Harsha Bhute
Harsha Bhute has a total of 32 co-authors. Here are the top co-authors by number of shared publications:
- Avinash Bhute
Pimpri Chinchwad College of Engineering
4 shared publications - Bharati P. Vasgi
Department of Information Technology Marathwada Mitra Mandal's College of Engineering Pune Maharashtra India
2 shared publications - Harsha Jitendra Sarode
Jagdishprasad Jhabarmal Tibrewala University Rajasthan India
2 shared publications - Manohar K. Kodmelwar
Department of Information Technology Vishwakarma Institute of Technology Pune Maharashtra India
2 shared publications Akhila Sanga
Pimpri Chinchwad College of Engineering
1 shared publicationAniket Patil
Pimpri Chinchwad College of Engineering Pune
1 shared publication
Latest Publications
- Open Access
- 14 July 2026
- Published by Atlantis Press SARL
- p. 247-259
Vision AI Driven Multimodal Assistive System with Memory Recall For Visually Impaired
- 14 July 2026
- Published by Atlantis Press SARL
- p. 247-259
Open AccessVision AI Driven Multimodal Assistive System with Memory Recall For Visually Impaired
- 5 February 2026
- Published by Elsevier in Indian Journal of Tuberculosis
- Vol. 73, S58-S61
Mobile-based health education interventions for improving tuberculosis knowledge in rural communities
- 5 February 2026
- Published by Elsevier in Indian Journal of Tuberculosis
- Vol. 73, S58-S61
Mobile-based health education interventions for improving tuberculosis knowledge in rural communities
- 2 January 2026
- Published by Springer Nature
- p. 487-501
Optimized Image Caption Generation for Social Media Using CNN and LSTM
- 2 January 2026
- Published by Springer Nature
- p. 487-501
Optimized Image Caption Generation for Social Media Using CNN and LSTM
- 28 November 2025
- Published by Elsevier in Indian Journal of Tuberculosis
- Vol. 72, S24-S30
Deep learning for automated sputum smear microscopy in tuberculosis diagnosis
- 28 November 2025
- Published by Elsevier in Indian Journal of Tuberculosis
- Vol. 72, S24-S30
Deep learning for automated sputum smear microscopy in tuberculosis diagnosis
- 17 October 2025
- Published by Elsevier in Indian Journal of Tuberculosis
- Vol. 72, S18-S23
Federated learning framework for predicting multi-drug resistant tuberculosis across regional databases
- 17 October 2025
- Published by Elsevier in Indian Journal of Tuberculosis
- Vol. 72, S18-S23
Federated learning framework for predicting multi-drug resistant tuberculosis across regional databases
- Open Access
- 3 October 2025
- Published by Academic Science Publications and Distributions in International Journal of Environmental Sciences
- p. 1594-1602
Next-Gen EHR: Advancing Healthcare with Integrated Recommendations and Decision Support
The study outlines a holistic strategy for modernizing healthcare administration through the fusion of electronic health records (EHR) and machine learning techniques, including Support Vector Machines (SVM) and Naive Bayes. It addresses challenges in manual patient record management, particularly in countries like India, by proposing an EHR system implementation. The developed web application combines frontend and backend development technologies to streamline record management, enhance data accuracy, and improve healthcare delivery. Additionally, leveraging Naive Bayes and SVM algorithms, the system prioritizes multimedia efficiency and offers health recommendation by analyzing unstructured EHRs. This research contributes to advancing predictive healthcare analytics by exploring the integration of EHRs and machine learning for health recommendation, aiming to enhance patient outcomes and healthcare quality. The findings of this underscore the strength of machine learning in revolutionizing health recommendation and personalized healthcare, thereby paving the way for improved patient care and management.
- 3 October 2025
- Published by Academic Science Publications and Distributions in International Journal of Environmental Sciences
- p. 1594-1602
Open AccessNext-Gen EHR: Advancing Healthcare with Integrated Recommendations and Decision Support
The study outlines a holistic strategy for modernizing healthcare administration through the fusion of electronic health records (EHR) and machine learning techniques, including Support Vector Machines (SVM) and Naive Bayes. It addresses challenges in manual patient record management, particularly in countries like India, by proposing an EHR system implementation. The developed web application combines frontend and backend development technologies to streamline record management, enhance data accuracy, and improve healthcare delivery. Additionally, leveraging Naive Bayes and SVM algorithms, the system prioritizes multimedia efficiency and offers health recommendation by analyzing unstructured EHRs. This research contributes to advancing predictive healthcare analytics by exploring the integration of EHRs and machine learning for health recommendation, aiming to enhance patient outcomes and healthcare quality. The findings of this underscore the strength of machine learning in revolutionizing health recommendation and personalized healthcare, thereby paving the way for improved patient care and management.
- 26 July 2025
- Published by Springer Nature
- p. 295-307
LLM Text Detection Using BERT-Base Models
- 28 April 2025
- Published by Edtech Publishers (OPC) Private Limited in International Journal of Scientific Research in Engineering and Management
- Vol. 09 (04) , 1-9
MockMate: AI-Powered Online Mock Interview Assessment and Evaluation System
In the current competitive job market, being well-prepared for interviews is essential to landing a job. Traditional mock interviews, however, are not scalable and can call for a large human resource commitment. By providing a tailored, automated, and interactive online platform driven by artificial intelligence, MockMate tackles this problem. The system mimics actual interview situations, assesses candidate responses in real time, and provides thorough, data-driven feedback by utilizing cutting-edge Natural Language Processing (NLP) and speech-to-text technologies. In addition to customizing their interviews and choosing career roles, candidates can also obtain performance ratings based on their ability to communicate, solve problems, and be technically proficient. The system analyzes the responses using NLP models like BERT, TF-IDF, and Cosine Similarity and records them using OpenAI Whisper. A thorough report that highlights areas for development and compares user responses with ideal answers is produced. Secure, effective, and high-availability service is guaranteed by MockMate, which is built with a scalable architecture utilizing Next.js, Node.js, and PostgreSQL and deployed using Docker and Kubernetes on cloud platforms like AWS/GCP. This platform enables users to increase their interview readiness, boost their confidence, and increase their chances of success by bridging the gap between traditional interview preparation and contemporary AI capabilities. Keywords- BERT, TF-IDF, Cosine Similarity, OpenAI Whisper, Evaluation .
- 28 April 2025
- Published by Edtech Publishers (OPC) Private Limited in International Journal of Scientific Research in Engineering and Management
- Vol. 09 (04) , 1-9
MockMate: AI-Powered Online Mock Interview Assessment and Evaluation System
In the current competitive job market, being well-prepared for interviews is essential to landing a job. Traditional mock interviews, however, are not scalable and can call for a large human resource commitment. By providing a tailored, automated, and interactive online platform driven by artificial intelligence, MockMate tackles this problem. The system mimics actual interview situations, assesses candidate responses in real time, and provides thorough, data-driven feedback by utilizing cutting-edge Natural Language Processing (NLP) and speech-to-text technologies. In addition to customizing their interviews and choosing career roles, candidates can also obtain performance ratings based on their ability to communicate, solve problems, and be technically proficient. The system analyzes the responses using NLP models like BERT, TF-IDF, and Cosine Similarity and records them using OpenAI Whisper. A thorough report that highlights areas for development and compares user responses with ideal answers is produced. Secure, effective, and high-availability service is guaranteed by MockMate, which is built with a scalable architecture utilizing Next.js, Node.js, and PostgreSQL and deployed using Docker and Kubernetes on cloud platforms like AWS/GCP. This platform enables users to increase their interview readiness, boost their confidence, and increase their chances of success by bridging the gap between traditional interview preparation and contemporary AI capabilities. Keywords- BERT, TF-IDF, Cosine Similarity, OpenAI Whisper, Evaluation .
- Open Access
- 14 April 2025
- Published by Springer Nature in Research Square
Multimodal Decision Support System for Improved Diagnosis and Healthcare Decision Making
Access to quality health care continues to be a major challenge for remote or overlooked regions that do not have the necessary medical infrastructures and human resources. This research proposes integrating structured data-laboratory results, vitals, and demographics- with unstructured medical data-clinical notes, free-text diagnosis-finalized into Multimodal Decision Support Systems (MDSS)-that closes the healthcare gap by enhancing the diagnostic accuracy and treatment recommendations. Our innovative approach entails employing Random Forest Classifiers for structured data and BERT-based embeddings for unstructured data and fuses their predictive outputs through late fusion technique. Among various evaluated fusion methods, including simple average, weighted average, and stacked fusion, the stacked fusion approach resulted in achieving maximum diagnostic accuracy, i.e., 87 against individual models, thus taking diagnostic accuracy improvement into huge consideration as well as significant reductions in misdiagnosis, in addition to last but not least, personalized healthcare recommendation, especially to rural populations. The evaluation of this system used the MIMIC-IV dataset and showed improved performance in risk prediction and analysis of patient outcomes. The first generation of our smart health care assistant will feature video consultations and multi-lingual support, as well as real-time processing capabilities to allow access to high-quality health care for these populations. Future improvements in optimizing data imputation, enhancing interpretability, and ensuring HIPAA and GDPR compliance will make for secure and ethical data usage. This work will build ground work for AI-Personal Healthcare Solutions with a long-term goal of bridging the gap between rural and urban patient populations in access to care.
- 14 April 2025
- Published by Springer Nature in Research Square
Open AccessMultimodal Decision Support System for Improved Diagnosis and Healthcare Decision Making
Access to quality health care continues to be a major challenge for remote or overlooked regions that do not have the necessary medical infrastructures and human resources. This research proposes integrating structured data-laboratory results, vitals, and demographics- with unstructured medical data-clinical notes, free-text diagnosis-finalized into Multimodal Decision Support Systems (MDSS)-that closes the healthcare gap by enhancing the diagnostic accuracy and treatment recommendations. Our innovative approach entails employing Random Forest Classifiers for structured data and BERT-based embeddings for unstructured data and fuses their predictive outputs through late fusion technique. Among various evaluated fusion methods, including simple average, weighted average, and stacked fusion, the stacked fusion approach resulted in achieving maximum diagnostic accuracy, i.e., 87 against individual models, thus taking diagnostic accuracy improvement into huge consideration as well as significant reductions in misdiagnosis, in addition to last but not least, personalized healthcare recommendation, especially to rural populations. The evaluation of this system used the MIMIC-IV dataset and showed improved performance in risk prediction and analysis of patient outcomes. The first generation of our smart health care assistant will feature video consultations and multi-lingual support, as well as real-time processing capabilities to allow access to high-quality health care for these populations. Future improvements in optimizing data imputation, enhancing interpretability, and ensuring HIPAA and GDPR compliance will make for secure and ethical data usage. This work will build ground work for AI-Personal Healthcare Solutions with a long-term goal of bridging the gap between rural and urban patient populations in access to care.
- Open Access
- 1 January 2025
- Published by Sakun Publishing House
- Vol. 24 (3) , 688-697
Assessing the Impact of Climate-Change & Developing A Predictive Model for Heatstroke Incidents in Pune, India
The increase in occurrences of heatwaves in Pune, India, creates the urgency of developing predictive models to safeguard public health. This research endeavours to construct a robust predictive model to estimate the impact of heatwaves on Pune's population, leveraging machine learning algorithms and historical weather data spanning five years (2016-2020). By analysing meteorological attributes like temperature, humidity, dew point, and wind speed, we aim to discern patterns indicative of heatwave occurrences, enabling proactive mitigation strategies. Through an exhaustive literature review, we explored the potential of machine learning in forecasting heat-related illnesses and identified suitable algorithms for our predictive model. Random Forest Regression, CatBoost Regressor, and XGBoost Algorithm emerged as promising candidates due to their effectiveness in similar contexts. The dataset used comprises comprehensive weather attributes, supplemented by synthetic data representing heat stroke cases under Indian weather conditions. Our study's findings hold promise for policymakers and healthcare authorities, offering actionable insights to mitigate the adverse effects of heatwaves on public health in Pune and beyond.
- 1 January 2025
- Published by Sakun Publishing House
- Vol. 24 (3) , 688-697
Open AccessAssessing the Impact of Climate-Change & Developing A Predictive Model for Heatstroke Incidents in Pune, India
The increase in occurrences of heatwaves in Pune, India, creates the urgency of developing predictive models to safeguard public health. This research endeavours to construct a robust predictive model to estimate the impact of heatwaves on Pune's population, leveraging machine learning algorithms and historical weather data spanning five years (2016-2020). By analysing meteorological attributes like temperature, humidity, dew point, and wind speed, we aim to discern patterns indicative of heatwave occurrences, enabling proactive mitigation strategies. Through an exhaustive literature review, we explored the potential of machine learning in forecasting heat-related illnesses and identified suitable algorithms for our predictive model. Random Forest Regression, CatBoost Regressor, and XGBoost Algorithm emerged as promising candidates due to their effectiveness in similar contexts. The dataset used comprises comprehensive weather attributes, supplemented by synthetic data representing heat stroke cases under Indian weather conditions. Our study's findings hold promise for policymakers and healthcare authorities, offering actionable insights to mitigate the adverse effects of heatwaves on public health in Pune and beyond.