Kaustav Choudhury | VIT University (original) (raw)

Papers by Kaustav Choudhury

Research paper thumbnail of Performance Analysis of Deep Learning Models for Detection of Autism Spectrum Disorder from EEG Signals

Traitement Du Signal, Jun 30, 2021

Autism Spectrum Disorder (ASD) starts showing symptoms in the early formative years of an individ... more Autism Spectrum Disorder (ASD) starts showing symptoms in the early formative years of an individual, affecting brain development and negatively impacting social and communication skills. Subjective diagnostic methods for ASD detection require lengthy questionnaires, trained medical personnel, and occupational therapists, and are subject to observer variability. Recent years have seen a rise in the usage of machine learning techniques for detecting ASD, which stems from a requirement for objective and accurate detection methods. This research analyzes the performance of various deep convolutional architectures for the detection of ASD. The primary objective of this work is to select a method capable of performing automatic feature extraction and classification with a relatively high degree of accuracy. Several experiments were conducted with different stateof-the-art deep architectures, out of which the ResNet50 performed the best, with an average accuracy of 81%. The performances of these architectures were analyzed in terms of precision, recall, and accuracy.

Research paper thumbnail of Performance Analysis of Deep Learning Models for Detection of Autism Spectrum Disorder from EEG Signals

Traitement du Signal

Autism Spectrum Disorder (ASD) starts showing symptoms in the early formative years of an individ... more Autism Spectrum Disorder (ASD) starts showing symptoms in the early formative years of an individual, affecting brain development and negatively impacting social and communication skills. Subjective diagnostic methods for ASD detection require lengthy questionnaires, trained medical personnel, and occupational therapists, and are subject to observer variability. Recent years have seen a rise in the usage of machine learning techniques for detecting ASD, which stems from a requirement for objective and accurate detection methods. This research analyzes the performance of various deep convolutional architectures for the detection of ASD. The primary objective of this work is to select a method capable of performing automatic feature extraction and classification with a relatively high degree of accuracy. Several experiments were conducted with different stateof-the-art deep architectures, out of which the ResNet50 performed the best, with an average accuracy of 81%. The performances of these architectures were analyzed in terms of precision, recall, and accuracy.

Research paper thumbnail of Modelling and simulation of carrier transport in quantum dot memory device for longer data retention and minimized power consumption

Journal of Computational Electronics

The performance of a group III–V material quantum dot (QD) nanostructure memory is investigated u... more The performance of a group III–V material quantum dot (QD) nanostructure memory is investigated using a self-consistent Schrodinger solver, eight-band k·p model, and carrier dynamics modelling. This model is used to explore the information loss due to the carrier emission rate in the QDs as a function of temperature, size and confinement potential. The results reveal the dominant emission mechanisms that should occur at different operating temperatures. To minimize the loss and improve the performance at room temperature, our findings reveal an increase in the carrier storage time and a reduction in the power dissipation with increasing dot size. It is further illustrated that electrons are advantageous as information carriers over holes and that the inclusion of high-bandgap barrier layers favours longer-duration data retention. The model is extended to include trap states in realistic QDs, whose effect is found to become more prominent with performance optimization. The computed results are in close agreement with other experimental data for different QDs along with barrier layer. This validates the efficacy of the model, which can be utilized as a design tool for fabricating nanoscale memories with better data retention capability.

Research paper thumbnail of Performance Analysis of Deep Learning Models for Detection of Autism Spectrum Disorder from EEG Signals

Traitement Du Signal, Jun 30, 2021

Autism Spectrum Disorder (ASD) starts showing symptoms in the early formative years of an individ... more Autism Spectrum Disorder (ASD) starts showing symptoms in the early formative years of an individual, affecting brain development and negatively impacting social and communication skills. Subjective diagnostic methods for ASD detection require lengthy questionnaires, trained medical personnel, and occupational therapists, and are subject to observer variability. Recent years have seen a rise in the usage of machine learning techniques for detecting ASD, which stems from a requirement for objective and accurate detection methods. This research analyzes the performance of various deep convolutional architectures for the detection of ASD. The primary objective of this work is to select a method capable of performing automatic feature extraction and classification with a relatively high degree of accuracy. Several experiments were conducted with different stateof-the-art deep architectures, out of which the ResNet50 performed the best, with an average accuracy of 81%. The performances of these architectures were analyzed in terms of precision, recall, and accuracy.

Research paper thumbnail of Performance Analysis of Deep Learning Models for Detection of Autism Spectrum Disorder from EEG Signals

Traitement du Signal

Autism Spectrum Disorder (ASD) starts showing symptoms in the early formative years of an individ... more Autism Spectrum Disorder (ASD) starts showing symptoms in the early formative years of an individual, affecting brain development and negatively impacting social and communication skills. Subjective diagnostic methods for ASD detection require lengthy questionnaires, trained medical personnel, and occupational therapists, and are subject to observer variability. Recent years have seen a rise in the usage of machine learning techniques for detecting ASD, which stems from a requirement for objective and accurate detection methods. This research analyzes the performance of various deep convolutional architectures for the detection of ASD. The primary objective of this work is to select a method capable of performing automatic feature extraction and classification with a relatively high degree of accuracy. Several experiments were conducted with different stateof-the-art deep architectures, out of which the ResNet50 performed the best, with an average accuracy of 81%. The performances of these architectures were analyzed in terms of precision, recall, and accuracy.

Research paper thumbnail of Modelling and simulation of carrier transport in quantum dot memory device for longer data retention and minimized power consumption

Journal of Computational Electronics

The performance of a group III–V material quantum dot (QD) nanostructure memory is investigated u... more The performance of a group III–V material quantum dot (QD) nanostructure memory is investigated using a self-consistent Schrodinger solver, eight-band k·p model, and carrier dynamics modelling. This model is used to explore the information loss due to the carrier emission rate in the QDs as a function of temperature, size and confinement potential. The results reveal the dominant emission mechanisms that should occur at different operating temperatures. To minimize the loss and improve the performance at room temperature, our findings reveal an increase in the carrier storage time and a reduction in the power dissipation with increasing dot size. It is further illustrated that electrons are advantageous as information carriers over holes and that the inclusion of high-bandgap barrier layers favours longer-duration data retention. The model is extended to include trap states in realistic QDs, whose effect is found to become more prominent with performance optimization. The computed results are in close agreement with other experimental data for different QDs along with barrier layer. This validates the efficacy of the model, which can be utilized as a design tool for fabricating nanoscale memories with better data retention capability.