Gitanjali S. Mate
Research Subjects
Gitanjali S. Mate has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Abdul K. Kureshi
Department of Electronics, Maulana Mukhtar Ahmad Nadvi Technical Campus, Malegaon 423203, India
2 shared publications - Bhupesh Kumar Singh
Amity University Rajasthan
2 shared publications - Kamal Kishore
School of Mechanical Engineering, Shandong University of Technology, Zibo, China
1 shared publication Manmath Jete
Department Of Information Technology JSPM's Rajarshi Shahu College Of Engineering, Pune, Maharashtra, India
1 shared publicationN And Cowley
Department Of Information Technology JSPM's Rajarshi Shahu College Of Engineering, Pune, Maharashtra, India
1 shared publicationNikhil Zade
Department Of Information Technology JSPM's Rajarshi Shahu College Of Engineering, Pune, Maharashtra, India
1 shared publication
Latest Publications
- Open Access
- 1 January 2026
- Published by Elsevier in SSRN Electronic Journal
“VisionX”: Deep Learning and Explainable AI for Automated Retinal Disease Detection
This article introduces VisionX, a new computerized retinal disease detection system that blends state-of-the-art deep learning models with explainable artificial intelligence (XAI) for improved clinical decision support. Different from current systems that mostly utilize cloud-based processing and black-box models, VisionX uses a hybrid model that incorporates EfficientNet-B4 and attention mechanisms for precise detection of Diabetic Macular Edema (DME), Choroidal Neovascularization (CNV), and Drusen while being clinically interpretable explanations by sophisticated visualization methods. Our system has offline processing capability, real-time inference, and an exhaustive XAI dashboard for clinical workflows. Extensive testing on multi-centre datasets containing 28,734 retinal fundus images demonstrates high performance with 96.7% accuracy, 95.8% sensitivity, and 97.2% specificity. Clinical validation with 15 ophthalmologists in 200 cases provides 94.2% diagnostic agreement and high user satisfaction (4.7/5.0) for explanation quality. The system's offline architecture guarantees data privacy while keeping sub-second inference times, hence being appropriate for deployment in resource-limited settings. VisionX is a major breakthrough in AI-enabled ophthalmology by bridging the disparity between algorithmic performance and clinical usability.
- 1 January 2026
- Published by Elsevier in SSRN Electronic Journal
Open Access“VisionX”: Deep Learning and Explainable AI for Automated Retinal Disease Detection
This article introduces VisionX, a new computerized retinal disease detection system that blends state-of-the-art deep learning models with explainable artificial intelligence (XAI) for improved clinical decision support. Different from current systems that mostly utilize cloud-based processing and black-box models, VisionX uses a hybrid model that incorporates EfficientNet-B4 and attention mechanisms for precise detection of Diabetic Macular Edema (DME), Choroidal Neovascularization (CNV), and Drusen while being clinically interpretable explanations by sophisticated visualization methods. Our system has offline processing capability, real-time inference, and an exhaustive XAI dashboard for clinical workflows. Extensive testing on multi-centre datasets containing 28,734 retinal fundus images demonstrates high performance with 96.7% accuracy, 95.8% sensitivity, and 97.2% specificity. Clinical validation with 15 ophthalmologists in 200 cases provides 94.2% diagnostic agreement and high user satisfaction (4.7/5.0) for explanation quality. The system's offline architecture guarantees data privacy while keeping sub-second inference times, hence being appropriate for deployment in resource-limited settings. VisionX is a major breakthrough in AI-enabled ophthalmology by bridging the disparity between algorithmic performance and clinical usability.
- Open Access
- 23 December 2023
- Published by International Research Journal of Modernization in Engineering Technology and Science in International Research Journal of Modernization in Engineering Technology and Science
Automatic Text Summarization and Translation using NLP With its functioning in AI: A Comprehensive Review
- 23 December 2023
- Published by International Research Journal of Modernization in Engineering Technology and Science in International Research Journal of Modernization in Engineering Technology and Science
Open AccessAutomatic Text Summarization and Translation using NLP With its functioning in AI: A Comprehensive Review
- Open Access
- 11 October 2021
- Published by Wiley in Journal of Healthcare Engineering
- Vol. 2021, 1
Corrigendum to “An Efficient CNN for Hand X-Ray Classification of Rheumatoid Arthritis”
[This corrects the article DOI: 10.1155/2021/6712785.].
- 11 October 2021
- Published by Wiley in Journal of Healthcare Engineering
- Vol. 2021, 1
Open AccessCorrigendum to “An Efficient CNN for Hand X-Ray Classification of Rheumatoid Arthritis”
[This corrects the article DOI: 10.1155/2021/6712785.].
- Open Access
- 14 June 2021
- Published by Wiley in Journal of Healthcare Engineering
- Vol. 2021, 1-10
An Efficient CNN for Hand X-Ray Classification of Rheumatoid Arthritis
Hand Radiography (RA) is one of the prime tests for checking the progress of rheumatoid joint inflammation in human bone joints. Recognizing the specific phase of RA is a difficult assignment, as human abilities regularly curb the techniques for it. Convolutional neural network (CNN) is the center for hand recognition for recognizing complex examples. The human cerebrum capacities work in a high-level way, so CNN has been planned depending on organic neural-related organizations in humans for imitating its unpredictable capacities. This article accordingly presents the convolutional neural network (CNN) which has the ability to naturally gain proficiency with the qualities and anticipate the class of hand radiographs from an expansive informational collection. The reproduction of the CNN halfway layers, which depict the elements of the organization, is likewise appeared. For arrangement of the model, a dataset of 290 radiography images is utilized. The result indicates that hand X-rays are rated with an accuracy of 94.46% by the proposed methodology. Our experiments show that the network sensitivity is observed to be 0.95 and the specificity is observed to be 0.82.
- 14 June 2021
- Published by Wiley in Journal of Healthcare Engineering
- Vol. 2021, 1-10
Open AccessAn Efficient CNN for Hand X-Ray Classification of Rheumatoid Arthritis
Hand Radiography (RA) is one of the prime tests for checking the progress of rheumatoid joint inflammation in human bone joints. Recognizing the specific phase of RA is a difficult assignment, as human abilities regularly curb the techniques for it. Convolutional neural network (CNN) is the center for hand recognition for recognizing complex examples. The human cerebrum capacities work in a high-level way, so CNN has been planned depending on organic neural-related organizations in humans for imitating its unpredictable capacities. This article accordingly presents the convolutional neural network (CNN) which has the ability to naturally gain proficiency with the qualities and anticipate the class of hand radiographs from an expansive informational collection. The reproduction of the CNN halfway layers, which depict the elements of the organization, is likewise appeared. For arrangement of the model, a dataset of 290 radiography images is utilized. The result indicates that hand X-rays are rated with an accuracy of 94.46% by the proposed methodology. Our experiments show that the network sensitivity is observed to be 0.95 and the specificity is observed to be 0.82.
- 1 August 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Improving Efficiency of Similarity of Document Network Using Bisect K-Means
In today's internet world, most of the document related to various fields are stored in digital or electronic format. It allows easy access to users and also save the space. So the main challenge is to retrieve the interested document from huge set of document dataset. For this problem, document clustering becomes popular solution in the area of data mining. In this approach, all similar kind of document are stored collectively. This paper focus on the documents of clinical cases. The proposed system makes cluster of similar kind of clinical cases documents and improves the search efficiency. For document clustering two kinds of techniques are used named as K-means and topic detection. Initially, Document network is formed by linking various kinds of clinical reports. On this network, Bisect K-means clustering algorithm is applied after cluster formation, system identify the unique topic for each cluster. The performance of the system is tested on Vaccine Adverse Event Reporting System (VAERS) dataset and evaluation results prove that Bisect K-means is more accurately divide the number of clinical reports in appropriate cluster, than K-means clustering algorithm. The performance of the system is rely on three terms: precision, recall, and F-measure.
- 1 August 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Improving Efficiency of Similarity of Document Network Using Bisect K-Means
In today's internet world, most of the document related to various fields are stored in digital or electronic format. It allows easy access to users and also save the space. So the main challenge is to retrieve the interested document from huge set of document dataset. For this problem, document clustering becomes popular solution in the area of data mining. In this approach, all similar kind of document are stored collectively. This paper focus on the documents of clinical cases. The proposed system makes cluster of similar kind of clinical cases documents and improves the search efficiency. For document clustering two kinds of techniques are used named as K-means and topic detection. Initially, Document network is formed by linking various kinds of clinical reports. On this network, Bisect K-means clustering algorithm is applied after cluster formation, system identify the unique topic for each cluster. The performance of the system is tested on Vaccine Adverse Event Reporting System (VAERS) dataset and evaluation results prove that Bisect K-means is more accurately divide the number of clinical reports in appropriate cluster, than K-means clustering algorithm. The performance of the system is rely on three terms: precision, recall, and F-measure.