Narne Sravanthi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Narne Sravanthi has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
R. Thamizhamuthu
SRM Institute of Science and Technology, Kattankulathur, India
2 shared publications- Subarno Bhattacharyya
O.P. Jindal Global University, India
2 shared publications Afshan Fatima
Shadan Women's College of Engineering and Technology, Hyderabad, India
1 shared publication- Ajay Pagare
Jaipur National University Jaipur ; suresh gyan vihar university jaipur
1 shared publication Ben Sujitha
Noorul Islam Centre for Higher Education
1 shared publicationDeemed to be University
Mullana, India
1 shared publication
Latest Publications
- Open Access
A Differentiable Keras-Based Adaptive Neuro-Fuzzy Inference System for Diabetes Prediction Using NHANES 1999–2018 Data
Diabetes mellitus is a major public health concern, creating a need for reliable computational approaches for diabetes status classification and risk-oriented clinical assessment. Conventional machine learning models can capture predictive relationships in clinical data but may offer limited interpretability, while purely fuzzy approaches may have limited learning capability and scalability. This study proposes a differentiable Keras-based Adaptive Neuro-Fuzzy Inference System (ANFIS) for diabetes prediction using clinically relevant variables from the National Health and Nutrition Examination Survey (NHANES) 1999–2018 dataset. The proposed framework integrates trainable Gaussian membership functions, fuzzy-rule-based inference, and neural consequent learning within a unified TensorFlow–Keras computational graph. The fuzzy and neural parameters are jointly optimised via gradient-based learning with the Adam optimiser and binary cross-entropy loss. The experimental dataset contains 18,085 samples and uses glucose, blood pressure, body mass index (BMI), and age as input features. On the independent test set, the proposed model achieved 92.54% accuracy, 71.66% precision, 62.04% sensitivity, 96.67% specificity, 66.50% F1-score, 92.57% ROC-AUC, and 79.35% balanced accuracy. The results show that the proposed framework combines nonlinear predictive learning with structured fuzzy reasoning. The Keras-based implementation further provides a flexible foundation for integration with modern deep learning and healthcare analytics pipelines. Nevertheless, further validation using larger, heterogeneous, and multi-institutional clinical datasets is required before clinical deployment.
Open AccessA Differentiable Keras-Based Adaptive Neuro-Fuzzy Inference System for Diabetes Prediction Using NHANES 1999–2018 Data
Diabetes mellitus is a major public health concern, creating a need for reliable computational approaches for diabetes status classification and risk-oriented clinical assessment. Conventional machine learning models can capture predictive relationships in clinical data but may offer limited interpretability, while purely fuzzy approaches may have limited learning capability and scalability. This study proposes a differentiable Keras-based Adaptive Neuro-Fuzzy Inference System (ANFIS) for diabetes prediction using clinically relevant variables from the National Health and Nutrition Examination Survey (NHANES) 1999–2018 dataset. The proposed framework integrates trainable Gaussian membership functions, fuzzy-rule-based inference, and neural consequent learning within a unified TensorFlow–Keras computational graph. The fuzzy and neural parameters are jointly optimised via gradient-based learning with the Adam optimiser and binary cross-entropy loss. The experimental dataset contains 18,085 samples and uses glucose, blood pressure, body mass index (BMI), and age as input features. On the independent test set, the proposed model achieved 92.54% accuracy, 71.66% precision, 62.04% sensitivity, 96.67% specificity, 66.50% F1-score, 92.57% ROC-AUC, and 79.35% balanced accuracy. The results show that the proposed framework combines nonlinear predictive learning with structured fuzzy reasoning. The Keras-based implementation further provides a flexible foundation for integration with modern deep learning and healthcare analytics pipelines. Nevertheless, further validation using larger, heterogeneous, and multi-institutional clinical datasets is required before clinical deployment.
- 25 September 2026
- Published by IGI Global Scientific Publishing
- p. 351-374
Big Data Analytics Framework for Real-Time Food Authenticity Verification and Supply Chain Traceability Enhancement
The chapter suggests a big data analytics framework to verify food authenticity and enhance agri-food supply chain traceability in modern agri-food supply chains in real-time. The framework combines the data from the IoT sensors, batch identification via RFID/QR code, product inspection data, transaction logs, machine learning, deep learning and blockchain validation. Multi-source data is cleaned, synchronized and transformed to batch level risk features. A hybrid analytics model classifies food batches as authentic, suspicious, or high-risk, while blockchain preserves verified traceability events and tamper-resistant audit records. Experimental results show that the proposed framework improves authenticity classification, trace retrieval, missing-event detection, and tamper identification. The integrated approach performs better than product-only, IoT-only, and blockchain-only baselines because it combines product evidence with supply chain behavior. The chapter shifts food verification from passive traceability toward real-time authenticity intelligence.
- 25 September 2026
- Published by IGI Global Scientific Publishing
- p. 351-374
Big Data Analytics Framework for Real-Time Food Authenticity Verification and Supply Chain Traceability Enhancement
The chapter suggests a big data analytics framework to verify food authenticity and enhance agri-food supply chain traceability in modern agri-food supply chains in real-time. The framework combines the data from the IoT sensors, batch identification via RFID/QR code, product inspection data, transaction logs, machine learning, deep learning and blockchain validation. Multi-source data is cleaned, synchronized and transformed to batch level risk features. A hybrid analytics model classifies food batches as authentic, suspicious, or high-risk, while blockchain preserves verified traceability events and tamper-resistant audit records. Experimental results show that the proposed framework improves authenticity classification, trace retrieval, missing-event detection, and tamper identification. The integrated approach performs better than product-only, IoT-only, and blockchain-only baselines because it combines product evidence with supply chain behavior. The chapter shifts food verification from passive traceability toward real-time authenticity intelligence.
- 25 September 2026
- Published by IGI Global Scientific Publishing
- p. 161-188
Digital Forensics in IoT-Enabled Heterogeneous and Intelligent Network Environments
The excessive proliferation of Internet of Things (IoT) ecosystems, which are characterized by the high number of devices, transient data generation, and constantly changing cyber threats have presented serious challenges to digital forensic investigations. To solve these problems, Adaptive Forensic Intelligence Model (AFIM) is suggested as a combined model. AFIM is a multi-modal evidence-gathering mechanism, a blockchain-based secure evidence management mechanism, and an AI-based forensic analytics engine. The model has been experimentally tested on a hybrid dataset of real data of the IoT and simulated cyberattack scenarios. The federated learning framework enables models to be developed in a decentralized manner as well as using adaptive thresholding for anomaly detection “on the fly”, thereby creating scalable and resilient systems to operate in distributed environments.
- 25 September 2026
- Published by IGI Global Scientific Publishing
- p. 161-188
Digital Forensics in IoT-Enabled Heterogeneous and Intelligent Network Environments
The excessive proliferation of Internet of Things (IoT) ecosystems, which are characterized by the high number of devices, transient data generation, and constantly changing cyber threats have presented serious challenges to digital forensic investigations. To solve these problems, Adaptive Forensic Intelligence Model (AFIM) is suggested as a combined model. AFIM is a multi-modal evidence-gathering mechanism, a blockchain-based secure evidence management mechanism, and an AI-based forensic analytics engine. The model has been experimentally tested on a hybrid dataset of real data of the IoT and simulated cyberattack scenarios. The federated learning framework enables models to be developed in a decentralized manner as well as using adaptive thresholding for anomaly detection “on the fly”, thereby creating scalable and resilient systems to operate in distributed environments.
- 20 February 2026
- Published by IGI Global Scientific Publishing
- p. 303-332
Quantum-Enhanced Neuromorphic Computing for Real-World Applications in Vision Speech Healthcare and Smart Cities
Smart systems deployed in vision, speech, healthcare, and smart city settings are increasingly working under severe requirements of latency, energy use and uncertainty, revealing inherent weaknesses in standard deep learning methods, as well as in single- shots of quantum computing. Although both neuromorphic computing and quantum-enhanced learning have been demonstrated to be promising, independently, neither approach can provide a tradeoff between real-time performance and global reasoning. This paper discusses the absence of a single and deployment focused model that would bring neuromorphic efficiency and quantum expressiveness to a single system. In our proposal, a quantum-enhanced neuromorphic computing system, event-driven spiking neural networks are used to do continuous, low-power perception and quantum-assisted machinery used to invoke uncertainty-aware classification and global optimization.
- 20 February 2026
- Published by IGI Global Scientific Publishing
- p. 303-332
Quantum-Enhanced Neuromorphic Computing for Real-World Applications in Vision Speech Healthcare and Smart Cities
Smart systems deployed in vision, speech, healthcare, and smart city settings are increasingly working under severe requirements of latency, energy use and uncertainty, revealing inherent weaknesses in standard deep learning methods, as well as in single- shots of quantum computing. Although both neuromorphic computing and quantum-enhanced learning have been demonstrated to be promising, independently, neither approach can provide a tradeoff between real-time performance and global reasoning. This paper discusses the absence of a single and deployment focused model that would bring neuromorphic efficiency and quantum expressiveness to a single system. In our proposal, a quantum-enhanced neuromorphic computing system, event-driven spiking neural networks are used to do continuous, low-power perception and quantum-assisted machinery used to invoke uncertainty-aware classification and global optimization.