[LGPR] Publication & Patent Success Series- Dr Sheshang Degadwala
- Scientific Progress
- Aug 19, 2025
- 4 min read
Updated: Oct 5, 2025
Journal of Trends in Computer Science and Smart technology (TCSST),7(3) 2025 | e-ISSN: 2582-4104
SCOPUS Source Link: https://suggestor.step.scopus.com/progressTracker/index.cfm?trackingID=5CD43F4D06CC7A14
Cite Score: TBA
Journal of Trends in Computer Science and Smart technology (TCSST) is an international forum for scientists, research scholars, and engineers involved in all the different aspects of computer science and technology to publish state-of-the-art and high quality research articles. This multidisciplinary journal will present readers with a diverse range of publications, which include independent/group research works, review articles that are of interest to the international research community.
The aim of the journal is to update the researchers on the significant research breakthroughs in a wide range of topics by presenting novel experiments, approaches, and concepts that enhance understanding of diverse areas of computer science. This journal is particularly interested in (but not limited to) the following sub-fields: computer architectures, programming languages, computing algorithms, sentiment analysis, natural language processing, GUI/computer graphics, computer security, augmented/virtual reality, artificial intelligence, and also the research articles that report on different topical areas and research progress initiated by computer scientists across the globe.
Emphasizes the emerging and interdisciplinary research progress on the computer science domain.
Investigates the theory, algorithms and methods involved in computer engineering practices.
Archives novel research and innovative applications in all the aspects of computer science.
Promotes the research insight and understanding on the state-of-the-art computer technologies.
Automating Histologic Assessment of Prostate Cancer with a ResNet50-Based Hybrid Vision Model https://doi.org/10.36548/jtcsst.2025.3.001
Author Details:
Sheshang Degadwala
Lincoln University College, 47301, Petaling Jaya, Selangor Darul Ehsan, Malaysia
Divya Midhunchakkaravarthy
Lincoln University College, 47301, Petaling Jaya, Selangor Darul Ehsan, Malaysia
Shakir Khan
College of Computer and Information Sciences, Imam Mohammad Ibn Saud Islamic University (IMSIU), Riyadh Saudi Arabia.
University Centre for Research and Development, Chandigarh University, Mohali, India
Abstract
Precise Gleason grading of prostate biopsy specimens is vital for determining the appropriate clinical management of prostate cancer. However, traditionally, subjective manual evaluation by pathologists is susceptible to inter-observer variability, contributing to variable diagnoses and a likelihood of less-than-optimal treatment decisions. Therefore, we present a hybrid deep-learning architecture, wherein a modified ResNet50 convolutional backbone has been amalgamated with a Vision Transformer (ViT) module with the aim of automated and standardized Gleason classification. The ResNet50 portion consists of 50 layers with bottleneck residual blocks inserted for texture and glandular pattern localization in contrast-enhanced histopathological images. The spatially rich feature maps are then forwarded to the ViT module that extracts long-range dependencies and contextual relationships across image patches through a combination of multi-head self-attention mechanisms and transformer encoders. In this manner, a combination of local feature extraction and global attention facilitates the model's learning of subtle morphological variations that are crucial for the differentiation of six different Gleason patterns on a large scale. The model was trained and validated on a balanced multiclass dataset of prostate biopsy images, achieving a classification accuracy of 99%, which is better than several existing deep-learning baselines. This hybrid architecture aims to enhance diagnostic consistency while providing a realistic, interpretable framework for implementation in clinical workflows geared toward high-throughput prostate cancer screening, especially in resource-limited healthcare settings.
Journal of Innovative Image Processing,7(1) 2025 | e-ISSN: 2582-4252
SCOPUS Source Link: https://www.scopus.com/sourceid/21101274722
Cite Score: 1.6
Journal of Innovative Image Processing, intends to provide an effective research interaction on the outcomes of high-quality theoretical and applied research from all the areas of image processing and analysis. JIIP publishes the research contributions that propose novel image interpretation and analysis methodology or the appropriate application of image processing methods in real-world settings.
The coverage of the Journal of Innovative Image Processing (JIIP) includes: image analysis, image coding and enhancement, image interpretation and retrieval, embedded image processing and real-time image processing applications, spectral imaging systems, medical imaging devices, pattern recognition, neural networks, software, specialized computer architectures and other related areas.
Prostate Biopsy Image Gleason Grading Classification using Machine Learning https://doi.org/10.36548/jiip.2025.1.007
Author Details:
Sheshang Degadwala
Lincoln University College, 47301, Petaling Jaya, Selangor Darul Ehsan, Malaysia
Divya Midhunchakkaravarthy
Lincoln University College, 47301, Petaling Jaya, Selangor Darul Ehsan, Malaysia
Shakir Khan
College of Computer and Information Sciences, Imam Mohammad Ibn Saud Islamic University (IMSIU), Riyadh Saudi Arabia.
University Centre for Research and Development, Chandigarh University, Mohali, India
Abstract
Prostate cancer diagnosis utilizes Gleason grading to analyze biopsy images to establish cancer severity levels. The analysis of prostate biopsy images is an important step in automating the Gleason grading system, which helps in prostate cancer diagnosis and prognosis. The subjective evaluation of manual grading methods exposes vulnerabilities since they lead to inconsistent results so automated solutions have become essential for precision and reliability. Present machine learning algorithms show insufficient robustness because they incorporate inadequate feature extraction approaches together with inadequate classifier choices. An ensemble Extra Trees model with characteristics from prostate biopsy images serves as the proposal for Gleason grading classification. The HSV color space produces three statistics (Mean, Standard Deviation, and Skewness) from colors with addition of entropy alongside four texture features derived from GLCM analysis which includes Contrast, Energy, Homogeneity, and Correlation. The proposed model receives evaluation against several classifiers which include Nearest Neighbors, Linear SVM, Decision Tree, and Random Forest. The ensemble Extra Trees classifier reaches 99% accuracy during testing which proves better than baseline models thus indicating its potential in trustworthy prostate cancer grading. The significance of this research is to improve the accuracy and efficiency of Gleason grading in prostate biopsy images using machine learning, aiding in early diagnosis and better treatment planning for prostate cancer.
Indexing
Patents
PROSTATE CANCER DETECTION AI DEVICE
Handheld Computer Device for Rapid Prostate Cancer Screening using Biosensor Strip, UK Design Patent Number: 6472064, Grant date: 23 September 2025
Excellent work