Manohar K. Kodmelwar
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Manohar K. Kodmelwar has a total of 31 co-authors. Here are the top co-authors by number of shared publications:
- Chandrakant D. Kokane
Department of Computer Science and Engineering (Artificial Intelligence), Vishwakarma Institute of Technology, Pune, Maharashtra, India
2 shared publications - Kishor Renukadas Pathak
Vishwakarma Institute of Information Technology
2 shared publications - Neeta Deshpande
Department of Computer Engineering, GES's R H Sapat College of Engineering Management Studies and Research, Nashik, India
2 shared publications - Suhas Chavan
Department of Computer Enginerering, Vishwakarma University, Kondhwa, Pune, Maharashtra, India
2 shared publications - Adwait Gondhalekar
Vishwakarma Institute of Information Technology
1 shared publication Anand Daulatabad
Department of Humanities Nutan Maharashtra Institute of Engineering & Technology Talegaon(D Pune Maharashtra India
1 shared publication
Latest Publications
- 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
- 31 October 2025
- Published by Elsevier in Indian Journal of Tuberculosis
- Vol. 72, S31-S38
CNN-based image recognition of Acid-Fast Bacilli in sputum smears for enhanced tuberculosis diagnosis
- 31 October 2025
- Published by Elsevier in Indian Journal of Tuberculosis
- Vol. 72, S31-S38
CNN-based image recognition of Acid-Fast Bacilli in sputum smears for enhanced tuberculosis diagnosis
- 31 October 2025
- Published by Springer Nature
- p. 124-134
Enhancing IOT-Based Fruit Picking with Reinforcement Learning, Transfer Learning and Neuroevolution
- 31 October 2025
- Published by Springer Nature
- p. 124-134
Enhancing IOT-Based Fruit Picking with Reinforcement Learning, Transfer Learning and Neuroevolution
- 24 August 2025
- Published by Springer Nature
- p. 541-549
Bridging the Past: Neural Machine Translation of Maharashtri Prakrit to English
- 24 August 2025
- Published by Springer Nature
- p. 541-549
Bridging the Past: Neural Machine Translation of Maharashtri Prakrit to English
- 1 January 2025
- Published by Researcher Connect Innovation and Impact Private Limited and MRI in Journal of Entomological Research
- Vol. 49 (suppl) , 1141-1147
Pest outbreaks in the context of climate change: Predicting future scenarios
Present studies aimed to identify the primary factors influencing the response of pests to climate change particularly changes in temperature, rainfall patterns, and surrounding events. Case studies reveal how complex these connections are by illustrating instances of temperature-related insect outbreaks. Further study also discusses advantages and drawbacks as well as related prediction models using temperature data to project insect behaviour relationship with climate change. The findings make it abundantly evident that we must have flexible management strategies including forecasts of climate change while controlling pests. By applying fresh technology and approaches from several domains, stakeholders may raise their preparedness for next pest outbreaks.
- 1 January 2025
- Published by Researcher Connect Innovation and Impact Private Limited and MRI in Journal of Entomological Research
- Vol. 49 (suppl) , 1141-1147
Pest outbreaks in the context of climate change: Predicting future scenarios
Present studies aimed to identify the primary factors influencing the response of pests to climate change particularly changes in temperature, rainfall patterns, and surrounding events. Case studies reveal how complex these connections are by illustrating instances of temperature-related insect outbreaks. Further study also discusses advantages and drawbacks as well as related prediction models using temperature data to project insect behaviour relationship with climate change. The findings make it abundantly evident that we must have flexible management strategies including forecasts of climate change while controlling pests. By applying fresh technology and approaches from several domains, stakeholders may raise their preparedness for next pest outbreaks.
- Open Access
- 20 November 2024
- Published by Science Research Society in Panamerican Mathematical Journal
- Vol. 35 (2s) , 482-501
A Smart Waste Management: an Adaptive Algorithm for Classification of Degradable and Non-Degradable Waste using Machine Learning Model
Urban waste management poses significant challenges due to the growing volume and complexity of waste generated in cities. Effective classification of waste into degradable and non-degradable categories is critical for optimizing waste disposal, promoting recycling, and mitigating environmental impacts. This report presents a comprehensive exploration of an adaptive machine learning algorithm for image-based waste classification, designed to enhance municipal waste management systems. The proposed approach integrates advanced image processing techniques with machine learning models, such as Convolutional Neural Networks (CNNs) and their efficient variants, to classify waste with high accuracy. A citizen reporting interface allows users to capture and upload waste images, which are processed in real-time by a centralized system. The adaptive algorithm incorporates preprocessing steps, feature extraction, and model inference to deliver actionable insights, enabling municipal authorities to deploy appropriate waste collection mechanisms promptly. Through a detailed literature review, the report examines existing methodologies, benchmarks their performance, and identifies key challenges, including data quality, scalability, and integration with urban infrastructure. Case studies from global implementations illustrate the transformative potential of AI-driven waste management systems. Furthermore, the report outlines future directions, such as leveraging advanced deep learning architectures, real-time deployment using edge computing, and expanding classification to include hazardous and recyclable waste categories. This work contributes to the growing field of smart waste management by demonstrating how machine learning can address critical urban challenges. By fostering collaboration between technology and sustainability, the proposed framework lays the foundation for cleaner and more sustainable cities.
- 20 November 2024
- Published by Science Research Society in Panamerican Mathematical Journal
- Vol. 35 (2s) , 482-501
Open AccessA Smart Waste Management: an Adaptive Algorithm for Classification of Degradable and Non-Degradable Waste using Machine Learning Model
Urban waste management poses significant challenges due to the growing volume and complexity of waste generated in cities. Effective classification of waste into degradable and non-degradable categories is critical for optimizing waste disposal, promoting recycling, and mitigating environmental impacts. This report presents a comprehensive exploration of an adaptive machine learning algorithm for image-based waste classification, designed to enhance municipal waste management systems. The proposed approach integrates advanced image processing techniques with machine learning models, such as Convolutional Neural Networks (CNNs) and their efficient variants, to classify waste with high accuracy. A citizen reporting interface allows users to capture and upload waste images, which are processed in real-time by a centralized system. The adaptive algorithm incorporates preprocessing steps, feature extraction, and model inference to deliver actionable insights, enabling municipal authorities to deploy appropriate waste collection mechanisms promptly. Through a detailed literature review, the report examines existing methodologies, benchmarks their performance, and identifies key challenges, including data quality, scalability, and integration with urban infrastructure. Case studies from global implementations illustrate the transformative potential of AI-driven waste management systems. Furthermore, the report outlines future directions, such as leveraging advanced deep learning architectures, real-time deployment using edge computing, and expanding classification to include hazardous and recyclable waste categories. This work contributes to the growing field of smart waste management by demonstrating how machine learning can address critical urban challenges. By fostering collaboration between technology and sustainability, the proposed framework lays the foundation for cleaner and more sustainable cities.
- Open Access
- 7 October 2024
- Published by Science Research Society in Panamerican Mathematical Journal
- Vol. 34 (4) , 481-498
Mathematical Evaluation of Deep Learning Architecture with Feature Fusion for Cervical Cancer Detection Classification
Cervical cancer is still one of the main reasons women die around the world, which shows how important it is to have accurate and quick ways to diagnose the disease. This study looks into how to improve the accuracy of finding cervical cancer using a deep learning-based classification system that combines features. In this mixed classic feature extraction methods, like Histogram of Oriented Gradients (HOG) and Scale-Invariant Feature Transform (SIFT), with deep learning architectures, like Graph Convolutional Networks (GCNs) and Capsule Networks. Each of these added its own benefits to the fusion model. Both HOG and SIFT are good at capturing important texture and structure features, even when the image conditions change. GCNs are used to learn relationship data between picture areas. This helps get a better sense of how things fit together in space, which is very important for medical imaging. Capsule Networks, on the other hand, let you understand the subtleties of hierarchical patterns in pictures of cervical cells. They do this by looking at spatial hierarchies and rotational changes that are very important for finding cancer. Our suggested fusion model combines these multi-level features, taking advantage of how the best features of deep learning and standard learning methods work together. Validation on standard cervical cancer image datasets showed that this fusion approach was much more accurate and precise at classifying than individual methods. Quantitative testing showed that the fusion model had higher sensitivity and specificity, which shows that it can reduce the number of fake negatives and positives, which is very important in clinical diagnosis. Our results show that the feature fusion model not only gives us a strong and accurate way to classify cervical cancer, but it also lays the groundwork for more uses in other complicated areas of medical imaging.
- 7 October 2024
- Published by Science Research Society in Panamerican Mathematical Journal
- Vol. 34 (4) , 481-498
Open AccessMathematical Evaluation of Deep Learning Architecture with Feature Fusion for Cervical Cancer Detection Classification
Cervical cancer is still one of the main reasons women die around the world, which shows how important it is to have accurate and quick ways to diagnose the disease. This study looks into how to improve the accuracy of finding cervical cancer using a deep learning-based classification system that combines features. In this mixed classic feature extraction methods, like Histogram of Oriented Gradients (HOG) and Scale-Invariant Feature Transform (SIFT), with deep learning architectures, like Graph Convolutional Networks (GCNs) and Capsule Networks. Each of these added its own benefits to the fusion model. Both HOG and SIFT are good at capturing important texture and structure features, even when the image conditions change. GCNs are used to learn relationship data between picture areas. This helps get a better sense of how things fit together in space, which is very important for medical imaging. Capsule Networks, on the other hand, let you understand the subtleties of hierarchical patterns in pictures of cervical cells. They do this by looking at spatial hierarchies and rotational changes that are very important for finding cancer. Our suggested fusion model combines these multi-level features, taking advantage of how the best features of deep learning and standard learning methods work together. Validation on standard cervical cancer image datasets showed that this fusion approach was much more accurate and precise at classifying than individual methods. Quantitative testing showed that the fusion model had higher sensitivity and specificity, which shows that it can reduce the number of fake negatives and positives, which is very important in clinical diagnosis. Our results show that the feature fusion model not only gives us a strong and accurate way to classify cervical cancer, but it also lays the groundwork for more uses in other complicated areas of medical imaging.
- Open Access
- 1 January 2024
- Published by Skyler Publication
- Vol. 13 (2) , 188-203
Machine Learning Models for Early Detection of Hepatic Disorders Using Clinical Data
Hepatic disarranges, which incorporate a wide extend of liver sicknesses, are a major world wellbeing issue since they are exceptionally common and can get more regrettable over time, driving to genuine conditions like cirrhosis and liver cancer. Early distinguishing proof is imperative for compelling administration and treatment, but it's still difficult to do since these disarranges are so complicated and the early signs are so gentle. This article talks around how to form and utilize progressed machine learning models to discover liver maladies early on utilizing clinical information. By utilizing machine learning strategies, we trust to move forward the precision of analyze and make it simpler for individuals to urge offer assistance when they require it, which is able lead to way better comes about for patients. A large set of clinical information, counting liver function tests, chemistry markers, and statistic data, is utilized within the think about. To figure in case somebody encompasses a liver issue, we utilize numerous sorts of machine learning models, such as back vector machines (SVM), choice trees, irregular woodlands, and profound learning neural systems. Measurements like precision, accuracy, review, and the range beneath the recipient working characteristic bend (AUC-ROC) are utilized to judge how well each show works. This lets us compare their capacity to form forecasts in a point by point way. Our inquire about appears that machine learning models can effectively discover patterns in clinical information that point to liver issues. A few models are exceptionally precise and dependable. In specific, the random woodland demonstrate does distant better;a much better;a higher;a stronger;an improved">a higher work of being precise and simple to get it. It gives us useful information almost how critical distinctive clinical characteristics are. Adding deep learning strategies too appears like a great thought for finding complicated nonlinear associations within the information, which would make acknowledgment indeed better. The ponder too talks approximately the problems that can happen when machine learning is utilized in therapeutic diagnostics. These issues incorporate awful information, models that are difficult to get it, and the chance of overfitting. We recommend ways to make strides the models, like choosing the proper highlights and doing cross-validation, to form beyond any doubt the comes about are dependable and can be utilized in other circumstances. Too talked approximately are the ethical issues that come up when utilizing machine learning in healthcare, like securing persistent protection and information, which stresses how vital it is to utilize AI in a clear and dependable way.
- 1 January 2024
- Published by Skyler Publication
- Vol. 13 (2) , 188-203
Open AccessMachine Learning Models for Early Detection of Hepatic Disorders Using Clinical Data
Hepatic disarranges, which incorporate a wide extend of liver sicknesses, are a major world wellbeing issue since they are exceptionally common and can get more regrettable over time, driving to genuine conditions like cirrhosis and liver cancer. Early distinguishing proof is imperative for compelling administration and treatment, but it's still difficult to do since these disarranges are so complicated and the early signs are so gentle. This article talks around how to form and utilize progressed machine learning models to discover liver maladies early on utilizing clinical information. By utilizing machine learning strategies, we trust to move forward the precision of analyze and make it simpler for individuals to urge offer assistance when they require it, which is able lead to way better comes about for patients. A large set of clinical information, counting liver function tests, chemistry markers, and statistic data, is utilized within the think about. To figure in case somebody encompasses a liver issue, we utilize numerous sorts of machine learning models, such as back vector machines (SVM), choice trees, irregular woodlands, and profound learning neural systems. Measurements like precision, accuracy, review, and the range beneath the recipient working characteristic bend (AUC-ROC) are utilized to judge how well each show works. This lets us compare their capacity to form forecasts in a point by point way. Our inquire about appears that machine learning models can effectively discover patterns in clinical information that point to liver issues. A few models are exceptionally precise and dependable. In specific, the random woodland demonstrate does distant better;a much better;a higher;a stronger;an improved">a higher work of being precise and simple to get it. It gives us useful information almost how critical distinctive clinical characteristics are. Adding deep learning strategies too appears like a great thought for finding complicated nonlinear associations within the information, which would make acknowledgment indeed better. The ponder too talks approximately the problems that can happen when machine learning is utilized in therapeutic diagnostics. These issues incorporate awful information, models that are difficult to get it, and the chance of overfitting. We recommend ways to make strides the models, like choosing the proper highlights and doing cross-validation, to form beyond any doubt the comes about are dependable and can be utilized in other circumstances. Too talked approximately are the ethical issues that come up when utilizing machine learning in healthcare, like securing persistent protection and information, which stresses how vital it is to utilize AI in a clear and dependable way.
- Open Access
- 7 October 2023
- Published by Auricle Technologies, Pvt., Ltd. in International Journal on Recent and Innovation Trends in Computing and Communication
- Vol. 11 (11s) , 656-664
Supply Chain Tracing and Anti-Counterfeiting with Distributed Ledger Technology
In recent times, there has been a rampant proliferation of counterfeit products that has left a trail of devastation in the manufacturing sectors. The repercussions of this extend to companies, impacting their brand reputation, revenue streams and overall profitability. Industries like agriculture, banking, electronics, and high-value deliveries uses the emergence of blockchain technology as a powerful tool to discern between authentic and counterfeit items. Its potential as a means to curtail the influx of fake products in the market is substantial. Blockchain technology, at its core, operates as a decentralized and distributed digital ledger system, meticulously recording transactions within interconnected blocks across multiple databases. The inherent security of this technology ensures the immutability of these blocks, rendering them invulnerable to alteration or hacking. By leveraging blockchain technology, consumers can independently verify the authenticity of a product, eliminating the need for reliance on third-party intermediaries. Incorporating recent technological advancements, the utilization of Quick Response (QR) codes offers a robust approach to combat the proliferation of counterfeit goods. The integration of blockchain technology with QR codes serves as a means to uphold the integrity of products. This innovative system securely stores product details and unique codes in the form of blocks, where QR codes play a pivotal role in collecting and matching these unique codes with entries in the blockchain database. If the QR code matches with entries in the database, the user receives a confirmation of the product's authenticity; otherwise, an alert is triggered, signaling the presence of a counterfeit product.
- 7 October 2023
- Published by Auricle Technologies, Pvt., Ltd. in International Journal on Recent and Innovation Trends in Computing and Communication
- Vol. 11 (11s) , 656-664
Open AccessSupply Chain Tracing and Anti-Counterfeiting with Distributed Ledger Technology
In recent times, there has been a rampant proliferation of counterfeit products that has left a trail of devastation in the manufacturing sectors. The repercussions of this extend to companies, impacting their brand reputation, revenue streams and overall profitability. Industries like agriculture, banking, electronics, and high-value deliveries uses the emergence of blockchain technology as a powerful tool to discern between authentic and counterfeit items. Its potential as a means to curtail the influx of fake products in the market is substantial. Blockchain technology, at its core, operates as a decentralized and distributed digital ledger system, meticulously recording transactions within interconnected blocks across multiple databases. The inherent security of this technology ensures the immutability of these blocks, rendering them invulnerable to alteration or hacking. By leveraging blockchain technology, consumers can independently verify the authenticity of a product, eliminating the need for reliance on third-party intermediaries. Incorporating recent technological advancements, the utilization of Quick Response (QR) codes offers a robust approach to combat the proliferation of counterfeit goods. The integration of blockchain technology with QR codes serves as a means to uphold the integrity of products. This innovative system securely stores product details and unique codes in the form of blocks, where QR codes play a pivotal role in collecting and matching these unique codes with entries in the blockchain database. If the QR code matches with entries in the database, the user receives a confirmation of the product's authenticity; otherwise, an alert is triggered, signaling the presence of a counterfeit product.