Monitoring FAZ area enlargement enables physicians to monitor progression of the DR. At present, it is difficult to discern the FAZ area and to measure its enlargement in an objective manner using digital fundus images. A semi-automated approach for determination of FAZ using color images has been developed. Here, a binary map of retinal blood vessels is computer generated from the digital fundus image to determine vessel ends and pathologies surrounding FAZ for area analysis. The proposed method is found to achieve accuracies from 66.67% to 98.69% compared to accuracies of 18.13-95.07% obtained by manual segmentation of FAZ regions from digital fundus images.
The use of vascular intersection aberration as one of the signs when monitoring and diagnosing diabetic retinopathy from retina fundus images (FIs) has been widely reported in the literature. In this paper, a new hybrid approach called the combined cross-point number (CCN) method able to detect the vascular bifurcation and intersection points in FIs is proposed. The CCN method makes use of two vascular intersection detection techniques, namely the modified cross-point number (MCN) method and the simple cross-point number (SCN) method. Our proposed approach was tested on images obtained from two different and publicly available fundus image databases. The results show a very high precision, accuracy, sensitivity and low false rate in detecting both bifurcation and crossover points compared with both the MCN and the SCN methods.
Ultrasound is an important and low cost imaging modality used to study the internal organs of human body and blood flow through blood vessels. It uses high frequency sound waves to acquire images of internal organs. It is used to screen normal, benign and malignant tissues of various organs. Healthy and malignant tissues generate different echoes for ultrasound. Hence, it provides useful information about the potential tumor tissues that can be analyzed for diagnostic purposes before therapeutic procedures. Ultrasound images are affected with speckle noise due to an air gap between the transducer probe and the body. The challenge is to design and develop robust image preprocessing, segmentation and feature extraction algorithms to locate the tumor region and to extract subtle information from isolated tumor region for diagnosis. This information can be revealed using a scale space technique such as the Discrete Wavelet Transform (DWT). It decomposes an image into images at different scales using low pass and high pass filters. These filters help to identify the detail or sudden changes in intensity in the image. These changes are reflected in the wavelet coefficients. Various texture, statistical and image based features can be extracted from these coefficients. The extracted features are subjected to statistical analysis to identify the significant features to discriminate normal and malignant ultrasound images using supervised classifiers. This paper presents a review of wavelet techniques used for preprocessing, segmentation and feature extraction of breast, thyroid, ovarian and prostate cancer using ultrasound images.
Fatty liver disease (FLD) is reversible disease and can be treated, if it is identified at an early stage. However, if diagnosed at the later stage, it can progress to an advanced liver disease such as cirrhosis which may ultimately lead to death. Therefore, it is essential to detect it at an early stage before the disease progresses to an irreversible stage. Several non-invasive computer-aided techniques are proposed to assist in the early detection of FLD and cirrhosis using ultrasound images. In this work, we are proposing an algorithm to discriminate automatically the normal, FLD and cirrhosis ultrasound images using curvelet transform (CT) method. Higher order spectra (HOS) bispectrum, HOS phase, fuzzy, Kapoor, max, Renyi, Shannon, Vajda and Yager entropies are extracted from CT coefficients. These extracted features are subjected to locality sensitive discriminant analysis (LSDA) feature reduction method. Then these LSDA coefficients ranked based on F-value are fed to different classifiers to choose the best performing classifier using minimum number of features. Our proposed technique can characterize normal, FLD and cirrhosis using probabilistic neural network (PNN) classifier with an accuracy of 97.33%, specificity of 100.00% and sensitivity of 96.00% using only six features. In addition, these chosen features are used to develop a liver disease index (LDI) to differentiate the normal, FLD and cirrhosis classes using a single number. This can significantly help the radiologists to discriminate FLD and cirrhosis in their routine liver screening.
Shear wave elastography (SWE) examination using ultrasound elastography (USE) is a popular imaging procedure for obtaining elasticity information of breast lesions. Elasticity parameters obtained through SWE can be used as biomarkers that can distinguish malignant breast lesions from benign ones. Furthermore, the elasticity parameters extracted from SWE can speed up the diagnosis and possibly reduce human errors. In this paper, Shearlet transform and local binary pattern histograms (LBPH) are proposed as an original algorithm to differentiate malignant and benign breast lesions. First, Shearlet transform is applied on the SWE images to acquire low frequency, horizontal and vertical cone coefficients. Next, LBPH features are extracted from the Shearlet transform coefficients and subjected to dimensionality reduction using locality sensitivity discriminating analysis (LSDA). The reduced LSDA components are ranked and then fed to several classifiers for the automated classification of breast lesions. A probabilistic neural network classifier trained only with seven top ranked features performed best, and achieved 98.08% accuracy, 98.63% sensitivity, and 97.59% specificity in distinguishing malignant from benign breast lesions. The high sensitivity and specificity of our system indicates that it can be employed as a primary screening tool for faster diagnosis of breast malignancies, thereby possibly reducing the mortality rate due to breast cancer.
Age-related Macular Degeneration (AMD) and Diabetic Retinopathy (DR) are the most prevalent diseases responsible for visual impairment in the world. This work investigates discrimination potential in the texture of color fundus images to distinguish between diseased and healthy cases by avoiding the prior lesion segmentation step. It presents a retinal background characterization approach and explores the potential of Local Tetra Patterns (LTrP) for texture classification of AMD, DR and Normal images. Five different experiments distinguishing between DR - normal, AMD - normal, DR - AMD, pathological - normal and AMD - DR - normal cases were conducted and validated using the proposed approach, and promising results were obtained. For all five experiments, different classifiers namely, AdaBoost, c4.5, logistic regression, naive Bayes, neural network, random forest and support vector machine were tested. We experimented with three public datasets, ARIA, STARE and E-Optha. Further, the performance of LTrP is compared with other texture descriptors, such as local phase quantization, local binary pattern and local derivative pattern. In all cases, the proposed method obtained the area under the receiver operating characteristic curve and f-score values higher than 0.78 and 0.746 respectively. It was found that both performance measures achieve over 0.995 for DR and AMD detection using a random forest classifier. The obtained results suggest that the proposed technique can discriminate retinal disease using texture information and has potential to be an important component for an automated screening solution for retinal images.
Accurate detection of diabetic retinopathy (DR) mainly depends on identification of retinal landmarks such as optic disc and fovea. Present methods suffer from challenges like less accuracy and high computational complexity. To address this issue, this paper presents a novel approach for fast and accurate localization of optic disc (OD) and fovea using one-dimensional scanned intensity profile analysis. The proposed method utilizes both time and frequency domain information effectively for localization of OD. The final OD center is located using signal peak-valley detection in time domain and discontinuity detection in frequency domain analysis. However, with the help of detected OD location, the fovea center is located using signal valley analysis. Experiments were conducted on MESSIDOR dataset, where OD was successfully located in 1197 out of 1200 images (99.75%) and fovea in 1196 out of 1200 images (99.66%) with an average computation time of 0.52s. The large scale evaluation has been carried out extensively on nine publicly available databases. The proposed method is highly efficient in terms of quickly and accurately localizing OD and fovea structure together compared with the other state-of-the-art methods.
The range of effectiveness of the novel corona virus, known as COVID-19, has been continuously spread worldwide with the severity of associated disease and effective variation in the rate of contact. This paper investigates the COVID-19 virus dynamics among the human population with the prediction of the size of epidemic and spreading time. Corona virus disease was first diagnosed on January 30, 2020 in India. From January 30, 2020 to April 21, 2020, the number of patients was continuously increased. In this scientific work, our main objective is to estimate the effectiveness of various preventive tools adopted for COVID-19. The COVID-19 dynamics is formulated in which the parameters of interactions between people, contact tracing, and average latent time are included. Experimental data are collected from April 15, 2020 to April 21, 2020 in India to investigate this virus dynamics. The Genocchi collocation technique is applied to investigate the proposed fractional mathematical model numerically via Caputo-Fabrizio fractional derivative. The effect of presence of various COVID parameters e.g. quarantine time is also presented in the work. The accuracy and efficiency of the outputs of the present work are demonstrated through the pictorial presentation by comparing it to known statistical data. The real data for COVID-19 in India is compared with the numerical results obtained from the concerned COVID-19 model. From our results, to control the expansion of this virus, various prevention measures must be adapted such as self-quarantine, social distancing, and lockdown procedures.
In recent years, the global Internet of Medical Things (IoMT) industry has evolved at a tremendous speed. Security and privacy are key concerns on the IoMT, owing to the huge scale and deployment of IoMT networks. Machine learning (ML) and blockchain (BC) technologies have significantly enhanced the capabilities and facilities of healthcare 5.0, spawning a new area known as "Smart Healthcare." By identifying concerns early, a smart healthcare system can help avoid long-term damage. This will enhance the quality of life for patients while reducing their stress and healthcare costs. The IoMT enables a range of functionalities in the field of information technology, one of which is smart and interactive health care. However, combining medical data into a single storage location to train a powerful machine learning model raises concerns about privacy, ownership, and compliance with greater concentration. Federated learning (FL) overcomes the preceding difficulties by utilizing a centralized aggregate server to disseminate a global learning model. Simultaneously, the local participant keeps control of patient information, assuring data confidentiality and security. This article conducts a comprehensive analysis of the findings on blockchain technology entangled with federated learning in healthcare. 5.0. The purpose of this study is to construct a secure health monitoring system in healthcare 5.0 by utilizing a blockchain technology and Intrusion Detection System (IDS) to detect any malicious activity in a healthcare network and enables physicians to monitor patients through medical sensors and take necessary measures periodically by predicting diseases. The proposed system demonstrates that the approach is optimized effectively for healthcare monitoring. In contrast, the proposed healthcare 5.0 system entangled with FL Approach achieves 93.22% accuracy for disease prediction, and the proposed RTS-DELM-based secure healthcare 5.0 system achieves 96.18% accuracy for the estimation of intrusion detection.
During the coronavirus disease (COVID-19) pandemic, different technologies, including telehealth, are maximised to mitigate the risks and consequences of the disease. Telehealth has been widely utilised because of its usability and safety in providing healthcare services during the COVID-19 pandemic. However, a systematic literature review which provides extensive evidence on the impact of COVID-19 through telehealth and which covers multiple directions in a large-scale research remains lacking. This study aims to review telehealth literature comprehensively since the pandemic started. It also aims to map the research landscape into a coherent taxonomy and characterise this emerging field in terms of motivations, open challenges and recommendations. Articles related to telehealth during the COVID-19 pandemic were systematically searched in the WOS, IEEE, Science Direct, Springer and Scopus databases. The final set included (n = 86) articles discussing telehealth applications with respect to (i) control (n = 25), (ii) technology (n = 14) and (iii) medical procedure (n = 47). Since the beginning of the pandemic, telehealth has been presented in diverse cases. However, it still warrants further attention. Regardless of category, the articles focused on the challenges which hinder the maximisation of telehealth in such times and how to address them. With the rapid increase in the utilization of telehealth in different specialised hospitals and clinics, a potential framework which reflects the authors' implications of the future application and opportunities of telehealth has been established. This article improves our understanding and reveals the full potential of telehealth during these difficult times and beyond.
Automated evaluation of a subject's neurocognitive performance (NCP) is a relevant topic in neurological and clinical studies. NCP represents the mental/cognitive human capacity in performing a specific task. It is difficult to develop the study protocols as the subject's NCP changes in a known predictable way. Sleep is time-varying NCP and can be used to develop novel NCP techniques. Accurate analysis and interpretation of human sleep electroencephalographic (EEG) signals is needed for proper NCP assessment. In addition, sleep deprivation may cause prominent cognitive risks in performing many common activities such as driving or controlling a generic device; therefore, sleep scoring is a crucial part of the process. In the sleep cycle, the first stage of non-rapid eye movement (NREM) sleep or stage N1 is the transition between wakefulness and drowsiness and becomes relevant for the study of NCP. In this study, a novel cascaded recurrent neural network (RNN) architecture based on long short-term memory (LSTM) blocks, is proposed for the automated scoring of sleep stages using EEG signals derived from a single-channel. Fifty-five time and frequency-domain features were extracted from the EEG signals and fed to feature reduction algorithms to select the most relevant ones. The selected features constituted as the inputs to the LSTM networks. The cascaded architecture is composed of two LSTM RNNs: the first network performed 4-class classification (i.e. the five sleep stages with the merging of stages N1 and REM into a single stage) with a classification rate of 90.8%, and the second one obtained a recognition performance of 83.6% for 2-class classification (i.e. N1 vs REM). The overall percentage of correct classification for five sleep stages is found to be 86.7%. The objective of this work is to improve classification performance in sleep stage N1, as a first step of NCP assessment, and at the same time obtain satisfactory classification results in the other sleep stages.
B-mode ultrasound imaging is used extensively in medicine. Hence, there is a need to have efficient segmentation tools to aid in computer-aided diagnosis, image-guided interventions, and therapy. This paper presents a comprehensive review on automated localization and segmentation techniques for B-mode ultrasound images. The paper first describes the general characteristics of B-mode ultrasound images. Then insight on the localization and segmentation of tissues is provided, both in the case in which the organ/tissue localization provides the final segmentation and in the case in which a two-step segmentation process is needed, due to the desired boundaries being too fine to locate from within the entire ultrasound frame. Subsequenly, examples of some main techniques found in literature are shown, including but not limited to shape priors, superpixel and classification, local pixel statistics, active contours, edge-tracking, dynamic programming, and data mining. Ten selected applications (abdomen/kidney, breast, cardiology, thyroid, liver, vascular, musculoskeletal, obstetrics, gynecology, prostate) are then investigated in depth, and the performances of a few specific applications are compared. In conclusion, future perspectives for B-mode based segmentation, such as the integration of RF information, the employment of higher frequency probes when possible, the focus on completely automatic algorithms, and the increase in available data are discussed.
Due to the continued evolution of the SARS-CoV-2 pandemic, researchers worldwide are working to mitigate, suppress its spread, and better understand it by deploying digital signal processing (DSP) and machine learning approaches. This study presents an alignment-free approach to classify the SARS-CoV-2 using complementary DNA, which is DNA synthesized from the single-stranded RNA virus. Herein, a total of 1582 samples, with different lengths of genome sequences from different regions, were collected from various data sources and divided into a SARS-CoV-2 and a non-SARS-CoV-2 group. We extracted eight biomarkers based on three-base periodicity, using DSP techniques, and ranked those based on a filter-based feature selection. The ranked biomarkers were fed into k-nearest neighbor, support vector machines, decision trees, and random forest classifiers for the classification of SARS-CoV-2 from other coronaviruses. The training dataset was used to test the performance of the classifiers based on accuracy and F-measure via 10-fold cross-validation. Kappa-scores were estimated to check the influence of unbalanced data. Further, 10 × 10 cross-validation paired t-test was utilized to test the best model with unseen data. Random forest was elected as the best model, differentiating the SARS-CoV-2 coronavirus from other coronaviruses and a control a group with an accuracy of 97.4 %, sensitivity of 96.2 %, and specificity of 98.2 %, when tested with unseen samples. Moreover, the proposed algorithm was computationally efficient, taking only 0.31 s to compute the genome biomarkers, outperforming previous studies.
In this paper, an algorithm based on independent component analysis (ICA) for extracting the fetal heart rate (FHR) from maternal abdominal electrodes is presented. Three abdominal ECG channels are used to extract the FHR in three steps: first preprocessing procedures such as DC cancellation and low-pass filtering are applied to remove noise. Then the algorithm for multiple unknown source extraction (AMUSE) algorithm is fed to extract the sources from the observation signals include fetal ECG (FECG). Finally, FHR is extracted from FECG. The method is shown to be capable of completely revealing FECG R-peaks from observation leads even with a SNR=-200dB using semi-synthetic data.
Effects of different boundary conditions prescribed across the boundaries of radiofrequency ablation (RFA) models of liver cancer are investigated for the case where the tumour is at the liver boundary. Ground and Robin-type conditions (electrical field) and body temperature and thermal insulation (thermal field) conditions are examined. 3D models of the human liver based on publicly-available CT images of the liver are developed. An artificial tumour is placed inside the liver at the boundary. Simulations are carried out using the finite element method. The numerical results indicated that different electrical and thermal boundary conditions led to different predictions of the electrical potential, temperature and thermal coagulation distributions. Ground and body temperature conditions presented an unnatural physical conditions around the ablation site, which results in more intense Joule heating and excessive heat loss from the tissue. This led to thermal damage volumes that are smaller than the cases when the Robin type or the thermal insulation conditions are prescribed. The present study suggests that RFA simulations in the future must take into consideration the choice of the type of electrical and thermal boundary conditions to be prescribed in the case where the tumour is located near to the liver boundary.
Fast diagnostic methods can control and prevent the spread of pandemic diseases like coronavirus disease 2019 (COVID-19) and assist physicians to better manage patients in high workload conditions. Although a laboratory test is the current routine diagnostic tool, it is time-consuming, imposing a high cost and requiring a well-equipped laboratory for analysis. Computed tomography (CT) has thus far become a fast method to diagnose patients with COVID-19. However, the performance of radiologists in diagnosis of COVID-19 was moderate. Accordingly, additional investigations are needed to improve the performance in diagnosing COVID-19. In this study is suggested a rapid and valid method for COVID-19 diagnosis using an artificial intelligence technique based. 1020 CT slices from 108 patients with laboratory proven COVID-19 (the COVID-19 group) and 86 patients with other atypical and viral pneumonia diseases (the non-COVID-19 group) were included. Ten well-known convolutional neural networks were used to distinguish infection of COVID-19 from non-COVID-19 groups: AlexNet, VGG-16, VGG-19, SqueezeNet, GoogleNet, MobileNet-V2, ResNet-18, ResNet-50, ResNet-101, and Xception. Among all networks, the best performance was achieved by ResNet-101 and Xception. ResNet-101 could distinguish COVID-19 from non-COVID-19 cases with an AUC of 0.994 (sensitivity, 100%; specificity, 99.02%; accuracy, 99.51%). Xception achieved an AUC of 0.994 (sensitivity, 98.04%; specificity, 100%; accuracy, 99.02%). However, the performance of the radiologist was moderate with an AUC of 0.873 (sensitivity, 89.21%; specificity, 83.33%; accuracy, 86.27%). ResNet-101 can be considered as a high sensitivity model to characterize and diagnose COVID-19 infections, and can be used as an adjuvant tool in radiology departments.
Flow of an electrically conducting fluid characterizing blood through the arteries having irregular shaped multi-stenoses in the environment of a uniform transverse magnetic-field is analysed. The flow is considered to be axisymmetric with an outline of the irregular stenoses obtained from a three-dimensional casting of a mild stenosed artery, so that the physical problem becomes more realistic from the physiological point of view. The marker and cell (MAC) and successive-over-relaxation (SOR) methods are respectively used to solve the governing unsteady magnetohydrodynamic (MHD) equations and pressure-Poisson equation quantitatively and to observe the flow separation. The results obtained show that the flow separates mostly towards the downstream of the multi-stenoses. However, the flow separation region keeps on shrinking with the increasing intensity of the magnetic-field which completely disappears with sufficiently large value of the Hartmann number. The present observations certainly have some clinical implications relating to magnetotherapy which help reducing the complex flow separation zones causing flow disorder leading to the formation and progression of the arterial diseases.
Diabetes foot ulceration (DFU) and amputation are a cause of significant morbidity. The prevention of DFU may be achieved by the identification of patients at risk of DFU and the institution of preventative measures through education and offloading. Several studies have reported that thermogram images may help to detect an increase in plantar temperature prior to DFU. However, the distribution of plantar temperature may be heterogeneous, making it difficult to quantify and utilize to predict outcomes. We have compared a machine learning-based scoring technique with feature selection and optimization techniques and learning classifiers to several state-of-the-art Convolutional Neural Networks (CNNs) on foot thermogram images and propose a robust solution to identify the diabetic foot. A comparatively shallow CNN model, MobilenetV2 achieved an F1 score of ∼95% for a two-feet thermogram image-based classification and the AdaBoost Classifier used 10 features and achieved an F1 score of 97%. A comparison of the inference time for the best-performing networks confirmed that the proposed algorithm can be deployed as a smartphone application to allow the user to monitor the progression of the DFU in a home setting.
This paper uses wavelets in the detection comparison of breast cancer among the three main races in Malaysia: Chinese, Malays, and Indians followed by a system that evaluates the radiologist's findings over a period of time to gauge the radiologist's skills in confirming breast cancer cases. The db4 wavelet has been utilized to detect microcalcifications in mammogram-digitized images obtained from Malaysian women sample. The wavelet filter's detection evaluation was done by visual inspection by an expert radiologist to confirm the detection results of those pixels that corresponded to microcalcifications. Detection was counted if the wavelet-detected pixels corresponded to the radiologist's identified microcalcification pixels. After the radiologist's detection confirmation a new client-server radiologist recording and evaluation system is designed to evaluate the findings of the radiologist over some period of cancer detection working time. It is a system that records the findings of the Malaysian radiologist for the presence of breast cancer in Malaysian patients and provides a way of registering the progress of detecting breast cancer of the radiologist by tracking certain metric values such as the sensitivity, specificity, and receiver operator curve (ROC). The initial findings suggest that no single race mammograms are easier for wavelets' detections of microcalcifications and for the radiologist confirmation even though for this study the Chinese race samples detection average were a few percentages less than the other two races, namely the Malay and Indian races.
Food recognition systems recently garnered much research attention in the relevant field due to their ability to obtain objective measurements for dietary intake. This feature contributes to the management of various chronic conditions. Challenges such as inter and intraclass variations alongside the practical applications of smart glasses, wearable cameras, and mobile devices require resource-efficient food recognition models with high classification performance. Furthermore, explainable AI is also crucial in health-related domains as it characterizes model performance, enhancing its transparency and objectivity. Our proposed architecture attempts to address these challenges by drawing on the strengths of the transfer learning technique upon initializing MobiletNetV3 with weights from a pre-trained model of ImageNet. The MobileNetV3 achieves superior performance using the squeeze and excitation strategy, providing unequal weight to different input channels and contrasting equal weights in other variants. Despite being fast and efficient, there is a high possibility for it to be stuck in the local optima like other deep neural networks, reducing the desired classification performance of the model. Thus, we overcome this issue by applying the snapshot ensemble approach as it enables the M model in a single training process without any increase in the required training time. As a result, each snapshot in the ensemble visits different local minima before converging to the final solution which enhances recognition performance. On overcoming the challenge of explainability, we argue that explanations cannot be monolithic, since each stakeholder perceive the results', explanations based on different objectives and aims. Thus, we proposed a user-centered explainable artificial intelligence (AI) framework to increase the trust of the involved parties by inferencing and rationalizing the results according to needs and user profile. Our framework is comprehensive in terms of a dietary assessment app as it detects Food/Non-Food, food categories, and ingredients. Experimental results on the standard food benchmarks and newly contributed Malaysian food dataset for ingredient detection demonstrated superior performance on an integrated set of measures over other methodologies.