International Journal of Biomedical Engineering and Technology (38 papers in press)

Regular Issues

• Exploration of Functional Connectivity of Brain to assess Cognitive and Physical Health Parameters using Brain-Computer Interface  
  by Murugavalli K, Ramalakshmi R. Ramar, Pallikonda Rajasekaran Murugan, Vaibhav Gandhi 
  Abstract: The neural brain activations are triggered or stimulated by predetermined external influences, including music, videos, audio, meditation and several others. The impact of diverse stimuli on the brain is the core investigation purpose of this research. This paper evaluates the response of the participants in different frequency bands, and also in the various brain regions, to better understand the impact. Sixty five peer-reviewed publications were examined depending on the stimuli: yoga and meditation, music, taste, scent, emotion, imagery and movement. Comprehensive research was undertaken to describe stimuli and their effects on brain functional connectivity. The importance and effect of the infinity walk on changes in humans’ cognitive and physical health parameters, as well as on mental health, is also investigated and perhaps to identify the active brain region in people who have practised the infinity walk. This technique assists in the identification and justification of the truth behind the infinity walk.
  Keywords: brain-computer interface; BCI; electroencephalography; EEG; functional connectivity; FC; infinity walk; figure-of-eight walk.
  DOI: 10.1504/IJBET.2022.10052922
   
  
• A Comprehensive Review on MRI to CT and MRI to PET Image Synthesis Using Deep learning.  
  by Meharban M. S, Sabu M. K, T. Santhanakrishnan 
  Abstract: Image synthesis is the process of generating a synthetic image with desired qualities. Although CT and PET images are suffering from ionising radiation, MRI images are free from such radiation. Due to this fact, we need a system to generate synthetic CT and PET images from MRI images. The system will be helpful to avoid such ionising radiation from CT and PET and makes a better patient treatment workflow. This work reviewed various deep learning synthetic CT and synthetic PET generation methods. More than 75 papers were selected from PubMed and ScienceDirect databases from 2017 to 2021. Recently CycleGAN variants produce better results and no need for paired data. However, an effective evaluation measure was not available to evaluate the efficacy of the proposed works. More blind tests involving radiologists are required to evaluate the visual quality of the synthesised image.
  Keywords: computed tomographic; CT; positron emission tomography; PET; magnetic resonance imaging; MRI; generative adversarial networks; GAN.
  DOI: 10.1504/IJBET.2022.10052929
   
  
• BLDA-CSWDT Autoimmune Thyroid Disease Risks Predictive Model using Machine Learning and Deep Feature Extraction Techniques  
  by Nagavali Saka, S.Murali Krishna 
  Abstract: Nowadays, different thyroid disorders are observed which are affecting the human population worldwide. Hence, to provide suitable treatment and be cost-consuming for the patients, an earlier diagnosis is required. To improve prediction, this paper proposed Bayes-linear discriminant analysis (B-LDA) and cuckoo search based weighted decision tree (CSWDT) models to predict the autoimmune thyroid risk assessment from the obtained dataset. Initially, after pre-processing, the features are extracted using the deep MLP model, and the significant features are fused by using the B-LDA model which overcomes the dimensionality reduction issue. Further, the classification is performed by using the optimised cuckoo search with a weighted decision tree model. In addition, K-fold cross-validation is performed and attains a better accuracy value of 99.5% in thyroid disease prediction.
  Keywords: autoimmune thyroid disease; deep MLP; cuckoo search optimisation; LDA; weighted decision tree; Bayes linear discriminant analysis; B-LDA; cuckoo search based weighted decision tree; CSWDT.
  DOI: 10.1504/IJBET.2022.10053111
   
  
• An efficient way of identification of protein coding regions of Eukaryotic genes using digital FIR filter governed by Ramanujan’s Sum  
  by Subhajit Kar, Madhabi Ganguly 
  Abstract: Finding protein coding regions, i.e., exons in a gene is a complex problem due to its diverse nature. In this paper, a novel FIR filtering governed by Ramanujan’s Sum is proposed for identification of protein coding regions in gene. The efficacy of the designed algorithms is tested on Caenorhabditis Elegans cosmid F56F11.4a, various benchmark datasets like GENSCAN, HMR195, ASP67, and, BG570, and compared to well-established algorithms based on Antinotch, Butterworth, and Comb filters. The numerical conversion of the biological sequence here is an integer sequence and Ramanujan’s Sum always generates a periodic sequence of integer numbers. This results in reduced quantisation error and simple hardware implementation. The evaluation of the designed Ramanujan’s Sum governed filtering is done at the exonic level, nucleotide level, and through ROC plots. The results obtained on gene F56F11.4 attain specificity of 82%, sensitivity 97%, and precision of 85% while the AUC value of ROC curve was calculated as 0.96 square units. These evaluation parameters reveal that the proposed method gives enhanced results while comparing it to other existing exon-finding techniques.
  Keywords: FIR filter; Ramanujan’s Sum; wavelet transform; exons.
  DOI: 10.1504/IJBET.2022.10053343
   
  
• Exploration of fibro-glandular region and breast density classification of digitized mammograms using least square support vector machine  
  by Vijaya Madhavi Mantragar, Christy Bobby T 
  Abstract: Breast tissue density is one of the significant risk-marker for identification of breast cancer in early stage. In the proposed work, fibro-glandular region is explored and classification of breast density as dense and non-dense is performed. Image pre-processing is performed to improve the image quality followed by segmentation of breast region to obtain region of interest (RoI). For the obtained RoI, pseudo colouring is performed to improve image acuity accompanied by R-image extraction and post-processing to obtain fibro-glandular breast tissues. Area, histogram, fractal, grey-level co-occurrence matrix and grey-level run length matrix features are derived from both fibro-glandular and RoI regions and ratiometric value of features are computed. Further, mutual-information-based feature ranking algorithm is applied on the derived ratiometric values and the significant features are identified. These significant features when fed to least square-support vector machine produced average classification accuracy (%) of 86.1
  Keywords: breast density; pseudo colouring; hue saturation value; HSV; ipsilateral; bilateral; LSSVM.
  DOI: 10.1504/IJBET.2022.10053387
   
  
• Socio-economic Implications of COVID-19 in India: Growing Stress and Educational Challenges  
  by Kiron Jayesh, Mahesh Jayaraman, Visweshwaran Baskaran, Nathiya Narayanaraju, Jagannath Mohan, Adalarasu Kanagasabai 
  Abstract: The global pandemic of COVID-19 has been a challenging period for people all over the world. While the main focus during this period has been to stop the transmission of the disease and increase the vaccination drive, a lot of people are going through unspoken problems on their own. This pandemic brought an imbalance in the well-being of families due to various reasons such as lockdown-related stress or financial instability. To evaluate all these impacts on the students, the female homemakers, and the family relationships, three online surveys were self-administered from various validated questionnaires. The survey concluded that female students (Likert scale 3.234) are more distracted from their classes during the lockdown compared to males (Likert scale 2.458) because females also spend a significant amount of time assisting their families with day-to-day chores. The female members of the family are significantly (p<0.05) more concerned about their familys well-being and relations than the males.
  Keywords: COVID-19; family; female homemakers; financial instability; lockdown; pandemic; stress; students; well-being.
  
  
• A NEW METHOD FOR DIAGNOSING EPILEPSY USING DICTIONARY LEARNING  
  by Ghazal Abbasi, Somayeh Saraf Esmili 
  Abstract: Epilepsy is a disorder of the central nervous system in which the activity of nerve cells in the brain is disrupted and leads to seizures An electroencephalograph is often used to diagnose epilepsy, which records the electrical potential generated in the brain In this study, we aim to diagnose epilepsy from the EEG signals using a new method of dictionary learning and sparse coding Most vital signals have a sparse representation that requires a dictionary to represent the sparse signals In the preprocessing, Butterworth and notch filters are used to remove noises, K-SVD algorithm is used to learn a dictionary to find a matrix of dictionary atoms, and in sparse coding, the orthogonal matching pursuit (OMP) algorithm is used to extract the features from the signals The extracted features were entered as input for classification of signals into two groups of epileptic and non-epileptic signals, using the feature vector machine
  Keywords: Dictionary learning; Epilepsy; K-SVD; Orthogonal matching pursuit (OMP); Sparse representation; Sparse coding; Electroencephalograph (EEG).
  DOI: 10.1504/IJBET.2022.10054052
   
  
• Comparative evaluation of geometrical, Zernike moments, and volumetric features of the corpus callosum for discrimination of ASD using machine learning algorithms  
  by Aditi Bhattacharya, Gokul Manoj, Vaibhavi Gupta, Abdul Aleem Shaik Gadda, Dhanvi Vedantham, A. Amilin Prince, Priya Rani, Anandh Kilpattu Ramaniharan, Jac Fredo A. R 
  Abstract: Autism spectrum disorder (ASD) is a neurodevelopmental condition associated with unusual structural changes in brain regions. In this study, we compared the performance of geometrical, Zernike moments, and volumetric features of corpus callosum (CC) to diagnose ASD. The data for the study was obtained from the open-access databases: ABIDE-I and ABIDE-II. Initially, the CC was segmented from the midsagittal view of 2D structural magnetic resonance imaging (sMRI) data using the distance regularized level set evolution (DRLSE). The segmented images were validated with the ground truth using similarity measures. The geometrical and Zernike moments were extracted from the 2D segmented region, and the volumetric features were extracted from 3D images of CC. The features extracted were then used to train support vector machine (SVM), bagging, and random forest (RF) classifiers. The segmented images were highly matched with the ground truth with mean similarity measure values of Sokal and Sneath-II= 0.9928 and Pearson and Heron-II=0.9924, which signified that the DRLSE method was able to segment the CC region successfully. We achieved the highest site-specific classification accuracy of 72.69% using the RF classifier
  Keywords: Autism spectrum disorder; Corpus callosum; sMRI; Level set method; Similarity measures; Geometric features; Volumetric features; Zernike moments; feature selection; Random Forest.
  DOI: 10.1504/IJBET.2022.10054054
   
  
• Integration of radiographic and histological images for the diagnosis of glioblastoma  
  by Fatiha Alim-Ferhat, Linda Ait Mohammed, Mohamed Abdelaziz 
  Abstract: As the number of cancer cases increases, the pathologist’s task becomes increasingly difficult. To classify tumours and define their level of aggressiveness, pathologists are faced with analysing a large number of pathological images, hundreds of thousands of them. Computer-aided methods, including artificial intelligence, can potentially improve tumour classification. It makes sense to implement such a system by taking advantage of the two complementary MRI and histological data. This study proposes to use multiple input convolutional neural networks to predict glioma grade. The proposed method was validated using data from the CPM: RAD-PATH 2020, achieved satisfactory results. We propose a dual path residual convolutional neural network model that trains simultaneously from MRI and pathology images. With this approach, we achieve a validation accuracy of 81%, showing that combining the two image sources yields better overall accuracy.
  Keywords: glioblastoma; digital pathology images; IRM; deep learning; tumour classification.
  DOI: 10.1504/IJBET.2023.10054076
   
  
• Effects of different cushioned insoles on ankle and knee joints biomechanics during load carriage running  
  by Tao Yang, Liangliang Xiang, Shanshan Ying, Jianshe Li, Justin Fernandez, Yaodong Gu 
  Abstract: Load carriage training resulted in substantial injuries among military recruits, particularly in their lower limbs and feet. This study analyzed the phase-specific effects of load carriage with three different material insoles on GRF, angle, and moment of ankle and knee joints during running with military boots. Eighteen male participants were recruited for this study from a local veteran club. A two-way repeated-measures analysis of variance (ANOVA) was conducted to determine statistical effects. The vertical active peak in the ortholite insole group was significantly lower than the control (p=0.002) and cork insoles (p=0.002) with the unloading condition. The control group's ankle dorsiflexion moment was greater than that of the ortholite and cork insoles at zero (p=0.001) and 15 kg load carriage (p=0.001) (46-83% stance). The findings show that the ortholite insole and cork insole improve cushioning performance in the lower limbs and stability of military boots compared with the control insole.
  Keywords: running; load carriage; cushioned insoles; impact force; biomechanics.
  DOI: 10.1504/IJBET.2022.10054077
   
  
• CLAHE Enhanced Hybrid Feature Descriptors for Classification of Acute Lymphoblastic Leukemia in Blood Smear Images  
  by Renuka Tali, Surekha Borra, Vijay Bhaskar Reddy Dinnepu 
  Abstract: Acute lymphoblastic leukemia (ALL) detection through a complete blood count test is often flagged to an expert pathologist for confirmation which is time-consuming, observer-specific, and involve intensive labor. The study proposes an efficient Computer Aided Diagnosis (CAD) method based on image processing and machine learning models to assist doctors in analyzing microscopic images. This study aimed to investigate the combined discriminative qualities of shape and texture features, as well as the best fit feature subset selection technique, to achieve high accuracy and a low false positive rate for classification of healthy and ALL infected leukocyte cell images. The approach begins with preprocessing ALLIDB pictures with the Contrast Limited Adaptive Histogram Equalization (CLAHE) enhancement model, followed by feature extraction using Gray Level Co-occurrence Matrix (GLCM), Local Binary Pattern (LBP), the Bag of Visual Words, and Histogram of Oriented Gradients (HOG). The list of the strongest discriminative feature set, as determined by Sequential Forward Selection (SFS) and Principal Component Analysis (PCA), is then utilized to train an SVM machine learning model.
  Keywords: Acute Lymphoblastic Leukemia; Computer Aided Diagnosis; Image Processing; Leukocytes; Machine Learning; Microscopic Images; Feature Extraction; Feature Selection; SVM.
  DOI: 10.1504/IJBET.2023.10054302
   
  
• Prediction of Wear in Total Knee Replacement Implants Using Artificial Neural Network  
  by Vipin Kumar, Anubhav Rawat, Ravi Prakash Tewar 
  Abstract: The current research work presents the development of an artificial neural network (ANN) based model in order to predict the linear wear depth by using wearing parameters such as non-dimensional contact stresses, sliding distance, and cross-shear ratio in the total knee replacement. The linear wear depth values are computed from knee wear models available in literature. The values of linear wear depth obtained from this model were used for training and testing of an artificial neural network model. Multi-layered feed-forward neural network is used for training and testing of the ANN model. Many architectures of neural networks were tried and the 3-6-6-6-1 architecture possessing 3, 6, and 1 neuron in its input layer, every hidden layer, and output layer respectively was found optimum. The sigmoid activation function was chosen for input and hidden layers, and the linear activation function was chosen for the output layer.
  Keywords: Artificial neural network (ANN); Linear wear depth; Total knee replacement; Wear model; and Cross-shear ratio.
  DOI: 10.1504/IJBET.2023.10054459
   
  
• MRI Segmentation Using Deep Learning Network for Brain Tumor Detection  
  by Ambily N, Suresh K 
  Abstract: Gliomas are a combination of infiltrating tumour cells and vasogenic edema. The abscission and radiation intensified in this region will improve survival. It is difficult to distinguish infiltrating cells with conventional imaging sequences. This paper presents an accurate and automatic method for defining areas of tumour infiltration in peritumoral edema in brain MRI, using a fully convolutional neural network, employing Semantic Segmentation technique. The architecture has a contracting path capturing the features and a symmetric expanding path enabling precise localization similar to U-Net. The expansive path yields a U shaped architecture. The multiparametric pattern analysis from clinical MRI sequences assists in identifying the tumor recurrence in peritumoral edema. This helps resection and strengthening of postoperative radiation therapy. In the proposed model, complete core and enhancing regions in Dice Similarity coefficient metric are (0.99, 0.98.0.98).
  Keywords: DNN; Semantic Segmentation; BrainTumour Detection.
  DOI: 10.1504/IJBET.2023.10054461
   
  
• Study of Biomarker Variation and Severity Prediction in Dementia using Intelligent System  
  by AHANA PRIYANKA, G. Kavitha 
  Abstract: Precise detection of dementia biomarkers in the brain enables early understanding of pathology variations. There is a need to study different dementia biomarker in MR images for its specific changes between normal and severity stages to categorize the prognostic difference. This study is an attempt to utilize an optimized framework with fused radiomic and deep features based on least absolute shrinkage and selection operator (LASSO) by using a hybrid meta-heuristic optimizer for classification. The investigation is attempted on ADNI database. The radiomic and deep features are extracted from the considered biomarkers and then fused. Further, the significant features are obtained using LASSO model. Then, these features are given to hybrid meta-heuristic optimizer with machine learning model for classification. Observed results show that hippocampus along with the brainstem gives higher classification accuracy of 97.87% to identify prognostic differences for considered classes. This quantifiable interpretation might improve clinical assessment.
  Keywords: Dementia; hybrid optimizer; fused feature; biomarker and prognostic difference.
  DOI: 10.1504/IJBET.2023.10054579
   
  
• Investigation of photocatalytic effects and extraction of genomic DNA from Staphylococcus aureus through Fe3O4/SiO2/TiO2 magnetic nanoparticles  
  by Farzaneh Firoozeh, Mohammadreza Rezayee Yazdi, Mohammad Zibaei, Hadiseh Rostami, Ali Sobhani Nasab, Azad Khaledi, Farzad Badmasti 
  Abstract: Staphylococcus aureus has been considered as one of the main pathogens that cause various diseases. Therefore, access to fast and reliable DNA-based methods is crucial for the detection and identification of this bacterium. DNA extraction and purification are fundamental primary steps in almost all molecular biology studies. Therefore, the purpose of this work is utilising Fe3O4/SiO2/TiO2 magnetic nanoparticles to extract genomic DNA of Staphylococcus aureus. This paper contains extracting genomic DNA from standard strain of Staphylococcus aureus ATCC 25923 using Fe3O4/SiO2/TiO2 nanostructures. The quality of extracted DNA was evaluated after electrophoresis on gel agarose, also DNA purity and concentrations were measured by a NanoDrop spectrophotometer. The concentration of genomic DNA extracted by Fe3O4/SiO2 magnetic nanoparticles from Staphylococcus aureus strain ATCC 25923 was 131.635 ng/?L. Also, A260/280 and A260/230 values of mentioned DNA were ranged 1.7 to 1.8 and 2 to 2.2 respectively. The obtained results showed that the DNA extracted by the synthesised magnetic nanoparticles has an acceptable concentration and purity for subsequent molecular biology studies in this bacterium.
  Keywords: Fe3O4/SiO2/TiO2; DNA extraction; Staphylococcus aureus; genomic DNA magnetic nanoparticles.
  DOI: 10.1504/IJBET.2023.10054605
   
  
• Multi-resolution dual-encoder self-constrained brain tumor MR image segmentation algorithm  
  by Weijie Hao, Wenyin Zhang, Yong Wu, Yifang Wang, Yuan Qi, Liang Wu, Ji Chen 
  Abstract: : Efficient segmentation of magnetic resonance (MR) brain tumour images is of the utmost implication for the assessment of the condition. Brain tumours proliferate, metastasize quickly, and easily infiltrate surrounding tissues, and there will be magnetic fields, imaging equipment, and patient movements that affect imaging quality during the imaging process. Therefore, automatic brain tumour MR image segmentation has always been among the most challenging scientific research problems. This paper proposes a multi-resolution dual-encoder self-constrained brain tumour MR image segmentation algorithm that can effectively segment the brain tumour lesion area and normal brain tissue. Experiments show that the dice indexes of brain tumour, cerebrospinal fluid, gray matter, and white matter obtained by this algorithm are: 0.91, 0.78, 0.82, and 0.86, respectively. By comparison, the proposed method demonstrates better efficiency and accuracy and has important implications for brain tumour segmentation.
  Keywords: medical image segmentation; brain tumour MR image; multiple resolution dual encoder; CSAM attention decoder; self-constrained network.
  DOI: 10.1504/IJBET.2023.10054646
   
  
• An advanced wavelet decomposition based denoising technique for de-speckling of all ultrasound images  
  by Mayank Singh, Indu Saini, Neetu Sood 
  Abstract: The ultrasound (US) image is well known for accessibility and low cost. Most importantly it is the only diagnostic technique which is radiation free. But, the presence of speckle noise, thoroughly limits its application for diagnosis. This paper aims to remove the noise using wavelet transformation. The US were transformed using discrete wavelet transform after log transformation. A threshold was obtained using the estimated noise variance for each sub-band. A multi-scale thresholding function was proposed to increase the thresholding flexibility. A large range of US were used (12,400, 926, 780, and 1,000 images of fetus, liver, breast and synthetic images respectively) to evaluate the performance. When compared with other thresholding techniques the proposed method has shown a maximum improvement of 172%, 340%, and 380% in peak signal to noise ratio, mean square error, and structural similarity index. With the referenceless metrics our technique has shown 47% improvement in US quality.
  Keywords: denoising; speckle noise; ultrasound images; wavelet transformation.
  DOI: 10.1504/IJBET.2023.10054699
   
  
• FEM-based Fatigue Analysis on a 4-Bar Polycentric Knee of Above-Knee Prostheses  
  by Mohammad-Reza S. Noorani, Saman Hoseini 
  Abstract: Developing safe and low cost artificial lower limbs with long working life is a necessity to help large population of people with amputation to recover the walking ability. So, in this paper we investigate on optimal design of pin components of a 4-bar polycentric knee used in a above-knee prosthesis. The pins suffer cyclic loads and wearing lead to damage and breakage therein, which is early cause of prosthesis failure. Here, we first exploit the ABAQUS to create a finite element model (FEM), then it is integrated with FEsafe software to obtain a fatigue life prediction according to Morrows fatigue criteria. Materials of SUS-304 stainless steel and Ti-6Al-4V titanium alloy are examined to achieve a fatigue life of over 3,000,000 cycles to meet the requirements of ISO 10328:2006. Pins with the diameter of 10 mm satisfies the requirements at all three stages of stance phase, i.e. heel strike, midstance, and push off, when SUS-304 is selected.
  Keywords: Prosthesis; Above-Knee Amputation; Finite Element Method; Stress Analysis; Morrow’s fatigue criteria.
  DOI: 10.1504/IJBET.2023.10056453
   
  
• A whole 3D Liver Reconstruction for personalized preoperative surgery based on FCN U-Net model segmentation  
  by Amina BENAHMED, Kamila Khemis, Salim Loudjedi 
  Abstract: The vascular mapping of liver is part of personalized medicine which allows optimal management of the patient during surgery. The work that we propose allows 3D reconstruction of the liver with its vascular tree. The main steps followed to achieve this processing are image pre-processing based on Hounsfield windowing, automatic segmentation of liver, then vascular tree extraction. Liver automatic segmentation is a challenging stage because of the inter-patient variability of the liver shape and it similar gray level with neighboring organs. Deep learning aproach fits to this problematic. We applied Fully Convolutional Networks “U-Net” with 38 layers, 40 connections, Adam's algorithm optimizer and learning rate of 0.001. We tested this model on two CT scan databases : 3Dircadb and Task08_HepaticVessels. The results obtained, validated quantitatively, are very satisfactory (Dice coefficient 98% and Loss function 2%), comparable to the literature and confirm the robustness of U-Net for automatic liver segmentation.
  Keywords: Deep Learning; Fully Convolutional Networks (FCNs); U-Net,Liver surgery; Personalized medicine ,3D Reconstruction; Segmentation.
  DOI: 10.1504/IJBET.2023.10055084
   
  
• Amalgamation of wavelet transform and neural network for COVID-19 detection  
  by Madhu Jain, Renu Sharma 
  Abstract: A zoonotic natured virus affecting almost every part of the globe is COVID-19. Early detection of such disease may lead to curable affairs. Since then, many research institutes have been trying to find effective methods for detecting and curing COVID-19. Real-time polymerase chain reaction test is also a method used for detection of the COVID-19. But, due to its accuracy rate and availability of kit it is not relied on. Here, a combination of machine learning and wavelet transform based algorithm for chest X-ray classification is proposed. Image pre-processing is done using wavelet transform and further the classification is done using convolution neural network. It is a multi-class classifier, which will classify whether input image is COVID-19 affected, pneumonia or not affected. The dataset collected for this study from an open-source repository. It comprises 2,550 images of each class. For quantitative analysis of the proposed architecture, parameters such as accuracy, precision, F1 score, recall and sensitivity are measured.
  Keywords: image classification; image enhancement; convolutional neural networks; CNN; X-rays; wavelet transforms; COVID-19.
  DOI: 10.1504/IJBET.2023.10055996
   
  
• Identification of Type-2 Diabetes by Electrocardiogram signal using Flexible analytical wavelet transform  
  by Bhanupriya Mishra, NeelamShobha Nirala 
  Abstract: Type-2 diabetes mellitus (T2DM) is a lifelong metabolic disease with worldwide prevalence. It can drastically decrease the life expectancy of any subject with a huge economic burden. The present study aimed to create a non-invasive and economical tool for automatic detection of T2DM using electrocardiogram (ECG) signals. The flexible analytic wavelet transform is used to evaluate the ECG by decomposing it into predictable sub-bands. Statistical and time-domain features were extracted from each sub-band. Different feature selection techniques were applied to obtain the most relevant features. The top nine features selected by the one-R attribute eval feature selection technique were fed into the various types of machine learning classifiers. In tested classifiers, the fine k-nearest neighbour and optimisable KNN classifiers have shown the highest average accuracy of 94.94% and 94.61% respectively. The results suggest that the proposed approach provides an efficient non-invasive T2DM detection method in regular applications.
  Keywords: electrocardiogram signal; flexible analytical wavelet transform; FAWT; type-2 diabetes; feature extraction; feature selection methods; machine learning techniques.
  DOI: 10.1504/IJBET.2023.10056600
   
  
• Plaque Rupture in Stenotic Coronary Artery Model: A Numerical Study  
  by Md Rakibuzzaman, Hyoung-Ho Kim, Sang-Ho Suh, A.K.M. Parvez Iqbal, Byoung-Kwon Lee, Hyuck Moon Kwon 
  Abstract: Atherosclerosis is a type of cardiovascular disease in which a wounded endothelial cell triggers a series of biochemical responses to repair the damage. As a mechanical response to the arterial wall, wall-shear-stress (WSS) is well investigated to become a significant contributing factor for atherogenesis and the development of atherosclerosis. This vascular wall behaviour could be utilised to predict plaque biomechanical instability and rupture susceptibility. Plaque has a significant function in causing blood vessel ruptures. This study used fluid-structure interaction (FSI) which is the combination of finite volume method (FVM) and finite element method (FEM) to investigate the deformable structure during internal fluid flow. Therefore this particular approach has been implemented to investigate the features of stenosed with and without plaque ruptures under various situations. The von Misses stress (VMS) and WSS were determined. Results revealed that on both sides of arterial branch, higher WSS took place than below the plaque. Moreover, the highest intensity of VMS was roughly equivalent, when the distance between the plaque and the inner wall of the vessel was less than 65 ?m, but the greater the distance, the greater the stress.
  Keywords: coronary artery; plaque rapture; FSI analysis; von misses stress; VMS; wall-shear-stress; WSS.
  DOI: 10.1504/IJBET.2023.10056792
   
  
• Novel approaches for classification COVID-19 and pneumonia disease from CT scans using radiomics features.  
  by Linda Ait Mohammed, Fatiha Alim-Ferhat, Mohamed Abdelaziz 
  Abstract: COVID-19 is a highly infectious and fatal pneumonia-like disease. Despite the time-consuming nature of RT-PCR (reverse transcription-polymerase chain reaction), it remains a proven testing method for detecting coronavirus infection. Therefore, COVID-19 screening can be adopted using X-ray and computed tomography (CT) images of an individual's lungs to assist the traditional RT-PCR method in making an accurate clinical diagnosis. This imaging-based diagnosis will facilitate the detection of coronavirus infection and provide insight into the status of the disease, its form, and its degree of risk to the patient's life. CT scanning has become the benchmark test for COVID-19 and other pneumonia diseases. It allows visualisation and precise localisation of lesions in the lungs and branches. This study aims to classify COVID-19 among other types of pneumonia using a new preprocessing method, a set of parameters relevant to matching the random forest model, and a hybrid segmentation method based on level set and morphological tools. With an accuracy of 98.30%, an AUC of 0.98 for classification, and a dice score of 76% for segmentation, this method yielded promising results comparable to those found in the literature and allowed automatic and accurate deferral between COVID-19 and other pneumonia infections.
  Keywords: CT scans; radiomics features; pneumonia; COVID-19; random forest.
  DOI: 10.1504/IJBET.2023.10057050
   
  
• An Efficient Brain Tumor Segmentation Approach Using Cascade Convolutional Neural Networks  
  by Ahmed Hechri, Abdelrahman Hamed, Ahmed Boudaka 
  Abstract: Brain tumours pose a significant threat to human life, as they are a major cause of death. Early detection of brain tumours is vital to improve treatment and reduce mortality rates. Automatic segmentation using deep learning methods is crucial for clinical evaluation and treatment planning, but remains challenging due to the diverse tumour locations and structures. In this work, we employed the concatenation of two different convolutional neural networks: the two-pathway architecture and the inception architecture. We also utilised a patch-based technique that combines global and local features to predict the output region. Our proposed system achieved dice scores of 0.86, 0.81, and 0.75 for the whole tumour, tumour core, and enhancing tumour on the BraTS 2018 dataset, respectively. For BraTS 2019, the dice scores were 0.85, 0.79, and 0.67, respectively. Compared to existing state-of-the-art CNN models, our proposed system significantly improves both qualitative and quantitative brain tumour segmentation results.
  Keywords: MR images; tumour segmentation; convolutional neural network; CNN; two pathways; inception.
  DOI: 10.1504/IJBET.2023.10057111
   
  
• Atherosclerotic Plaque Segmentation Using Modified UNet with Hybrid Pooling Layers  
  by Soni Singh, Pankaj Jain, Neeraj Sharma, Mausumi Pohit 
  Abstract: Atherosclerotic plaque segmentation is a vital task in cardiovascular image processing. Fuzzy nature of the carotid images makes it difficult to extract vital features from the plaque tissue region. UNet deep learning models use max-pooling layers for extraction of feature maps and are quite effective in medical image segmentation. In this study, we hypothesised that the UNet model with a hybrid pooling layer consisting of average pooling layer and max-pooling layers could exert more control on feature selection, and therefore be more effective solution for carotid plaque segmentation. We used a public database of 66 B-mode ultrasound images of the carotid artery for our experiments. We experimented with four cases of modified UNet model using a hybrid pooling layer with four different values of , and compared it with the standard UNet model. Modified UNet model with hybrid pooling layers shows nearly 5% improvements in DSC and JI values.
  Keywords: UNet; hybrid pooling layer; atherosclerosis; carotid plaque segmentation; deep learning.
  DOI: 10.1504/IJBET.2023.10057116
   
  
• EEG based variable node functional network comparison for multiclass brain disease detection using stacked ensemble model  
  by Mangesh Kose, Mitul Kumar Ahirwal, Mithilesh Atulkar 
  Abstract: The brain connectivity network (BCN) is considered to be an effective approach for analyzing brain functionality. The EEG-based BCN considers electrodes as a node and functional similarity between EEGs from corresponding nodes as edge. The EEG dataset available for the evaluation might contain variable number of nodes. The variable number of nodes provides biased results while performing graph classification. Hence, the study proposed a strategy to mitigate the aforementioned challenge. The proposed method characterises variable node BCN with the help of network level metrics as a feature vector. The extracted metrics characterises the network as a whole and do not rely on the number of nodes. Two public datasets, with 16 electrodes and 19 electrodes EEG data, are used to test the suggested method. The classification is performed with the stacked-ensemble classification technique. Finally, the quantitative analysis of the proposed approach represents a significant performance with the 92.34% classification accuracy.
  Keywords: brain connectivity network; BCN; electroencephalogram; EEG; graph theory-based metrics; SMOTE; stacked-ensemble classification.
  DOI: 10.1504/IJBET.2023.10057222
   
  
• Heart rate monitoring in COVID-19 patients: Methods, Performance, and Evaluation. A Review  
  by Maham Sarvat, Suhaib Masroor, Zain Anwar Ali, Sobia Shabbir, Bilal Ahmad 
  Abstract: Heart rate monitoring plays a key in assisting numerous cardiac disease. Heart rate devices improve the quality of patient care and able to perform continuous heart rate monitoring during physical activities and fitness tracking. This review paper discuss the effects of COVID-19 on heart rate, compares different heart rate measurement methods, and discuss the key signal processing techniques. Also propose a new system which can be used for the COVID-19 patients for continuous monitoring of heart rate. The proposed system offers wireless wristband monitoring of heart rate through the internet of things. The proposed concept is executed with the ESP8266 WIFI module, Arduino controller, and the real-time monitoring can be visualised by the Thing speak account which also monitors the tachycardia bradycardia condition of the heart. Thus, the proposed framework is proved to be more reliable for Heart rate monitoring during physical activity of high intensity and for fitness tracking.
  Keywords: cardiovascular diseases; CVD; heart rate; wireless monitoring; COVID-19; signal processing; wearable.
  DOI: 10.1504/IJBET.2023.10057474
   
  
• ADB-Net: An Attention-based Dilated Bridge model for fully automatic intra-tumor segmentation of Gliomas.  
  by Radhika Malhotra, Barjinder Saini, Savita Gupta 
  Abstract: Glioma segmentation is a complicated task due to the non-uniform and unstructured morphology of gliomas. Moreover, the requirement for trainable parameters grows exponentially with architectural advancements. In this work, a light-weight and modified attention-based dilated bridge net (ADB-Net) architecture is developed for accurate segmentation of glioma sub-regions. The proposed work has four main benefactions. Firstly, the bridging network in D-Link is enhanced by incorporating a deformed residual connection after each dilation convolutional block to promote the mapping of multi-level information between encoding/decoding units. Secondly, a proper selection of the dilation factor is included for dilated convolutional blocks. Thirdly, four modified attention skip modules (ASM) are also introduced to provide recognition of varied-sized tumours. Lastly, the proposed architecture outperforms its baselines while minimising the number of trainable parameters by more than 50%. It achieves dice scores for the complete tumour, tumour core, and enhancing tumour as 0.971, 0.979, and 0.962, respectively.
  Keywords: D-Link; attention; glioma; segmentation; loss function; BraTS; convolutional neural networks; CNN; attention skip modules; ASM.
  DOI: 10.1504/IJBET.2023.10057482
   
  
• The effect of heel height on the Achilles tendon and muscle activity in Latin dancers during a special-landing task  
  by Fengfeng Li, Datao Xu, Huiyu Zhou, Bálint Kovács, Minjun Liang 
  Abstract: Latin shoes are considered an essential part of Latin dance, but wearing the shoes may contribute to dance injuries. The primary purpose of this article was to compare the biomechanics and muscle activity differences between 7.5 cm heel-height Latin shoes and 5.5 cm Latin shoes based on the bounce step during Jive performance. One-dimensional statistical parametric mapping (SPM1D) was used to analyse the data. 7.5 cm Latin shoes showed significantly greater peak dorsiflexion during the 4.56%, as well as the VALR and peak vertical force significantly increased compared to 5.5 cm Latin shoes. These results suggest that a higher heel leads to a greater lower limb muscle activation in dancers and enhanced medial and lateral ankle muscle control. The lower extremity strength of dancers and the ability to control their limbs should be considered an essential element when reasonably raising Latin heels.
  Keywords: Latin dance footwear; biomechanics; electromyography; heel-height.
  DOI: 10.1504/IJBET.2023.10057554
   
  
• Video Analytics based multi-symptoms system for determining progression of Parkinson Disease  
  by Jignesh Sisodia, Dhananjay Kalbande 
  Abstract: : Parkinson disease is the second most common neurodegenerative disease, and its symptoms tend to increase progressively and affect numerous parts of the body. Parkinson disease patients suffer from symptoms such as rigidity in the body, Bradykinesia, tremors in hands, facial tremor, and freezing of gait. Traditionally assessment of Parkinson disease is based on clinician observation on the severity of symptoms of patients during the visit. Symptoms of Parkinson are highly episodic and cannot be completely observed at the doctors clinic. With the effect of COVID-19, physical visit of the elderly population to the clinic is considered unsafe. Video-based assessment at the patients home led to the solution of avoiding the patient to be exposed to the outside world. We propose a non-invasive video analytics-based assessment of progression of Parkinson disease based on finger tapping and tremor using UPDRS scale. We also propose a video-based technique utilising deep learning and convolutional neural networks which analyse the gait characteristics of patients to identify Parkinson. We intend to distinguish a healthy subject and progression of disease at different stages. These techniques can assist clinical experts for examination of patients to identify the progression of the disease.
  Keywords: Parkinson disease; video analytics; deep learning; convolutional neural network.
  DOI: 10.1504/IJBET.2023.10057647
   
  
• Effects of Music Therapy, Mantra Therapy, and Alpha Beta Binaural Beats on Human Energy Fields using RFI Technology  
  by Karan Sharma, Sudhanshu Choudhary 
  Abstract: This technical paper is based on research on human energy fields, also known as aura, which is radiated by the body externally under the application of various therapies. Music therapist helps patients to improve and maintain their health by using music. The internal energy state of the body is required to be measured to visually confirm a person's symptoms. It can be achieved with aura measurements and it is done using resonant field imaging (RFI) technology. Here, the music, mantra sounds, and alpha-beta binaural beats are considered as external therapies for aura interpretation. By using RFI, the energy field can be accurately identified and the function of a specific region in the human body. The analysis shows the health condition and energy of the human body samples before and after applying the therapies such as Gayatri mantra chanting, Mahamrityunjaya mantra chanting, listening to soft flute music, and listening to alpha beats.
  Keywords: music therapy; mantra therapy; alpha-beta binaural; energy field; resonant field imaging; RFI.
  DOI: 10.1504/IJBET.2023.10058077
   
  
• A Mixed Loss Joint Approach Based Deep Learning Strategy to Enhance SNR and Resolution of Arterial Spin Labeling MRI  
  by Shyna A, Ushadevi Amma C, Ansamma John, Kesavadas C, Bejoy Thomas, Anagha T. J 
  Abstract: The arterial spin labelling (ASL) MRI is a non-invasive technique to quantify cerebral blood flow (CBF) to diagnose various neurological disorders. However, low spatial resolution, which produces partial volume effects and low SNR of ASL images, hampered CBF quantification, which can be corrected with appropriate post-processing approaches. The deep learning-oriented super-resolution technique is a promising way to enhance the overall image quality of ASL-MRI. The proposed 2D CNN architecture uses a residual dense block (RDB) as the basic building unit and a mixed loss joint strategy to increase SNR and resolution while retaining edge information. The proposed model is validated on simulated ASL-MRI data generated from structural images of ADNI in terms of different metrics such as RMSE, PSNR, and SSIM and using clinical data by two independent radiologists in terms of the visual quality control score (VQCS). Experimental results show that the proposed model outperforms state-of-the-art methods.
  Keywords: arterial spin labelling; ASL; cerebral blood flow; CBF; partial volume effect; super resolution; deep learning.
  DOI: 10.1504/IJBET.2023.10058312
   
  
• EEG and Speech Signal Based Multi-class Recognition Manoeuvre by exploiting a Hyb-SGTS and a Dual stage deep CNN architecture for an early diagnosis of HC, AD and PD Neurological Diseases  
  by Chetan Balaji, D.S. Suresh 
  Abstract: Alzheimer's disease (AD) and Parkinson’s disease (PD) are the neurodegenerative illness of the brain that affects the nerve system of brain. The early detection and diagnosis of these disorders are essential in customizing patient's treatment plans to better manage the development and progression, this helps to achieve maximum treatment benefit before mental deterioration occurs. In this paper, the early detection of AD with PD diseases utilizing hybrid seagull and tunicate swarm (Hyb-SGTS) optimization based feature selection technique with dual stage deep convolutional neural network (DSDCNN) are proposed. The experimental results obtained from the employed optimization method yields a better performance and provide maximal classification accuracy with improved efficiency when compared with the existing methods, such as Support Vector Machines (SVM), Na
  Keywords: Neurodegenerative diseases; Alzheimer's disease; Parkinson's disease; Healthy controls; hybrid seagull and tunicate swarm optimization algorithm; dual stage deep convolutional neural network.
  DOI: 10.1504/IJBET.2023.10058868
   
  
• Unveiling the potential of complex network in coronavirus proliferation study  
  by S. Sankararaman 
  Abstract: The development of novel methods for understanding virus replication is the need of the time of the COVID-19 pandemic. The present work proposes a novel surrogate graph-based method for understanding SARS-CoV-2 replication. Constructing a time history pattern (THP) matrix from the video of the virus interaction with normal cells, the inertia moment (IM) and complex network features are determined. The variation of IM and the graph features are correlated with the proliferation of SARS-CoV-2. Thus, the work suggests the possibility of complex network and IM analyses to understand the kinetics of the virus infection.
  Keywords: graph theory; coronavirus; proliferation; inertia moment; complex network.
  DOI: 10.1504/IJBET.2022.10051583
   
  
• Development of a mathematical correlation for polydisperse non-spherical drug particle deposition in the human upper respiratory system  
  by Sanaz Aghaei, Hassan Khaleghi 
  Abstract: Estimating the drug particle deposition in the upper respiratory system is essential to provide more effective treatment for respiratory diseases. This study numerically investigates the effect of both particle size distribution and particle shape on the total deposition efficiency in the human upper respiratory system. To investigate the effect of particle size distribution, spherical monodisperse and polydisperse particles are compared. Non-spherical polydisperse particles are also studied to investigate the effect of sphericity. It is concluded that by decreasing particle size and increasing particle sphericity, the total deposition efficiency decreases. This means that more particles escape from the upper airways to the bronchi and bronchioles. Therefore, for lung disease treatment, finer particles with higher sphericity are more suitable. Furthermore, a mathematical correlation is developed to represent the total deposition efficiency as a function of Stokes number and sphericity. This correlation estimates the deposition of both spherical and non-spherical polydisperse particles.
  Keywords: polydisperse particles; non-spherical particles; total deposition efficiency; mathematical correlation; idealised upper respiratory model.
  DOI: 10.1504/IJBET.2022.10051642
   
  
• Retina blood vessels segmentation by combining deep learning networks  
  by Mohamed Elssaleh Bachiri, Adel Rahmoune, Fayçal Rahmoune 
  Abstract: In this paper, we propose two deep learning architectures for the segmentation and detection of the vascular networks of blood vessels in fundus images. First, we combined VGG16 with U-net, then, we used Resnet 34 in combination with U-net. Both architectures employ an encoding and a decoding path. In this paper, we used the DRIVE and STARE databases. After applying VGG 16+U-net on the DRIVE database, we obtained the accuracy value of 0.96955, 0.79929 sensitivity, 0.98624 specificity, 0.9805 recall, and 0.9833 F1-score. We applied VGG 16+U-net on STARE database and we got 0.95259 accuracy, 0.89996 sensitivity, 0.95530 specificity, 0.9933 recall, and 0.9742 F1-score. Concerning Resnet 34 + U-net, we got the value of 0.9692 accuracy, 0.7859 sensitivity, 0.9870 specificity, 0.9794 recall, and 0.9832 F1-score after applying on DRIVE database. Moreover, we got 0.9363 accuracy, 0.9335 sensitivity, 0.9246 specificity, 0.9961 recall, and 0.9649 F1-score after we applied Resnet 34+U-net on STARE.
  Keywords: retinal segmentation; convolution neuron network; U-Net; deep learning; VGG 16; Resnet 34.
  DOI: 10.1504/IJBET.2022.10051639
   
  
• Automated hard exudate segmentation using neural encoders and attention mechanisms for diabetic retinopathy diagnosis  
  by Pratiksha Gawas, Sowmya Kamath S. 
  Abstract: Diabetic retinopathy (DR) is a complication caused by increased blood glucose levels, which causes retinal damage in diabetic patients' eyes. If not discovered and treated early, it can lead to vision loss. Hard exudates (HE) are one of its characteristic signs. Identification of HE is a paramount step in early diagnosis of DR. In this work, the suitability of U-Net-based deep CNN with different encoder configurations and attention gates (AG) is experimented, for HE segmentation. The proposed models were benchmarked on the standard IDRiD dataset. To overcome the challenges related to the limited dataset, data augmentation techniques were also applied to generate image patches and used for model training. Extensive experiments on the dataset revealed that U-Net with AG achieved an accuracy of 98.8%. The U-Net with ResNet50 as the encoder backbone achieved an accuracy of 98.64%. The findings show that the presented models are effective and suitable for early-stage clinical diagnosis.
  Keywords: hard exudate; hard exudate segmentation; neural encoders; attention mechanism; diabetic retinopathy; diabetic retinopathy prediction; medical informatics; deep learning.
  DOI: 10.1504/IJBET.2022.10052447
   
  
• Asymmetry in people with transtibial and transfemoral amputation for the activities of daily living - a review  
  by Mohammad Shah Faizan, Swati Pal 
  Abstract: Asymmetry between the prosthetic and the intact leg may cause discomfort and seriously deteriorate people's quality of life. It is important to know the current status of asymmetry involved in the recent leg prosthetics so that efforts will be made to minimise it. In this paper, 31 articles that focus on the asymmetry in people with unilateral transtibial and transfemoral amputation were screened using PRISMA. The articles were further reviewed and computed for the symmetry index. The results revealed the presence of a high level of asymmetry during various activities performed. The level of asymmetry decreases with the activities associated with increasing speed. The microprocessor-controlled prosthetics have lesser asymmetry as compared to the non-microprocessor-controlled. The recent prosthetics were not fully effective in minimising the asymmetry, thus, more advanced research is needed in the design of prosthetics, by taking into consideration the varied nature of daily activities.
  Keywords: leg prosthetics; microprocessor-controlled prosthetics; non-microprocessor-controlled prosthetics; unilateral amputation; asymmetry; symmetry index; activities of daily living; biomedical devices.
  DOI: 10.1504/IJBET.2022.10052342