Shahil Merchant
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Shahil Merchant has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Eason He
Electrical and Computer Engineering, College of Engineering, The Ohio State University, Columbus, OH 43210, USA
2 shared publications - Jason Forrester
Electrical and Computer Engineering, College of Engineering, The Ohio State University, Columbus, OH 43210, USA
2 shared publications - John LaRocco
Ronin Institute for Independent Scholarship 2.0, Sacramento, CA 95816, USA
2 shared publications - Qudsia Tahmina
The Ohio State University, USA
2 shared publications - Saeedeh Ziaeefard
The Ohio State University, USA
2 shared publications - Ye Lin
Electrical and Computer Engineering, College of Engineering, The Ohio State University, Columbus, OH 43210, USA
2 shared publications
Latest Publications
- Open Access
- 1 November 2025
- Published by MDPI AG in Technologies
- Vol. 13 (11) , 501
Decoding Temporally Encoded 3D Objects from Low-Cost Wearable Electroencephalography
Decoding visual content from neural activity remains a central challenge at the intersections of engineering, neuroscience, and computational modeling. Prior work has primarily leveraged electroencephalography (EEG) with generative models to recover static images. In this study, we advance EEG-based decoding by introducing a temporal encoding framework that approximates dynamic object transformations across time. EEG recordings from healthy participants (n = 20) were used to model neural representations of objects presented in “initial” and “later” states. Individualized classifiers trained on time-specific EEG signatures achieved high discriminability, with Random Forest models reaching a mean accuracy and standard deviation of 92 ± 2% and a mean AUC-ROC and standard deviation of 0.87 ± 0.10, driven largely by gamma- and beta-band activity at the frontal electrodes. These results confirm and extend evidence of strong interindividual variability, showing that subject-specific models outperform intersubject approaches in decoding temporally varying object representations. Beyond classification, we demonstrate that pairwise temporal encodings can be integrated into a generative pipeline to produce approximated reconstructions of short video sequences and 3D object renderings. Our findings establish that temporal EEG features, captured using low-cost open-source hardware, are sufficient to support the decoding of visual content across discrete time points, providing a versatile platform for potential applications in neural decoding, immersive media, and human–computer interaction.
- 1 November 2025
- Published by MDPI AG in Technologies
- Vol. 13 (11) , 501
Open AccessDecoding Temporally Encoded 3D Objects from Low-Cost Wearable Electroencephalography
Decoding visual content from neural activity remains a central challenge at the intersections of engineering, neuroscience, and computational modeling. Prior work has primarily leveraged electroencephalography (EEG) with generative models to recover static images. In this study, we advance EEG-based decoding by introducing a temporal encoding framework that approximates dynamic object transformations across time. EEG recordings from healthy participants (n = 20) were used to model neural representations of objects presented in “initial” and “later” states. Individualized classifiers trained on time-specific EEG signatures achieved high discriminability, with Random Forest models reaching a mean accuracy and standard deviation of 92 ± 2% and a mean AUC-ROC and standard deviation of 0.87 ± 0.10, driven largely by gamma- and beta-band activity at the frontal electrodes. These results confirm and extend evidence of strong interindividual variability, showing that subject-specific models outperform intersubject approaches in decoding temporally varying object representations. Beyond classification, we demonstrate that pairwise temporal encodings can be integrated into a generative pipeline to produce approximated reconstructions of short video sequences and 3D object renderings. Our findings establish that temporal EEG features, captured using low-cost open-source hardware, are sufficient to support the decoding of visual content across discrete time points, providing a versatile platform for potential applications in neural decoding, immersive media, and human–computer interaction.
- Open Access
- 20 August 2025
- Published by MDPI AG in Preprints.org
Reconstructing Temporally Encoded 3D Objects from Low-Cost Electroencephalography
Turning images and scenes from imagination and memory has applications from engineering to artistic expression. Electroencephalography (EEG) is a non-invasive technique for recording the brain’s electrical activity via scalp electrodes, accessible with low-cost headsets. Previous work used EEG to encode images with the assistance of a generative adversarial network (GAN), allowing EEG-based image reconstruction. Successive images, encoding objects at separate temporal points, were used to train a classification system. EEG data from healthy participants (N = 20) were used to encode images, each divided into an “initial state” and a “later state.” A modified “one versus rest” system using a random forest classifier was used for both offline and online use. Compared to the intersubject model, the individualized models worked most reliably with gamma and beta features on frontal electrodes, reaching a mean accuracy of 92 ± 4%, a mean F1 score of 0.64 ± 0.08, and a mean AUC-ROC of 0.87 ± 0.09. In line with prior literature, changes in spectral activity across the brain were also observed. The “paired” images of objects were converted into short films and 3D objects with the assistance of a ComfyUI pipeline. The system uses temporal encoding to capture dynamic object transformations, reliably reconstructing time-specific representations from EEG despite limitations, demonstrating potential for scalable, real-time visual memory reconstruction in research, industry, and art.
- 20 August 2025
- Published by MDPI AG in Preprints.org
Open AccessReconstructing Temporally Encoded 3D Objects from Low-Cost Electroencephalography
Turning images and scenes from imagination and memory has applications from engineering to artistic expression. Electroencephalography (EEG) is a non-invasive technique for recording the brain’s electrical activity via scalp electrodes, accessible with low-cost headsets. Previous work used EEG to encode images with the assistance of a generative adversarial network (GAN), allowing EEG-based image reconstruction. Successive images, encoding objects at separate temporal points, were used to train a classification system. EEG data from healthy participants (N = 20) were used to encode images, each divided into an “initial state” and a “later state.” A modified “one versus rest” system using a random forest classifier was used for both offline and online use. Compared to the intersubject model, the individualized models worked most reliably with gamma and beta features on frontal electrodes, reaching a mean accuracy of 92 ± 4%, a mean F1 score of 0.64 ± 0.08, and a mean AUC-ROC of 0.87 ± 0.09. In line with prior literature, changes in spectral activity across the brain were also observed. The “paired” images of objects were converted into short films and 3D objects with the assistance of a ComfyUI pipeline. The system uses temporal encoding to capture dynamic object transformations, reliably reconstructing time-specific representations from EEG despite limitations, demonstrating potential for scalable, real-time visual memory reconstruction in research, industry, and art.