Showing papers in "International Journal of Biomedical Engineering and Technology in 2020"

Design of artificial pancreas based on the SMGC and self-tuning PI control in type-I diabetic patient

Abstract: Optimal closed loop control of blood glucose (BG) level has been a major focus for the past so many years to realise an artificial self-regulating insulin device for type-I diabetes mellitus (TIDM) patients. There is urgency for controlled drug delivery system to design with appropriate controller not only to regulate the blood glucose but also for other chronic clinical disorders requiring continuous long-term medication. As a solution to the above problem, a novel optimal self-tuning PI controller is proposed whose gains dynamically vary with respect to the error signal. The controller is verified with a nonlinear model of the diabetic patient under various uncertainties arises in various physiological conditions and wide range of disturbances. A comparative analysis of self-tuning PI controller performance has been done with the sliding mode Gaussian control (SMGC) and other optimal control techniques. Obtained results clearly reveal the better performance of the proposed method to regulate the BG level within the normoglycaemic range (70-120 mg/dl) in terms of accuracy, robustness and handling uncertainties.

Continuous user authentication using multimodal biometric traits with optimal feature level fusion

Abstract: The biometric process demonstrates the authenticity or approval of an individual in view of his/her physiological or behavioural characteristics. Subsequently, for higher security feature, the blend of at least two or more multimodal biometrics (multiple modalities) is requiring. Multimodal biometric technology gives potential solutions for continuous user-to-device authentication in high security. This research paper proposed continuous authentication (CA) process using multimodal biometric traits considers finger and iris print images to various feature extraction process. At that point, features are extracted into optimal feature level fusion (FLF) process. The final feature vector is acquired by concatenating directional information and centre area features. Disregard the optimal feature process the inspired fruit fly optimisation (FFO) model is considered, and then these model is fused into authentication procedure to find the matching score values (Euclidian distance) with imposter and genuine user. From the approach, results are accomplished most extreme accuracy, sensitivity and specificity compared with existing papers with better FPR and FRR value for the authentication process. The result shows 92.23% accuracy for the proposed model when compared to GA, PSO which is attained in MATLAB programming software.

Review on retinal blood vessel segmentation - an algorithmic perspective

Abstract: Medical image processing has progressed in leaps and bounds with the advent of radical medical imaging modalities. Blood vessel segmentation from the retinal fundus image is very useful in the diagnosis of chronic vascular diseases, arteriosclerosis, diabetic retinopathy, hypertension, etc. This review paper aims to bring out the existing algorithms developed for the segmentation of vessels in the fundus. This paper covers various segmentation approaches categorised under template matching, multi-scale approach, region growing, active contour model and pattern recognition methods. Pattern recognition is further classified as unsupervised, supervised and deep learning approaches. Performance metrics such as accuracy, specificity, sensitivity, and area under the curve measures for these algorithms performed on the appropriate retinal databases are tabulated and discussed. Moreover, this paper discusses the impact of retinal blood vessel segmentation in screening cardiovascular and cerebrovascular diseases. Also, this paper recommends a universal blood vessel segmentation algorithm for the medical vasculature images.

Machine learning algorithms using binary classification and multi model ensemble techniques for skin diseases prediction

Abstract: Skin disease has more touchiness as compared to any other disease. Regular skin issues are dermatitis. The main focus of this research paper will be on dermatology database which contains different eryhemato-squamous diseases class as psoriasis, seboreic dermatitis, lichen planus, pityriasisrosea, cronic dermatitis and pityriasisrubrapilaris. Each record is a collection of 33 attributes which are linear values and one attribute of them is nominal. The 75% of the dataset utilise for demonstrating and keep down 25% for approval. The purpose of this article is to achieve the best-performing classifier that will communicate in the collection of dermatological information. Therefore, k-nearest neighbours and support vector machines are used. By using ten-fold cross validation and assess calculations utilising the accuracy metric. This is a gross metric which will prove the developed model is best one.

Implementation of circular Hough transform on MRI images for eye globe volume estimation

Abstract: Eye globe volume estimation has gained attention in both medical and biomedical engineering field. However, most of the methods used manual analysis which is tedious and prone to errors due to various inter- or intraoperator variability studies. In the present study, we estimated the volume of eye globe, in MRI images of normal eye globe condition using the circular Hough transform (CHT) algorithm. To test the performance of the proposed method, 24 magnetic resonance images which constitute 14 males and ten females (normal eye globe condition) with T1-weighted MRI images are randomly selected from the database. The mean (±SD) of the eye globe volume for male was 6.75 (±0.82) cm3 and 6.17 (±0.74) cm3 on the right and left side respectively. For female, the mean (±SD) of the eye globe volume was 6.24 (±0.70) cm3 on the right side and 6.22 (±0.72) cm3 on the left side. The proposed method have provided the new insight in estimating eye globe volume with a new and unique, automated approach.

A tumour segmentation approach from FLAIR MRI brain images using SVM and genetic algorithm

Abstract: This paper puts forth a framework of a medical image analysis system for brain tumour segmentation. Image segmentation helps to segregate objects right from the background, thus proving to be a powerful tool in medical image processing. This paper presents an improved segmentation algorithm rooted in support vector machine and genetic algorithm. SVM is the basis technique used for segmentation and classification of medical images. The MRI database used consists of FLAIR images. The proposed system consists of two stages. The first stage performs preprocessing the MRI image, followed by block division. The second stage includes - feature extraction, feature selection and finally, the SVM-based training and testing. The feature extraction is done by first order histogram and co-occurrence matrix and GA using KNN is used to select subset features. The performance of the proposed system is evaluated in terms of specificity, sensitivity, accuracy, time elapsed and figure of merit.

Engineering approaches for ECG artefact removal from EEG: a review

Abstract: Electroencephalograms (EEGs) signal, obtained by recording the brain waves are used to analyse health problems related to neurology and clinical neurophysiology. This signal is often contaminated by a range of physiological and non-physiological artefacts, which leads to a misinterpretation in EEG signal analysis. Hence, artefact removal is one of the pre-processing step required for clinical usefulness of the EEG signal. One of the physiological artefact, i.e., electrocardiogram (ECG) contaminated EEG can affect the clinical analysis and diagnosis of brain health in various ways. This paper presents a review of engineering approaches adopted till date for ECG artefact identification and removal from contaminated EEG signal. In addition, the technical approach, computational extensiveness, input requirement and the results achieved with every method is discussed. Along with that, the feasibility study for real time implementation of the algorithms is discussed. Also, an analysis of these methods has been reported based on their performance.

Pose and occlusion invariant face recognition system for video surveillance using extensive feature set

Abstract: Face recognition presents a challenging problem in the field of image analysis and computer vision. Different video sequences of the same subject may contain variations in resolution, illumination, pose, and facial expressions. These variations contribute to the challenges in designing an effective video-based face-recognition algorithm. In this proposed method, we are presenting a face recognition method from video sequence with various pose and occlusion. Initially, shot segmentation process is done to separate the video sequence into frames. Then, face part is detected from each frame for further processing. Face detection is the first stage of a face recognition system. After detecting the face exactly the facial features are extracted. Here SURF features, appearance features, and holo-entropy is used to find out the uniqueness of the face image. The active appearance model (AAM) can be used to find out the appearance-based features in the face image. These features are used to select the optimal key frame in the video sequence which is based on the supervised learning method, modified artificial neural network (MANN) using bat algorithm. Here bat algorithm is used for optimising the weights of neurons. Finally, based on the feature library, the face image can be recognised.

Phase-based frame interpolation method for videos with high accuracy using odd frames

Abstract: In this venture, an imaginative unpredictability movement that has low vector preparing calculation toward the end side is proposed for movement repaid video vector frame interpolation. By handling this calculation, we typically demonstrates the issues of broken edges and disfigured structures in an inserted frame by progressively refining movement vectors on various square sizes. In the proposed strategy, the info has been taken as video rather than pictures in the current framework and the recuperation yield is taken as pictures and further process has been experienced to get the yield as video. There are some extraordinary systems in this strategy, for example, stage-based interpolation method, multistage movement repaid interpolation, and so forth is usually used to get high reason picture with diminished haze in the pictures to get the unmistakable picture of the information videos. Exploratory outcomes will demonstrate the visual quality, toughness in the video groupings containing quick moving video.

Plasma cell identification based on evidential segmentation and supervised learning

Abstract: Myeloma disease is among the most common type of cancer, it is characterised by proliferation of plasma cells, kind of white blood cell (WBC). Early diagnosis of the disease can improve the patient's survival rate. The manual diagnosis involves clinicians to visually examine microscopic bone marrow images for any signs of cells proliferation. This step is often laborious and can be highly subjective due to clinician's expertise. Automatic system based on WBC identification and counting provides more accurate result than manual method. This system is mainly based on three major steps: cell's segmentation, cell's characterisation and cell's classification. In the proposed system, microscopic images of bone marrow blood are segmented by the combination of watershed transform and the evidence theory, the segmented cells are characterised with shape and colour texture features, and then classified into plasma cells or not plasma cells with three supervised classifiers; support vector machines, K-nearest neighbour and decision tree. Experimental results show that the recognition of plasma cells with the K-nearest neighbour achieved 97% of correct rate with 100% of specificity.

Systematic CO2 monitoring using machine learning enabled WSN to develop the anti-hazard strategies for the future

Abstract: The life cycle of an organism does not complete without carbon dioxide (CO2). CO2 is an essential peculiar flavour ingredient gas that drives the world. On the other hand, the impact of excessive CO2 affects the atmosphere with rapid climate changes, greenhouse effect, rain with corrosive implications, and many more hazards. The excessive CO2 influences the natural resource and depletes it to the hazard nature. So the atmosphere needs attention towards monitoring under various conditions. In this paper, the technology of wireless sensor network (WSN) is used for monitoring the CO2 and other gases through which data is stored and fed to the machine learning methodology for future strategies. This paper gives an analysis of the feasibility and effectiveness of the various methods. During the investigation, the generalised regression neural network (GRNN) is identified to be a suitable algorithm for the learning and anti-hazard strategies for monitoring the CO2. The results produced by the GRNN method are promising, which particulates up to 96% of accuracy compared to other algorithms.

2D MRI intermodal hybrid brain image fusion using stationary wavelet transform

Abstract: Medical image fusion involves combination of multimodal sensor images to obtain both anatomical and functional data to be used by radiologists for the purpose of disease diagnosis, monitoring and research. This paper provides a comparative analysis of multiple fusion techniques that can be used to obtain accurate information from the intermodal MRI T1-T2 images. The source images are initially decomposed using stationary wavelet transform (SWT) and the approximations are reconstructed by discrete curvelet transform (DCT), the SWT and DCT are good for point and line discontinuities. The decomposed MRI approximation and detail components are fused using the different fusion rules. The reconstructed fused image is used to accurate identification of brain diseases such as 95.7% of brain lesion, 97.3% of Alzheimer's disease and 98% of brain tumour. Various performance parameters are evaluated to compare the fusion techniques and the proposed method which provides better result is analysed.

Enhanced cache sharing through cooperative data cache approach in MANET

Abstract: In a mobile ad hoc network (MANET) under normal cache sharing scenarios, when the data is transmitted from source to destination, all the nodes along the path store the information on the cache layer before reaching the destination. This may result in increased node overhead, increased cache memory utility and very high end-to-end delay. In this paper, we propose an enhanced cache sharing through cooperative data cache (ECSCDC) approach for MANETs. During the transmission of desired data from the data centre back to the request originator, the data packets will be cached by the intermediate caching nodes only if required, by using the asymmetric cooperative cache approach. Those caching nodes that can retain the data in its cache, for future data retrieval is selected based on scaled power community index. By simulation results, we show that the proposed technique reduces the communication overhead, access latency and average traffic ratio near the data centre while increasing the cache hit ratio.

Fluid structure interaction study on straight and undulated hollow fibre hemodialyser membranes

Abstract: In hemodialysis therapy, the dialyser is subjected to blood flow continuously for several hours and is also being reused; the stress experienced by the fibres owing to blood flow is of utmost importance because it reflects on the mechanical stability of the membrane. It is tedious to study the stress experienced by an individual fibre in real-time; computer aided techniques enables to gain better insights about the load bearing capacity of the membrane. A finite-element strategy is implemented to study the effect of flow induced stress in hemodialyser membrane. A 3D model of the membrane was developed in straight and undulated (crimped) fibre orientations. Fluid structure interaction study was conducted to analyse the stress distribution due to varying blood flow. It is observed that in both the fibre orientations, the stress varies inversely with the blood flow rate. The effect of varying the length of the fibre, wall thickness and crimp frequency is also studied. From the analysis it is found that the crimped fibres experiences less stress compared to straight fibre. Such analysis aids to predict and evaluate the performance of the hemodialyser membrane.

Recent advances on ankle foot orthosis for gait rehabilitation: a review

Abstract: Since the early 1980s, hydraulic and pneumatic device are used to explore methods of orthotic devices for lower limb. Over the past decades, significant development has been made by researchers in rehabilitation robotics associating assistive orthotic device for the lower limb extremities. The aim in writing this review article is to present a detailed insight towards the development of the controlled ankle foot orthosis (AFO) device for enhancing the functionality of people disabled by injury to the lower limb or by neuromuscular disorders such as multiple sclerosis, spinal muscular atrophy, etc. Different types of approaches towards design, actuation and control strategies of passive and active AFOs are analysed in this article considering gait rehabilitation. In currently available commercialised ankle foot orthotic devices for lower limb, to overcome the weakness and instability produced by drop foot and to follow natural gait is still a challenge. This paper also focuses the impact of active control of AFO device mainly to enhance the functionality of lower limb reducing the deformities. Researchers have put in huge amount of efforts in terms of modelling, simulating and controlling of such devices mainly for gait rehabilitation with kinematic and dynamics analysis.

Descendant adaptive filter to remove different noises from ECG signals

Abstract: Electrocardiogram (ECG) signals are electrical signals generated corresponding to activity of heart. ECG signals are recorded and analysed to monitor heart condition. In initial raw form, ECG signals are contaminated with different types of noises. These noises may be electrode motion artefact noise, baseline wander noise and muscle noise also known as electromyogram (EMG) noise etc. In this paper, a descendent structure consists of adaptive filters is used to eliminate the three different types of noises (i.e., motion artefact noise, baseline wander noise and muscle noise). The two different adaptive filtering algorithms have been implemented; least mean square (LMS) and recursive least square (RLS) algorithm. The performance of these filters are compared on the basis of different fidelity parameters such as mean square error (MSE), normalised root mean squared error (NRMSE), signal-to-noise ratio (SNR), percentage root mean squared difference (PRD), and maximum error (ME) has been observed.