrating AI-driven solutions into rehabilitation, enhancing both effectiveness and accessibility.
Keywords:
Physical therapy, Real-time pose estimation, heatmap regression, yolo-v8, keypoint detection,Corrective feed- back
Paper Title :Clinical Impact and Workflow Integration of Artificial Intelligence for Breast Cancer Detection and Diagnosis: A Scoping Review
Name : Gurjeet Kaur
Technical session : Computer Science & Artificial Intelligence
Abstract :
This scoping review explored the application of Artificial Intelligence (AI) in diagnosing and detecting breast cancer, examining various AI techniques and their effectiveness in enhancing early detection. By analysing current trends and advancements, the review highlights the potential of AI to improve breast cancer screening and identifies research gaps for better integration into clinical workflows. A comprehensive search across databases such as PubMed, Google Scholar, and MEDLINE yielded 425 articles, of which 47 met the inclusion criteria. Among these, deep learning (DL) techniques were most prevalent (50%), followed by machine learning (30%) and hybrid models (20%). Convolutional neural networks were commonly employed for image analysis, with accuracy rates exceeding 90% in mammography interpretation. Metrics like accuracy (70%), sensitivity (50%), and specificity (40%) were frequently reported. AI’s integration with traditional imaging methods enhanced detection rates, reduced false positives, and decreased radiologists’ reading times. The findings underscore AI’s significant role in breast cancer care, particularly in early detection, risk prediction, and diagnostic precision. Despite its promising performance, the study calls for improved assessment methodologies to maximize AI’s potential in clinical workflows, emphasizing its utility in enhancing screening efficiency and overall patient outcomes.
Keywords:
breast cancer, early detection, artificial intelligence, machine learning, diagnostic imaging
Paper Title :A Comprehensive Review of the Future of Medical Learning with Artificial Intelligence
Name : Siddhi Rathi
Technical session : Computer Science & Artificial Intelligence
Abstract :
Artificial Intelligence (AI) is revolutionizing health education by enhancing accessibility, personalization, and efficiency in learning. This paper explores how AI technologies, including machine learning, natural language processing, and predictive analytics, are transforming traditional health education methodologies. Key innovations include adaptive learning platforms, virtual patient simulations, and AI-driven decision-support tools that enhance clinical training and knowledge retention. AI fosters personalized learning experiences by tailoring content to individual needs, enabling real-time feedback, and promoting competency-based education. Additionally, it supports interdisciplinary collaboration by bridging gaps between medical specialties through shared data insights. The integration of AI into health education also addresses challenges like faculty shortages, uneven resource distribution, and the need for continuous skill updating in dynamic healthcare environments. Ethical considerations, including data privacy and equity in access to AI tools, remain critical for responsible implementation. By reshaping how health professionals are trained, AI not only prepares learners for complex clinical scenarios but also contributes to improved patient outcomes. This paper emphasizes the transformative potential of AI in creating innovative, inclusive, and efficient health education systems, setting a foundation for future advancements in the healthcare workforce.
Keywords:
Artificial Intelligence, health education, personalized learning, adaptive technology, virtual simulation, clinical training, AI ethics, healthcare innovation.
Paper Title :Bridging Visuals and Words by Automating Image Captioning and Translation
Name : Padmavati E Gundgurti
Technical session : Computer Science & Artificial Intelligence
Abstract :
This AI-based system automates image captioning, multilingual translation, and text-to-speech synthesis using state-of-the-art deep learning techniques. For image captioning, it employs the BLIP (Bootstrapped Language-Image Pretraining) model, enabling conditional generation of highly descriptive captions. These captions are translated into various Indian languages, including Hindi, Tamil, and Telugu, using the Deep Translator powered by the Google Translator API. To enhance accessibility, the system integrates Google Text-to-Speech (gTTS), converting both English and translated captions into audio files. A user-friendly web interface, developed with Gradio, allows users to input images via upload or webcam, select their desired output language, and receive both textual and audio outputs. This versatile system supports applications such as assistive technologies for visually impaired users, multilingual content localization, and efficient image indexing for diverse audiences.
Paper Title :Study of Tagset for Natural Languages with Special Reference to Marathi Language
Name : Swati Prakash Sonawane
Technical session : Computer Science & Artificial Intelligence
Abstract :
The study of tagsets for natural languages is central to the development of computational tools in Natural Language Processing (NLP), particularly in tasks like Part-of-Speech (POS) tagging. POS tagging assigns grammatical categories to each word in a sentence, which is essential for various NLP applications such as syntactic parsing, machine translation, and information retrieval. For languages with complex morphology, such as Marathi, the task becomes increasingly intricate. This paper focuses on understanding the tagsets used for Marathi POS tagging, examining their linguistic intricacies and the challenges involved. A detailed overview of popular Marathi tagsets like Universal Dependencies (UD), the Marathi Syntax Treebank (MST), and the Indian Language Treebank (ILCI) is provided, highlighting their advantages and shortcomings. Additionally, the study delves into the significance of corpus development, the impact of word segmentation, and the importance of context in accurate tagging. Through this analysis, we aim to provide insights into how Marathi-specific characteristics influence POS tagging and the evolution of tagset frameworks for morphologically rich languages.
Keywords:
Marathi language, Part-of-Speech tagging, Universal Dependencies, Tagsets, Natural Language Processing, Morphological analysis.
Paper Title :Data-Driven Approach for Kidney Disease Prediction
Name : Kavita Patil
Technical session : Computer Science & Artificial Intelligence
Abstract :
Chronic Kidney Disease is a very significant health problem affecting millions of patients every year and requires early diagnosis for good treatment results. The review will analyze a data-driven approach for the prediction of kidney diseases: machine learning methods. Recent studies on early-stage detection and prediction of progression using various algorithms define it as a success rate of over 95%: Support-Vector-Machines, Random Forests, Logistics Regression, and Decision Tree. Cited works in this analysis will be from vast literature review types of validation in various types of the model. While traditional machine learning methods provide clear predictions, deep learning and hybrid evolving models add higher accuracy and generalizability with broader distributions of input. These techniques hold particular promise in resource-limited settings where conventional diagnostic protocols may be impractical or hard to administer.
Keywords:
Chronic Kidney Disease, Support Vector
Paper Title :Analyzing the Issues of Thyroid Disease Diagnosis Using Machine Learning and Deep Learning: A Review
Name : Kalpana K Harish
Technical session : Computer Science & Artificial Intelligence
Abstract :
In day-to-day life millions of people throughout the world are afflicted with thyroid disease such as Hypothyroid, Hyperthyroid, Thyroid Cancer, thyroiditis, Hashimoto’s thyroiditis, goiter etc. These diseases may occur due to the deficiency of Vitamin D. Among all these thyroid deceases more people afflicted with hypothyroid, goiter and hyperthyroid. To deal with, these diseases many researchers has proposed various models using different methodologies such as Machine Learning, Deep Learning, Soft Computing techniques, Data mining, optimizations etc. From Machine learning methodologies like LR, DT, and random forest so on, the logistic regression is providing best accuracy. Further, from deep learning techniques namely CNN, ANN, RNN etc., the CNN is offering the best accuracy.
Keywords:
Thyroid, Computer-Aided-Design, Artificial neural networks, Machine Learning, Deep Learning