Subham Ghosh
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Subham Ghosh has a total of 3 co-authors. Here are the top co-authors by number of shared publications:
- Arnab Nandi
Department of Electronics and Communication Engineering, National Institute of Technology, Silchar, Assam, India
4 shared publications - Banani Basu
Department of Electronics and Communication Engineering, National Institute of Technology, Silchar, Assam, India
4 shared publications - Marami Das
Department of Neurology, Gauhati Medical College and Hospital, Guwahati, Assam, India
4 shared publications
Latest Publications
- 20 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Letters
- Vol. 10 (4) , 1-4
Parkinson's Disease Identification With Antenna and Shallow Hybrid Network
Parkinson's disease (PD), a prevalent neurodegenerative disorder, is frequently misdiagnosed with similar symptoms. This study investigates the feasibility of a wrist-worn wearable antenna for identifying actual PD from false PD (essential and physiological tremors). Antenna sensitivity was measured with the biological tissues. Reflection properties were measured while patients performed four motor activities. To overcome class imbalance, a generative adversarial network was used for data augmentation. Signals were normalized and converted into 2-D representations using the Gaussian-windowed Gabor wavelet transform and the short-time Fourier synchrosqueezed transform. A deep-learning model (DLM) shallow hybrid convolutional neural network (CNN)-transformer network is utilized for obtaining local and global features, with benchmark DLMs serving as comparisons. The original and enhanced datasets were trained and evaluated using fivefold subjectwise cross-validation. The suggested model, with only 0.7 million parameters, obtained 93.22% precision, 91.12% accuracy, and 91.12% F1-score.
- 20 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Letters
- Vol. 10 (4) , 1-4
Parkinson's Disease Identification With Antenna and Shallow Hybrid Network
Parkinson's disease (PD), a prevalent neurodegenerative disorder, is frequently misdiagnosed with similar symptoms. This study investigates the feasibility of a wrist-worn wearable antenna for identifying actual PD from false PD (essential and physiological tremors). Antenna sensitivity was measured with the biological tissues. Reflection properties were measured while patients performed four motor activities. To overcome class imbalance, a generative adversarial network was used for data augmentation. Signals were normalized and converted into 2-D representations using the Gaussian-windowed Gabor wavelet transform and the short-time Fourier synchrosqueezed transform. A deep-learning model (DLM) shallow hybrid convolutional neural network (CNN)-transformer network is utilized for obtaining local and global features, with benchmark DLMs serving as comparisons. The original and enhanced datasets were trained and evaluated using fivefold subjectwise cross-validation. The suggested model, with only 0.7 million parameters, obtained 93.22% precision, 91.12% accuracy, and 91.12% F1-score.
- 6 September 2024
- Published by Elsevier in Measurement
- Vol. 241
Antenna for automated classification of mouth/neck activities using convolutional neural networks and smoothed time–frequency representations
- 6 September 2024
- Published by Elsevier in Measurement
- Vol. 241
Antenna for automated classification of mouth/neck activities using convolutional neural networks and smoothed time–frequency representations
- 5 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 24 (18) , 29056-29064
Nearfield for Classification of Mouth–Neck Activities Based on Wavelet Filtering and Gabor Transformation Using DCNN
This paper proposes a novel method for detecting human mouth-neck activities using the nearfield effect of a single antenna used as the non-invasive sensor capable of capturing very complex and subtle variations of the surroundings. The time-varying signatures of reflection coefficient (S 11 ) due to perturbed nearfield radiation reveal minute distinct characteristics gathered from the Flexible-RT/Duroid (F-RTD) antenna, which can pinpoint the distinct qualities and coexisting changes in radiation of these actions for various mouth-neck motions. The F-RTD antenna demonstrates good biocompatibility to electromagnetic absorption. The wavelet-based denoising of perturbed S 11 signatures are obtained using decomposition, elimination of coefficients, and reconstruction process. A deep learning (DL) model is used to extract deep features from 2D time-frequency (TF) images generated by a Gaussian windowed Gabor wavelet transform (GWT) of the denoised signals. Two more pretrained benchmark models, namely Squeeze-Net and Alex-Net, are used, and it is found that the proposed DL model produces the best outcome in every case. The antenna is placed in two distinct locations, on the neck, and near the mouth. The neck mount is shown to be effective localization for detecting six different motions. The classification performance is analyzed S 11 magnitude and phase, as well as several sliding time windows. At the optimal sliding frame of 2.5 seconds, the accuracy, F1-score, sensitivity, and precision metrics for neck-mount are found to be 94.68%, 94.7%, 94.68%, and 94.72%, respectively. Similarly, the corresponding metrics for mouth-mount are 90.88%, 90.45%, 90%, and 90.88%.
- 5 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 24 (18) , 29056-29064
Nearfield for Classification of Mouth–Neck Activities Based on Wavelet Filtering and Gabor Transformation Using DCNN
This paper proposes a novel method for detecting human mouth-neck activities using the nearfield effect of a single antenna used as the non-invasive sensor capable of capturing very complex and subtle variations of the surroundings. The time-varying signatures of reflection coefficient (S 11 ) due to perturbed nearfield radiation reveal minute distinct characteristics gathered from the Flexible-RT/Duroid (F-RTD) antenna, which can pinpoint the distinct qualities and coexisting changes in radiation of these actions for various mouth-neck motions. The F-RTD antenna demonstrates good biocompatibility to electromagnetic absorption. The wavelet-based denoising of perturbed S 11 signatures are obtained using decomposition, elimination of coefficients, and reconstruction process. A deep learning (DL) model is used to extract deep features from 2D time-frequency (TF) images generated by a Gaussian windowed Gabor wavelet transform (GWT) of the denoised signals. Two more pretrained benchmark models, namely Squeeze-Net and Alex-Net, are used, and it is found that the proposed DL model produces the best outcome in every case. The antenna is placed in two distinct locations, on the neck, and near the mouth. The neck mount is shown to be effective localization for detecting six different motions. The classification performance is analyzed S 11 magnitude and phase, as well as several sliding time windows. At the optimal sliding frame of 2.5 seconds, the accuracy, F1-score, sensitivity, and precision metrics for neck-mount are found to be 94.68%, 94.7%, 94.68%, and 94.72%, respectively. Similarly, the corresponding metrics for mouth-mount are 90.88%, 90.45%, 90%, and 90.88%.