Dr. P. Madhavasarma is a faculty at the Department of Electronics & Instrumentaion Engineering, SASTRA Deemed University, Thanjavur. Prior to joining the institute, he was with Saraswathy College of Engineering &Technology, Tindivanam. His research interests span the fields of model prediction and fracture healing analysis, process modeling and simulation, control relevant process identification, biomedical engineering and soft computing techniques. His main focus is to impart value based quality education in the field of engineering and do research which is useful to the society. His current research interst in the field of interdicilpinary work along with medical practitioners for fracture healing analysis using soft computing and image processing based 3D pinting for scaffold for fracture treatment. He has worked with electrical related companies as a consultant to improve their product quality. He has been reviewer for the international journals such as 1)Transaction of the I
EDUCATION
• Ph.D. Control &Instrumentation SASTRA University, Thanjavur (2009)
• M.Tech. Control & Instrumentation, National Institute of Technology (NIT), Trichy (2005)
• M.B.A Marketing Management, Madurai Kamaraj University Madurai (2001)
• B.E Electrical and Electronics Engineering, Bharathidasan University (1996)
RESEARCH, TEACHING, or OTHER INTERESTS
Biomedical Engineering, Agricultural and Biological Sciences, Multidisciplinary, Biophysics
Deepfakes Unleashed: Exploring the Role of Blockchain in Managing AI- Generated Content M. Sridevi, P. Madhavasarma, Jayaprakash J. Stanly, Kumar M. Santhosh Securing AI Generated Media with Blockchain Technologies, 2025 Deepfakes, which are driven by generative AI, have revolutionized synthetic media, presenting both exciting opportunities and formidable obstacles. Deepfakes may be utilized for immersive experiences and imaginative storytelling, but they also present risks including false information, identity theft, and a decline in public confidence. The crucial role that blockchain technology plays in controlling AI-generated material and reducing the dangers of deepfakes is examined in this study. Digital media authenticity and provenance may be effectively verified thanks to blockchain's decentralized, unchangeable ledger. Blockchain and generative AI may be used to provide a strong foundation for content authentication, manipulation detection, and media production and distribution responsibility.
TFL-Net: A Hybrid Deep Learning Framework for Tibia Fracture Detection and Localization Deepak Puthanpura, Adithya Senthilkumar, Renuga Devi T, Muthukumar K, Madhavasarma P, Sridevi M Proceedings of 2025 1st International Conference on Radio Frequency Communication and Networks Rfcon 2025, 2025 Tibia fractures present significant diagnostic challenges in orthopedics due to their subtle presentation and variable healing patterns, often resulting in delayed treatment and adverse outcomes for athletes and patients. This work presents a comprehensive, hybrid deep learning framework called TFL-net (Tibia Fracture detection and Localization network) for tibia fracture diagnosis, with a particular emphasis on rapid and precise localization using the YOLOv8 object detection model. TFL-net integrates three key methodologies: (1) VGG16-based deep feature extraction combined with a Support Vector Machine (SVM) classifier for accurate fracture classification; (2) a custom-trained YOLOv8 network for real-time localization of fracture regions, leveraging advanced spatial attention, multi-scale feature integration, and grid-based prediction to achieve a mean average precision (mAP@0.5) of 0.964 and an inference time of 21.6 ms per image; and (3) an Extreme Gradient Boosting (XGBoost) module for individualized healing time prediction based on patient-specific clinical and demographic parameters. The YOLOv8 model, central to the localization task, demonstrates robust detection performance and efficiency, making it suitable for clinical deployment. The integrated system, accessible via a graphical user interface, streamlines the workflow from initial diagnosis to prognosis, supporting clinicians with accurate, real-time insights and actionable treatment planning for tibia fractures.
AI and IoMT Convergence for Emergency Healthcare: Aadhaar Authentication, Secure Data Management, and Automated Fracture Detection Krishnakumar V, Madhavasarma P, Venkata Subramanian D, Sridevi M, Prabakar T N, Venkatesh T Conference Proceedings 2025 IEEE Silchar Subsection Conference IEEE Silcon 2025, 2025 The Internet of Medical Things (IoMT) and artificial intelligence enhance real-time data management by optimizing the integration of emergency units. In this study, we propose an AI-based IoMT framework to facilitate quick patient identification and delivery of critical support. In emergencies, when facial recognition is not possible, the biometric fingerprint system (Unique Identification Authority of India [UIDAI] Aadhaar repo) will assist in accurately identifying the patient and retrieving their full medical history. IoMT-enabled ambulance devices read physiological signals (ECG, SpO2, BP, Temperature), which are analyzed by fog computing nodes for anomalies and securely transmitted to a cloud server. The AI algorithm will perform predictive risk assessments, support triage prioritization, and optimize treatment. The blockchain-based asset management layer will ensure record accuracy and prevent tampering. Federated identifiability will improve care quality and services at the point of care. This approach will reduce or eliminate the need to defer or palliate care when patients cannot consent, as interoperability across biometrics, IoMT, cloud, and hospital infrastructure will enhance response times.
Bio-inspired feature selection for early diagnosis of Parkinson’s disease through optimization of deep 3D nested learning S. Priyadharshini, K. Ramkumar, Subramaniyaswamy Vairavasundaram, K. Narasimhan, S. Venkatesh, P. Madhavasarma, Ketan Kotecha Scientific Reports, 2024 Parkinson’s disease (PD) is one of the most common neurodegenerative disorders that affect the quality of human life of millions of people throughout the world. The probability of getting affected by this disease increases with age, and it is common among the elderly population. Early detection can help in initiating medications at an earlier stage. It can significantly slow down the progression of this disease, assisting the patient to maintain a good quality of life for a more extended period. Magnetic resonance imaging (MRI)-based brain imaging is an area of active research that is used to diagnose PD disease early and to understand the key biomarkers. The prior research investigations using MRI data mainly focus on volume, structural, and morphological changes in the basal ganglia (BG) region for diagnosing PD. Recently, researchers have emphasized the significance of studying other areas of the human brain for a more comprehensive understanding of PD and also to analyze changes happening in brain tissue. Thus, to perform accurate diagnosis and treatment planning for early identification of PD, this work focuses on learning the onset of PD from images taken from whole-brain MRI using a novel 3D-convolutional neural network (3D-CNN) deep learning architecture. The conventional 3D-Resent deep learning model, after various hyper-parameter tuning and architectural changes, has achieved an accuracy of 90%. In this work, a novel 3D-CNN architecture was developed, and after several ablation studies, the model yielded results with an improved accuracy of 93.4%. Combining features from the 3D-CNN and 3D ResNet models using Canonical Correlation Analysis (CCA) resulted in 95% accuracy. For further enhancements of the model performance, feature fusion with optimization was employed, utilizing various optimization techniques. Whale optimization based on a biologically inspired approach was selected on the basis of a convergence diagram. The performance of this approach is compared to other methods and has given an accuracy of 97%. This work represents a critical advancement in improving PD diagnosis techniques and emphasizing the importance of deep nested 3D learning and bio-inspired feature selection.
An electrical stimulation data based model to predict the healing period of fractured limb P. Madhavasarma, M. Sridevi, S. Kumaravel, P. Veeraragavan Mathematical and Computer Modelling of Dynamical Systems, 2019 In this work, diagnosing of reunion of human tibia fracture across limbs using a simple mathematical model is demonstrated. At present in practice, the fracture reunion is predicted using repeated radiographs. Frequent exposure to such radiation causes harmful health effects in patients. Hence, as an alternative, modelling technique using electrical data recorded across patients stimulated with DC electric voltage of range 0.1–1V is proposed. Various model structures, namely P1D and P1DZ models were tried. An error analysis was performed and it was observed that the measured data fitted P1DZ model with an error less than 5%. Model parameters namely process gain and time constant were observed. When the model parameter process gain becomes constant, the time constant reduces significantly indicating the healing of fracture. Reunion was also confirmed with simultaneously taken radiographs. The fact that human bone is a biological semi-conductor therefore exhibits electrical properties and bone does behave like a capacitor is proved by empirical methods in our study is the novelty of the work.
Design and implementation of the monitoring and control system for unified power quality conditioner using soft computing method S. Arulkumar, P. Madhavasarma, P. Veeraragavan Proceedings 2017 2nd International Conference on Recent Trends and Challenges in Computational Models Icrtccm 2017, 2017 This paper presents a design and simulation of UPQC system for limitations of non linear load factors. The fuzzy logic controller is designed for tuning the PI controller parameters for achieving the optimal voltage is the solar power connected in the UPQC system. Solar power used to improve the power for load side power system designed in network. The MATLAB simulation software is used for simulation of results. From the simulation results, the Total Harmonics Distortion value is reduced from 2.77% to 1.29% in the voltage side similarly 4.88% to 2.01% in the current side. Also the waveform of the supply system is improved.
Model based evaluation of controller using pole placement technique for nonlinear spherical tank process Research Journal of Pharmaceutical Biological and Chemical Sciences, 2016
Leak diagnosis in pilot plant using soft computing technique Research Journal of Pharmaceutical Biological and Chemical Sciences, 2016
Transformerless hybrid power filter based on a sixswitch two-leg inverter for reduction of Total harmonic Distortion and improve the voltage performance with different aspects Arpn Journal of Engineering and Applied Sciences, 2016
Reduction of Total Harmoic Distortion with facts devices using PI and fuzzy controller International Journal of Control Theory and Applications, 2016
Modeling and performance analysis of a process based on conductivity measurement using neural networks Research Journal of Pharmaceutical Biological and Chemical Sciences, 2016
Classification of tibia fracture across limb in patients treated using DC electric stimulation based on observed healing pattern Research Journal of Pharmaceutical Biological and Chemical Sciences, 2016
Determination of optimum voltage for tibia fracture across limb in patients treated using DC electric stimulation Research Journal of Pharmaceutical Biological and Chemical Sciences, 2016
Identification of time constant for healing process of limb with fractured tibia bone using step response techniques Research Journal of Pharmaceutical Biological and Chemical Sciences, 2016
Tibia fracture healing diagnosis: A review Research Journal of Pharmaceutical Biological and Chemical Sciences, 2016
Evaluating the effect of capacitance model on tibia fractured limb healing diagnosis Research Journal of Pharmaceutical Biological and Chemical Sciences, 2015
Comparative Analysis of Fracture Healing Predicted Using Mathematical Model and Soft Computing Technique International Journal of Applied Engineering Research, 2015
Artificial intelligence in fracture healing diagnosis Research Journal of Pharmaceutical Biological and Chemical Sciences, 2015
Power quality improvement in three phase power system by combined operation of shunt active filter and photovoltaic system International Journal of Applied Engineering Research, 2014
SR/FST/ETI-153/2005 Strengthening facilities for laboratory Rs 25, 00000
SET/Scet/project-1/2018 Embedded system based patient monitoring system Rs 5, 00000
SSIPLADMIN/02 2019 A leak Measurement in an air brake system using soft computing methods Rs 5, 00000
RESEARCH OUTPUTS (PATENTS, SOFTWARE, PUBLICATIONS, PRODUCTS)
202041045778 System and method for fracture healing prediction 20/10/20
02041045898 A system and a method for controlling water supply conductivity in an industrial power plant 21/10/20
Industry, Institute, or Organisation Collaboration
1. MOU has signed between SASTRA University Thanjavur and Sarswathy College of engineering Tindivanam. For the period of three year from 2019 September to 2022 September under AICTE MARGDARSHAN SCHEME.
2. MOU has signed between Suja Shoei Industries Private limited Patheri Tindivanam and Sarswathy College of engineering Tindivanam for the period of one year March 2022 to March 2023.
