research

• Security of Sensitive Data in Face Recognition System Applications: A Novel Encryption Approach 🔗

  Published in IEEE Access, 25 September 2025

  Proposed a novel encryption-based framework to enhance the security and privacy of sensitive biometric data in face recognition systems, with emphasis on safeguarding against unauthorized access and misuse.

  Keywords: Face recognition, face identification, facial encodings, encryption, decryption

---

• Securing Communication: Enhanced Table-Based End-to-End Encryption Using Polynomial Functions

  Under review in IEEE Open Journal of the Computer Society

  Proposed a novel encryption technique, analyzed for resistance against common cryptographic attacks and evaluated for efficiency through time and computational complexity comparisons.

  Keywords: Cipher table, Cryptography, Decryption, Endto-End Encryption, IP address.

---

• Multi-Regional Medicinal Plant Leaf Identification using Explainable Deep Learning with Attention Mechanism

  Under review in Journal of Ayurveda and Integrative Medicine

  This study presents a multi-regional, large-scale medicinal plant leaf identification framework that addresses dataset imbalance and diversity through a curated collection of 145 species and 29,000 images. Attention-enhanced deep learning models achieve near-perfect accuracy, with explainable AI techniques providing interpretability and reinforcing trust in automated plant identification systems.

  Keywords: Plant Leaf Classification, Deep Learning, Attention Mechanism, Explainable AI, Multi-Regional Dataset, Sustainable Development Goals

---

• An Explainable Multi-Model AI Framework for Early Detection of Sickle Cell Anemia from Blood Smear Images

  This study aims to systematically review existing machine learning and deep learning approaches for sickle cell anemia detection, evaluate and compare data-driven models across multiple datasets, and analyze their performance, efficiency, and generalization capabilities. Additionally, the work emphasizes interpretability through explainable AI techniques, providing both quantitative and qualitative insights into model behavior, reliability, and practical deployment considerations.

  Keywords: deep learning, explainability, machine learning, sickle cell anemia detection, sustainable development goals

---

• DeFi-driven Research Grant System

  This work proposes an upgradeable, DeFi-enabled research grant management smart contract using OpenZeppelin, enabling decentralized funding, proposal submission, peer review, and governance without loss of state.

  Keywords: Upgradeable Smart Contracts, DeFi, Research Grant Management, DAO Governance, IPFS, OpenZeppelin, Peer Review, Blockchain Transparency