Biomedical Engineering: Applications, Basis and Communications

Electrocardiogram (ECG) is one of the best representatives of physiological signal that provides the state of the autonomic nervous system, primarily responsible for the cardiac activity. The ECG data compression plays a significant role in localized digital storage or efficient communication channel utilization in telemedicine applications. The lossless and lossy compression system’s compressor efficiency depends on the methodologies used for compression and the quality measure used to evaluate distortion. Based on domain ECG, data compression can be performed either one-dimensional (1D) or two-dimensional (2D) for utilization of inter and inter with intra beat correlation, respectively. In this paper, a comparative study between 1D and 2D ECG data compression methods was taken out from the existing literature to provide an update in this regard. ECG data compression techniques and algorithms in 1D and 2D domain have their own merits and limitations. Recently, numerous research and techniques in 1D ECG data compression have been developed, including direct and transformed domain. Additionally, 2D ECG data compression research is reported based on period normalization and complexity sorting in recent times. Finally, several practical issues highlight the assessment of reconstructed signal quality and performance comparisons with an average comparative of exhaustive existing 1D and 2D ECG compression methods based on the utilized digital signal processing systems.

Autism is one of the developmental disorders which make major behavioral and communication changes in the individual and induce a major drawback in the walking pattern. The main aim of this study is to analyze and compare the stride patterns of an autistic patient and a healthy individual. With the help of the proposed device, the yaw, pitch and roll of the limbs can be easily measured and analyzed. These parameters will identify the lacking movement patterns and help the physiotherapists in improving the walking pattern of the autistic patients by adopting the correct approach of remedial actions/treatment. It can also be used by the people who have undergone hip replacements or knee replacements and have an abnormal gait pattern. This device is a simple accumulation of sensors and memory storage unit. Two gyroscope sensors have been used, one each for the knee and the ankle. These sensors are strapped onto the knee and ankle with the help of free-sized elastic bands. The data received by the sensors are stored in the storage device automatically and are evaluated in a specially designed Web application. However, the raw data from the gyroscope sensors have not been used directly and various filters have been used to remove the errors. The results have been compared between an autistic patient and a normal healthy subject and showed a higher degree of angle difference for the autistic patient than the normal individual.

Electromyography (EMG) signal recording equipment is comparatively modern. Still, there are enough restrictions in detection, recording, and characterization of EMG signals because of nonlinearity in the equipment, which leads to noise components. The most commonly affecting artifacts are Power Line Interference (PLI-Noise), Baseline Wander noise (BW-Noise), and Electrocardiogram noise (ECG-Noise). Adaptive filters are advanced and effective solutions for EMG signal denoising, but the improper tuning of filter coefficients leads to noise components in the denoised EMG signal. This defect in adaptive filters triggers or motivates us to optimize the filter coefficients with existing meta-heuristics optimization algorithms. In this paper, Least Mean Squares (LMS) filter and Recursive Least Squares (RLS) adaptive filter coefficients are optimized with a new Hybrid Firefly–Particle Swarm Optimization (HFPSO) by taking the advantages and disadvantages of both the algorithms. Experiments are conducted with the proposed HFPSO and it proved better in EMG signal denoising in terms of the measured parameters like signal-to-noise ratio (SNR) in dB, maximum error (ME), mean square error (MSE), etc. In the second part of the work, the denoised EMG signal features are extracted for the diagnosis of diseases related to myopathy and neuropathy as EMG signal reflects the neuromuscular function and EMG signal examination may contribute to the diagnosis of muscle disorder linked to myopathy and neuropathy.

The novel coronavirus (COVID-19) was first reported in the Wuhan City of China in 2019 and became a pandemic. The outbreak has caused shocking effects to the people across the globe. It is important to screen a majority of the population in every country and for the respective governments to take appropriate action. There is a need for a rapid screening system to triage and recommend the patients for appropriate treatment. Chest X-ray imaging is one of the potential modalities, which has ample advantages such as wide availability even in the villages, portability, fast data sharing option from the point of capturing to the point of investigation, etc. The aim of the proposed work is to develop a deep learning algorithm for screening COVID-19 cases by leveraging the widely available X-ray imaging. We have built a deep learning Convolutional Neural Network model utilizing a combination of the public domain (open-source COVID-19) and private data (pneumonia and normal cases). The dataset was used before and after the segmentation of the lung region for training and testing. The outcome of the classification after lung segmentation resulted in significant superiority. The average accuracy achieved by the proposed system was 96%. The heat maps incorporated in the system were helpful for our radiologists to cross-verify whether the appropriate features are identified. This system (COVID-Detect) can be used in remote places in the countries affected by COVID-19 for mass screening of suspected cases and suggesting appropriate actions, such as recommending confirmatory tests.

In this paper, we propose a novel method to detect Congenital Heart Diseases (CHDs) using digital palm images. An Axial triradius is one of the features of palm whose position can be used to detect CHDs. In palm print image, axial triradius is identified using a pattern matching algorithm which is a well-known algorithm in image processing. Along with axial triradius, two more triradii are identified. One triradius is located near the little finger and another one located near the index finger. The location of all these three triradii is obtained. Two vectors are drawn from axial triradius, one vector towards left triradius and another vector towards right triradius. The angle at axial triradius is calculated. The angle obtained is used to detect CHDs. Here template matching method has been proposed to identify triradii on palmprint images. Using this approach CHDs such as Fallot’s Tetralogy (FT), Atrial Septal Defect (ASD), Ventricular Septal Defect (VSD) and Coarctation of the Aorta (COA) can be detected. This paper deals with the angle at axial triradius which is calculated for two different people to determine the disease. Out of these two, one person is clinically diagnosed as suffering from FT and another person is healthy. The proposed method can predict that the person is suffering from TF. The proposed method has been used to detect the angle at axial triradius by taking palm images of 100 FT patients, 100 ASD patients, 100 VSD patients, 100 COA patients and 100 healthy people. The result obtained has proved the correctness of the results of previous research works to a certain extent. The present approach allows the automation towards the identification of triradii and evaluation of the position of axial triradius along with the results of its application on some sample images.

The image segmentation of histopathological tissue images has always been a challenge due to the overlapping of tissue color distributions, the complexity of extracellular texture and the large image size. In this paper, we introduce a new region-merging algorithm, namely, the Regional Pattern Merging (RPM) for interactive color image segmentation and annotation, by efficiently retrieving and applying the user’s prior knowledge of stroke-based interaction. Low-level color/texture features of each region are used to compose a regional pattern adapted to differentiating a foreground object from the background scene. This iterative region-merging is based on a modified Region Adjacency Graph (RAG) model built from initial segmented results of the mean shift to speed up the merging process. The foreground region of interest (ROI) is segmented by the reduction of the background region and discrimination of uncertain regions. We then compare our method against state-of-the-art interactive image segmentation algorithms in both natural images and histological images. Taking into account the homogeneity of both color and texture, the resulting semi-supervised classification and interactive segmentation capture histological structures more completely than other intensity or color-based methods. Experimental results show that the merging of the RAG model runs in a linear time according to the number of graph edges, which is essentially faster than both traditional graph-based and region-based methods.

The internet of medical things (IoMT) has a great role in improving the health around the world. IoMT is having a great impact in our life in which the clinical data of the patient is observed and checked and then can be transferred to the third party for using in the future such as the cloud. IoMT is a huge data system with a continuous developing rate, which implies that we should keep a lot of data secure. We propose a combined security architecture that fuses the standard architecture and new blockchain technology. Blockchain is a temper digital ledger which gives peer-to-peer communication and provides communication between non-trust individuals. Using standard in-depth strategy and blockchain, we are able to develop a method to collect vital signs data from IoMT and connected devices and use blockchain to store and retrieve the collected data in a secure and decentralized fashion within a closed system, suitable for healthcare providers such as private clinics, hospitals, and healthcare organizations were sharing data with each other is required. Right now initially examine the innovation behind Blockchain then propose IoMT-based security architecture utilizing Blockchain to guarantee the security of information transmission between associated nodes. Experimental analysis shows that the proposed scheme presents a non-significant overhead; yet it brings major advantages to meet the standard security and privacy requirements in IoMT.

Generally, radiation oncology applies evaluation and prediction in medical imaging and diagnosis, specifically for contouring organs, which results in the production of the clinical target volume (CTV) that corresponds to disease risk and organ exclusion. Medical physicists contour organs and combine computed tomography (CT) scans to digital imaging and communications in medicine (DICOM) radiation therapy (RT) to assist physicians for diagnosing tumors and calculating the dosages in treatments including radiation and chemotherapy. Thus, to generate RT images with high accuracy, this paper proposes a new Generator Adversarial Network (GAN) for RT images called radiation therapy GAN (RTGAN). We combine multiple loss functions with synthetic similarity DICOM-RT images and compare the results with Pinnacle, a radiation oncology treatment planning system. Further, we evaluate the method to get a score of 0.984 in structured similarity (SSIM) and 31.26 in peak signal-to-noise ratio (PSNR) and find that it costs 0.058 s to finish contouring one CT image. The proposed method is applied and tested in the department of radiation oncology at the Chung Shan Medical University Hospital, and the results are similar to the ground truth images. Thus, it not only effectively reduces the false-positive rate but also makes a breakthrough in medicine.