Back close

Automated Detection of Nonconvulsive Electrographic Seizures from continuous EEG 

Dept/Center/Lab: Amrita Center for Wireless Networks and Applications (AWNA)

Thematic Area: Biomedical Signal Processing and Analytics

Project Incharge:Dr. Shivapratap Gopakumar, Associate Professor, ASF
Co-Project Incharge:Dr. Siby Gopinath
Co-Project Incharge:Dr. Chandan Karmakar
Dr. Patrick Kwan, Neurologist, Monash Health, Melbourne
Dr. Terrence O’Brien, Neurology Head and Professor, Monash Health, Melbourne
Automated Detection of Nonconvulsive Electrographic Seizures from continuous EEG 

Our project aims to develop and implement an innovative automated seizure detection system tailored for ICU use. This project addresses a critical gap in ICU patient care by providing a reliable and efficient solution for non-convulsive seizure detection. The system’s potential to benefit a substantial proportion of ICU patients at risk of seizures underscores its significance in enhancing patient outcomes and optimising resource utilisation in critical care settings. 

By automating seizure detection, our project has the potential to significantly reduce the neurological consequences of non-convulsive seizures, improve patient survival rates, and enhance the quality of care delivered in ICUs worldwide. It represents a transformative step in the intersection of healthcare and technology, aligning with the broader goal of advancing critical care practices. 

Publication Details 

  • E. Ali, R. K. Udhayakumar, M. Angelova and C. Karmakar, “Performance Analysis of Entropy Methods in Detecting Epileptic Seizure from Surface Electroencephalograms,” 2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), Mexico, 2021, pp. 1082-1085. 
  • Ahsan Habib, Xiuxian Pham, Radhagayathri Udhayakumar, Emran Ali, Daniel Thom, Joshua Laing, Chandan Karmakar, Patrick Kwan, Terence O’Brien, “Deep Learning Model for Detection of Electrographic Seizures from Continuous EEG in ICU patients”, 2021 American Epilepsy Society. 

Proposed Future Work Details 

Our future work would explore the following: 

  • Continuously refine and improve the algorithm used for seizure detection by incorporating advanced machine learning techniques, such as deep learning and reinforcement learning. 
  • Integrate the seizure detection system with real-time monitoring capabilities to detect seizures and trigger interventions. 
  • Perform multi-centre trials to validate our findings. 
  • Perform cost-effectiveness analyses to evaluate the economic impact of implementing the seizure detection system in ICU settings in preparations for a product roadmap

Related Projects

Autonomous Navigational Platform for Elderly and Physically Challenged
Autonomous Navigational Platform for Elderly and Physically Challenged
Nanostructured iron-based mixed metal oxide photovoltaic thin films
Nanostructured iron-based mixed metal oxide photovoltaic thin films
Reconstructing Local Field Potential from realistic computational models for spontaneous and evoked stimuli
Reconstructing Local Field Potential from realistic computational models for spontaneous and evoked stimuli
Automatic Diagnostic Tool for Detection of Regional Wall Motion Abnormality from Echocardiogram
Automatic Diagnostic Tool for Detection of Regional Wall Motion Abnormality from Echocardiogram
An Edge-based Cyber-Physical System for Smart Polyhouse Solar Drying of Agricultural Food Products
An Edge-based Cyber-Physical System for Smart Polyhouse Solar Drying of Agricultural Food Products
Admissions Apply Now