BS, Aiswarya, Anusha SP, and Jerrin Thomas Panachakel. "Operational stage evaluation of urban roads in achieving carbon neutrality." Sustainable Transport and Livability 1.1 (2024): 1-21. Paper
Zakeer, Sumina, et al. "Application of GIS in road safety analysis: A case study of NH66 corridor in Kollam." Journal of Transportation Safety & Security (2024): 1-17. Paper
J.T. Panachakel, A.G. Ramakrishnan, "Decoding Imagined Speech from EEG Using Transfer Learning." IEEE Access 9 (2021): 135371-135383. Paper
J.T. Panachakel, A.G. Ramakrishnan, "Decoding Covert Speech from EEG - A Comprehensive Review", Frontiers in Neuroscience 15 (2021): 392. Paper | News Article
John, Angel Mary, Jerrin Thomas Panachakel, and S. P. Anusha. "Navigating AI Policy Landscapes: Insights into Human Rights Considerations Across IEEE Regions." 2024 IEEE 12th Region 10 Humanitarian Technology Conference (R10-HTC). IEEE, 2024. Paper
J.T. Panachake, et al. "Emotion Detection from EEG using Transfer Learning" 2023 45th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC). IEEE, 2023. Preprint
J.T. Panachakel, A.G. Ramakrishnan, A.S. Anusha, K. Sharma, "Can We Identify the Category of Imagined Phoneme From EEG?", Proceedings of IEEE EMBC 2021.
J.T. Panachakel, A.G. Ramakrishnan, "Classification of Phonological Categories in Imagined Speech using Phase Synchronization Measure", Proceedings of IEEE EMBC 2021.
J.T. Panachakel, A.G. Ramakrishnan, and K.P. Manjunath. "VR Glasses based Measurement of Responses to Dichoptic Stimuli: A Potential Tool for Quantifying Amblyopia", Proceedings of IEEE EMBC 2020. Paper | Presentation | Slides
J.T. Panachakel, A.G. Ramakrishnan, and T.V. Ananthapadmanabha. "Decoding Imagined Speech Using Wavelet Features and Deep Neural Networks", Proceedings of IEEE Indicon 2019, December 13-15, 2019. Paper | Slides
J.T. Panachakel, N.N. Vinayak, M. Nunna, A.G. Ramakrishnan, and K. Sharma. "An Improved EEG Acquisition Protocol Facilitates Localized Neural Activation", Proceedings of International Conference on Communication Systems and Networks (ComNet 2019), December 12 & 13, 2019. Paper | Slides
J.T. Panachakel, and KC Finitha. "Energy Efficient Compression of Shock Data using Compressed Sensing", Springer Intelligent Systems Technologies and Applications, 2016. Paper
J.T. Panachakel. "Contourlet Transform and Iterative Noise Free Filtering Based Bilayer Filter for Enhancing Echocardiogram", Proceedings of 2012 International Conference on Green Technologies (ICGT 2012), December 18 - 20, 2012. Paper
Ph.D. Thesis:
Title: Machine Learning for Decoding Imagined Words and Altered State of Consciousness from EEG
Guide: Prof. A. G. Ramakrishnan, Dept. of Electrical Engineering, Indian Institute of Science, Bangalore
Abstract: The thesis explores several architectures for accurately decoding the cognitive activity from EEG recorded during speech imagery and Rajayoga meditation. The thesis can be divided into three parts.
The first part of the thesis deals with the investigation on whether neural correlates of phonological categories exist in the EEG recorded during speech imagery. We have shown that neural correlates of phonological categories do exist in the EEG recorded during speech imagery. These correlates lead to statistically significant differences in the mean phase coherence (MPC) values of the EEG across several cortical regions. We have also shown that MPC values can be used for accurately classifying the EEG recorded during speech imagery based on the phonological category of the prompts. The proposed architecture for this task has an accuracy of 84.9%.
The second part deals with decoding imagined words from EEG. One of the challenges in designing systems for decoding imagined words from EEG is the limited availability of data. We have presented three architectures for decoding imagined words from EEG. All three architectures alleviate this problem of limited availability of data. The transfer learning-based architecture employs MPC and magnitude-squared coherence values along with a ResNet50-based network. The ResNet50-based network is employed as a fixed feature extractor. This architecture achieves an accuracy of 92.8% on a publicly available EEG dataset in classifying speech imagery.
The last part of the thesis deals with the classification of Altered State of Consciousness from Resting State. We have presented three architectures for classifying the altered state of consciousness during Rajayoga meditation from the resting state. Both CSP-LDA-LSTM (common spatial pattern-linear discriminant analysis-long short-term memory) and SVD-DNN (singular value decomposition-deep neural network) architectures are able to capture subject-invariant features. The best intra-subject accuracy obtained is 98.2% and the best inter-subject accuracy is 96.4%. Although it may seem that this problem is unrelated to the problems addressed in the first two parts, all the three problems are related since they are based on the same type of data and the results serve related goals.
M.Tech. Thesis:
Title: Prediction of Cerebrovascular Accident Using Neuroimaging
Guide: Dr. Jeena R.S., Department of Electronics and Communication Engineering, Government Engineering College, Idukki, Kerala
Abstract: Cerebrovascular accident (CVA) or stroke is the rapid loss of brain function due to disturbance in the blood supply to the brain. Statistically, stroke is the second leading cause of death. This has motivated us to suggest a two-tier system for predicting stroke; the first tier makes use of Artificial Neural Network (ANN) to predict the chances of a person suffering from stroke. The ANN is trained the using the values of various risk factors of stroke of several patients who had stroke. Once a person is classified as having a high risk of stroke, s/he undergoes another the tier-2 classification test where his/her neuro MRI (Magnetic Resonance Imaging) is analysed to predict the chances of stroke. The tier-2 uses non-negative matrix factorization (NMF) for feature extraction and support vector machine (SVM) for classification. The work concentrates only on ischemic stroke which accounts for 87% of CVAs.
M.Tech. Project
Title: "Diverse Domain Multilevel Watermarking
Guide: Prof. Anurenjan P.R., Dept. of Electronics and Communication Engineering, College of Engineering, Trivandrum, Kerala.
Abstract: The ever-increasing illegal manipulation of genuine audio products has been a dilemma for the music industry. Audio watermarking has been proposed as a possible solution since this technology embeds copyright information into audio files as a proof o their ownership. In our project we make use of Audio Watermarking. Here the information being watermarked is a text message. The algorithm being used is a cascade of three powerful mathematical transforms; the discrete wavelets transform (DWT), the discrete cosine transform (DCT) and the singular value decomposition (SVD). In this project, we propose an effective, robust, and an inaudible audio watermarking algorithm which is further strengthened using two novel detection techniques, AOT (Adaptively Optimized Threshold) and AOTx (AOT extended).By implementing multiple watermarking we were able to innovate the already existing technology. The results were analysed both in terms of perceptual quality (MOS) and robustness to various attacks including delay, inversion, amplification, interference from power supply etc.
B.Tech. Project
Title: Video Processing Engine for Vintage Videos
Guide: Dr. Luxy Mathews, Dept. of Electronics and Communication Engineering, Mar Baselios College of Engineering and Technology, Trivandrum, Kerala.
Abstract: Old black and white videos images are highly affected by noise, especially impulse noise like salt and pepper noise. Most of the video processing engines use all pixels within a window to filter out the impulse noise. They increase the size of neighboring pixels with the increase of noise density. However, this estimate of all neighboring pixels does not give promising results for high level of noise density. In contrast, in the project, we propose an impulse noise removal scheme that emphasizes on few noise-free pixels. The video images are first sampled into frames and the proposed iterative algorithm searches the noise-free pixels within a small neighborhood. The noisy-pixel is then replaced with the average estimated from noise-free pixels. The iterative process continues until all noisy-pixels of the corrupted image are filtered. This scheme is expected to provide superior performance as compared to the existing approaches, especially for high density salt-and-pepper noise.