Welcome to PSGR KCW's Digital Repository
Preserving Knowledge, Empowering Research
- 36 Undergraduate Programmes
- 17 Postgraduate Programmes
- 13 Research Programmes
PSGR Krishnammal College for Women (KCW), inaugurated in June 1963 under the aegis of the GRG Trust of Coimbatore has grown over the last six decades into a temple of learning and academic excellence. Founded on a motto of ‘empowering through education’, the ‘women-only’ KCW symbolizes knowledge, love, and service. KCW is an autonomous college of higher education for women. It is affiliated to the Bharathiar University, Coimbatore.
Access, Share, and Discover Research Materials and Scholarly Works at PSGR Krishnammal College for Women
Recent Submissions
INTERPRETABLE FEDERATED LEARNING FOR PRIVACY-CENTRIC GENOMIC DIAGNOSTICS: A MULTI-INSTITUTIONAL FRAMEWORK (Article)
(Fundacao de Pesquisas Cientificas de Ribeirao Preto, 2026) Sowmya M; Kaliappan A; Govindaraju S; Menaka S; Sangeetha T; Ranjani R; Arun Kumar R; Department of Computer Science (UG); Sangeetha T
The rapid digitization of healthcare systems has created large amounts of sensitive medical data that is generated from different diagnostic centers and hospitals. Deep learning has provided impressive performance and diagnostic accuracy. However, there are privacy, security, and compliance concerns related to centralized training methods. Therefore, this study presents the first healthcare diagnostics privacy-preserving framework, Hybrid Explainable Federated Attention Framework (HEFAF). The HEFAF framework combines explainable federated deep learning and a dual-level attention-based convolutional neural network with adaptive weighted aggregation for the improvement of performance and Explainability without raw patient data. The HEFAF framework has three st arting contributions. First is the private federated optimizer that provides data privacy (adaptive privacy-aware federated optimizer). This optimizer applies different local learning rate adjustments for each federated participant based on the degree of data heterogeneity. Second, is the explainable weighted aggregation, where aggregation is done selectively to explain the attributable data. This also reduces the need to explain the data. This also reduces explainable data aggregation and computational load. Third is the explainable module where the SHAP (Shapley Additive exPlanations) and attention methods combine to give clinical Explainability to diagnostic data. The model was assessed on a distributed dataset of medical images, with 18,500 diagnosis samples contributed by five medical centers. As found in the experiments, the proposed HEFAF model demonstrates a diagnostic accuracy of 95.1%, 4.6% better than independent local models and 2.3% better than traditional federated averaging models. Additionally, the adaptive aggregation mechanism explained 31% of the reduction in the communication cost, and the explainability validation aligns 93% of the model-identified regions with expert-specified regions. The results validate that the proposed framework demonstrates the ability of privacy preservation, model robustness, and interpretability. This research bridges the components of secure distributed learning and explainable clinical decision support, providing an easily scalable and regulation-compliant framework for intelligent healthcare systems.
Web Of Science (1993-2026)
(2026)
SUSTAINABLE DEVELOPMENT GOAL: SUSTAINABLE MANAGEMENT AND USE OF NATURAL RESOURCES IN TEXTILE AND APPAREL INDUSTRY (Book Chapter)
(Springer International Publishing, 2022) Department of Costume Design and Fashion; Radhakrishnan, Shanthi
The United Nations has envisaged a sustainable development plan for the year 2030 which initiates 17 sustainability development goals (SDGs) with objectives that promote all round development. This forum encourages contributions from all sectors—governments, industrial, civil organizations, public and private sectors—as opportunities for the fulfillment of these goals. The textile and fashion industries have been very popular in the extensive use of natural resources accompanied by waste and waste products that tend to pollute the environment causing hazards to the living organisms in the planet. Businesses and brands in the textile and apparel sector are earnestly working on aligning their production and management on the basis of sustainability, the pinnacle being the sustainability development goals. This chapter deals with the sustainable management and effective use of natural resources (SDG 12—Target 12.2)—water, energy and soil for the development of sustainable textile fibers and certification methodologies for sustainable reporting (SDG 12—Target 12.6). This can be achieved by sound management of chemicals and wastes occurring in the production cycle or life cycle of a product (SDG 12—Target 12.4). Green productivity in sustainable manufacturing calls for improved resource efficiency and waste reductions by implementing a cleaner manufacturing strategy. The specialized long value chain of the textile and fashion industry is poised to address the sustainability challenge to achieve the economic, social and environment development goals.
ADAPTIVE INTELLIGENT OPTIMIZATION MODEL FOR IMPROVED MEDICAL IMAGE ANALYSIS AND PREDICTIVE DIAGNOSIS USING HETEROGENEOUS HEALTHCARE DATA (Conference Paper)
(Institute of Electrical and Electronics Engineers Inc., 2025) Viji Gripsy J; Poongodi S.; Rajakumari S; Jayasree R; Nithya K.V; Sheeba L; Senthilkumaran B; Vaibhavi M; Department of Computer Science (UG); Viji Gripsy J; Poongodi S.; Jayasree R; Sheeba L
The explosion of electronic health records, imaging diagnostics, and genetic data has dramatically increased the quantity of data which can be analyzed to ultimately gain novel insights into disease process, response to therapy, and patient outcomes. In healthcare, due to the availability of mass and intricacy data leads to significant complexities in obtaining reliable insights and achieve high accuracy outcomes. The medical data attributes exhibit heterogeneity with high dimensionality. Traditional algorithmic approaches are failed to cope with these challenges. Conventional algorithms exhibit inadequate to maintain performance across different datasets. It is important to create innovative approaches to unlock maximize the use of the healthcare data. This Adaptive Intelligent Optimization (AIO) model proposed to handle intricate complex problems with multi-dimensional scenarios. Due to the unique challenges of medical data in to the healthcare that can hinder their direct integration may not provide optimal results. The primary goals of this research work focus on custom adaptive intelligent optimization technique to improve precise medical diagnosis and effective image analysis in healthcare. The proposed C-Meta- Opt model experimental results achieves a superior accuracy of 94.5% which is higher than existing optimization algorithms and 86.5% accuracy for diabetes, 89.2% for heart disease and 93.8% for breast cancer datasets.