Publications
19 results found
Bajaj N, Requena Carrion J, 2023, Deep Representation of EEG Signals Using Spatio-Spectral Feature Images, APPLIED SCIENCES-BASEL, Vol: 13
Bajaj N, Rajwadi M, Constance TG, et al., 2023, Deception detection in conversations using the proximity of linguistic markers, KNOWLEDGE-BASED SYSTEMS, Vol: 267, ISSN: 0950-7051
Wu H, Patel KHK, Li X, et al., 2022, A fully-automated paper ECG digitisation algorithm using deep learning, Scientific Reports, Vol: 12, ISSN: 2045-2322
There is increasing focus on applying deep learning methods to electrocardiograms (ECGs), with recent studies showing that neural networks (NNs) can predict future heart failure or atrial fibrillation from the ECG alone. However, large numbers of ECGs are needed to train NNs, and many ECGs are currently only in paper format, which are not suitable for NN training. We developed a fully-automated online ECG digitisation tool to convert scanned paper ECGs into digital signals. Using automated horizontal and vertical anchor point detection, the algorithm automatically segments the ECG image into separate images for the 12 leads and a dynamical morphological algorithm is then applied to extract the signal of interest. We then validated the performance of the algorithm on 515 digital ECGs, of which 45 were printed, scanned and redigitised. The automated digitisation tool achieved 99.0% correlation between the digitised signals and the ground truth ECG (n = 515 standard 3-by-4 ECGs) after excluding ECGs with overlap of lead signals. Without exclusion, the performance of average correlation was from 90 to 97% across the leads on all 3-by-4 ECGs. There was a 97% correlation for 12-by-1 and 3-by-1 ECG formats after excluding ECGs with overlap of lead signals. Without exclusion, the average correlation of some leads in 12-by-1 ECGs was 60–70% and the average correlation of 3-by-1 ECGs achieved 80–90%. ECGs that were printed, scanned, and redigitised, our tool achieved 96% correlation with the original signals. We have developed and validated a fully-automated, user-friendly, online ECG digitisation tool. Unlike other available tools, this does not require any manual segmentation of ECG signals. Our tool can facilitate the rapid and automated digitisation of large repositories of paper ECGs to allow them to be used for deep learning projects.
Patel K, Bajaj N, Statton B, et al., 2022, Bariatric surgery reverses ventricular repolarisation heterogeneity – mechanistic insights into fat-related arrhythmic risk, British Cardiovascular Society Annual Conference, ‘100 years of Cardiology’, 6–8 June 2022, Publisher: BMJ Publishing Group, Pages: A60-A61, ISSN: 1355-6037
Sivanandarajah P, Wu H, Bajaj N, et al., 2022, Is machine learning the future for atrial fibrillation screening?, Cardiovascular Digital Health Journal, Vol: 3, Pages: 136-145, ISSN: 2666-6936
Atrial fibrillation (AF) is the most common arrhythmia and causes significant morbidity and mortality. Early identification of AF may lead to early treatment of AF and may thus prevent AF-related strokes and complications. However, there is no current formal, cost-effective strategy for population screening for AF. In this review, we give a brief overview of targeted screening for AF, AF risk score models used for screening and describe the different screening tools. We then go on to extensively discuss the potential applications of machine learning in AF screening.
Bajaj N, Carrión JR, Bellotti F, 2022, PhyAAt: physiology of auditory attention to speech datasaet, Publisher: ArXiv
Auditory attention to natural speech is a complex brain process. Itsquantification from physiological signals can be valuable to improving andwidening the range of applications of current brain-computer-interface systems,however it remains a challenging task. In this article, we present a dataset ofphysiological signals collected from an experiment on auditory attention tonatural speech. In this experiment, auditory stimuli consisting ofreproductions of English sentences in different auditory conditions werepresented to 25 non-native participants, who were asked to transcribe thesentences. During the experiment, 14 channel electroencephalogram, galvanicskin response, and photoplethysmogram signals were collected from eachparticipant. Based on the number of correctly transcribed words, an attentionscore was obtained for each auditory stimulus presented to subjects. A strongcorrelation ($p<<0.0001$) between the attention score and the auditoryconditions was found. We also formulate four different predictive tasksinvolving the collected dataset and develop a feature extraction framework. Theresults for each predictive task are obtained using a Support Vector Machinewith spectral features, and are better than chance level. The dataset has beenmade publicly available for further research, along with a python library -phyaat to facilitate the preprocessing, modeling, and reproduction of theresults presented in this paper. The dataset and other resources are shared onwebpage - https://phyaat.github.io.
Bajaj N, Constance TG, Rajwadi M, et al., 2022, Fraud detection in telephone conversations for financial services using linguistic features, Publisher: ArXiv
Detecting the elements of deception in a conversation is one of the mostchallenging problems for the AI community. It becomes even more difficult todesign a transparent system, which is fully explainable and satisfies the needfor financial and legal services to be deployed. This paper presents anapproach for fraud detection in transcribed telephone conversations usinglinguistic features. The proposed approach exploits the syntactic and semanticinformation of the transcription to extract both the linguistic markers and thesentiment of the customer's response. We demonstrate the results on real-worldfinancial services data using simple, robust and explainable classifiers suchas Naive Bayes, Decision Tree, Nearest Neighbours, and Support Vector Machines.
Bajaj N, Carrion JR, Bellotti F, et al., 2021, Analysis of Factors Affecting the Auditory Attention of Non-native Speakers in e-Learning Environments, ELECTRONIC JOURNAL OF E-LEARNING, Vol: 19, Pages: 159-169, ISSN: 1479-4403
- Author Web Link
- Cite
- Citations: 1
Bajaj N, 2021, Wavelets for EEG Analysis, Wavelet Theory, Publisher: IntechOpen
<jats:p>This chapter introduces the applications of wavelet for Electroencephalogram (EEG) signal analysis. First, the overview of EEG signal is discussed to the recording of raw EEG and widely used frequency bands in EEG studies. The chapter then progresses to discuss the common artefacts that contaminate EEG signal while recording. With a short overview of wavelet analysis techniques, namely; Continues Wavelet Transform (CWT), Discrete Wavelet Transform (DWT), and Wavelet Packet Decomposition (WPD), the chapter demonstrates the richness of CWT over conventional time-frequency analysis technique e.g. Short-Time Fourier Transform. Lastly, artefact removal algorithms based on Independent Component Analysis (ICA) and wavelet are discussed and a comparative analysis is demonstrated. The techniques covered in this chapter show that wavelet analysis is well-suited for EEG signals for describing time-localised event. Due to similar nature, wavelet analysis is also suitable for other biomedical signals such as Electrocardiogram and Electromyogram.</jats:p>
Constance TG, Bajaj N, Rajwadi M, et al., 2021, Resolving Ambiguity in Hedge Detection by Automatic Generation of Linguistic Rules, Editors: Farkas, Masulli, Otte, Wermter, Publisher: SPRINGER INTERNATIONAL PUBLISHING AG, Pages: 369-380, ISBN: 978-3-030-86382-1
Paranthaman PK, Bajaj N, Solovey N, et al., 2021, Comparative Evaluation of the EEG Performance Metrics and Player Ratings on the Virtual Reality Games, IEEE Conference on Games (IEEE CoG), Publisher: IEEE, Pages: 604-611, ISSN: 2325-4270
Bajaj N, Requena Carrión J, Bellotti F, et al., 2020, Automatic and tunable algorithm for EEG artifact removal using wavelet decomposition with applications in predictive modeling during auditory tasks, Biomedical Signal Processing and Control, Vol: 55, Pages: 101624-101624, ISSN: 1746-8094
Bajaj N, Bellotti F, Berta R, et al., 2019, Auditory Attention, Implications for Serious Game Design, Editors: Gentile, Allegra, Sobke, Publisher: SPRINGER INTERNATIONAL PUBLISHING AG, Pages: 201-209, ISBN: 978-3-030-11547-0
Bajaj N, Bellotti F, Berta R, et al., 2016, A Neuroscience Based Approach to Game Based Learning Design, Editors: Bottino, Jeuring, Veltkamp, Publisher: SPRINGER INTERNATIONAL PUBLISHING AG, Pages: 444-454, ISBN: 978-3-319-50181-9
Kharbanda M, Bajaj N, 2013, An Exploration of Fractal Art in Fashion Design, 2nd IEEE International Conference on Communications and Signal Processing (ICCSP), Publisher: IEEE, Pages: 226-230
- Author Web Link
- Cite
- Citations: 6
Deora D, Bajaj N, 2012, INDIAN SIGN LANGUAGE RECOGNITION, IEEE 1st International Conference on Emerging Technology Trends in Electronics, Communication and Networking, Publisher: IEEE
Bajaj N, Kashyap R, 2012, EXTENSION OF WAVELET FAMILY IN FRACTIONAL FOURIER DOMAIN, IEEE 1st International Conference on Emerging Technology Trends in Electronics, Communication and Networking, Publisher: IEEE
Bajaj N, Thakur A, 2011, Enhancement of RC5 for image encryption
In recent years, image encryption has grown much to provide information security to end user. There are various algorithms for encrypting the image. RC5 is a block cipher and can be used for image encryption. In this paper, weaknesses of RC5 for image encryption are discussed and analysed. Further, improvement in RC5 algorithm is done. Improvement shows the robustness of algorithm against different attacks. Implementation of enhanced RC5 algorithm has been done on Matlab R2009a and tested with NIST randomness test suit and obtained satisfactory results. © 2011 IEEE.
Bajaj N, 2011, Enhancement of A5/1: Using variable feedback polynomials of LFSR, Pages: 55-60
The Global System for Mobile communication, GSM is the most widely used cellular system in the world, with over a billion customers around the world. GSM was the first cellular system which seriously considered security threats. Previous cellular systems had practically no security, and they were increasingly the subject of criminal activity such as eavesdropping on cellular calls, phone cloning, and call theft. The GSM voice calls are encrypted using a family of algorithms collectively called A5. A5/1 is the stream cipher which encrypts the information transmitted from mobile user. Initially A5 algorithm was kept secret to ensure the security but as algorithm was disclosed many cryptanalytic attacks were proposed and proved the A5 algorithm cryptographically weak. In this paper the modification in A5/1 is proposed to make it robust and resistive to the attacks. Modification is done in two ways (1) feedback tapping mechanism which is enhanced by variable taps for LFSR (Linear Feedback Shift Register) and random shuffling of LFSRs, which increases the complexity of the algorithm without compromising the properties of randomness and (2) clocking rule. The modification has been proposed keeping the ease of implementation in mind. This modified algorithm has been simulated in MATLAB and tested its randomness properties by Randomness test suit given by NIST-National Institute of Standard and Technology and obtained satisfactory results. © 2011 IEEE.
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.