The feasibility of a novel system to reliably estimate the normalized central blood pressure (CBPN) from the radial photoplethysmogram (PPG) is investigated. Right-wrist radial blood pressure and left-wrist PPG were simultaneously recorded in five different days. An industry-standard applanation tonometer was employed for recording radial blood pressure. The CBP waveform was amplitude-normalized to determine CBPN. A total of fifteen second-order autoregressive models with exogenous input were investigated using system identification techniques. Among these 15 models, the model producing the lowest coefficient of variation (CV) of the fitness during the five days was selected as the reference model. Results show that the proposed model is able to faithfully reproduce CBPN (mean fitness = 85.2% ± 2.5%) from the radial PPG for all 15 segments during the five recording days. The low CV value of 3.35% suggests a stable model valid for different recording days.
The present study was conducted to examine the effects of body weight on intradiscal pressure (IDP) and annulus stress of intervertebral discs at lumbar spine. Three-dimensional finite element model of osseoligamentous lumbar spine was developed subjected to follower load of 500 N, 800 N, and 1200 N which represent the loads for individuals who are normal and overweight with the pure moments at 7.5 Nm in flexion and extension motions. It was observed that the maximum IDP was 1.26 MPa at L1-L2 vertebral segment. However, the highest increment of IDP was found at L4-L5 segment where the IDP was increased to 30% in flexion and it was more severe at extension motion reaching to 80%. Furthermore, the maximum annulus stress also occurred at the L1-L2 segment with 3.9 MPa in extension motion. However, the highest increment was also found at L4-L5 where the annulus stress increased to 17% in extension motion. Based on these results, the increase of physiological loading could be an important factor to the increment of intradiscal pressure and annulus fibrosis stress at all intervertebral discs at the lumbar spine which may lead to early intervertebral disc damage.
In tongue diagnosis, colour information of tongue body has kept valuable information regarding the state of disease and its correlation with the internal organs. Qualitatively, practitioners may have difficulty in their judgement due to the instable lighting condition and naked eye's ability to capture the exact colour distribution on the tongue especially the tongue with multicolour substance. To overcome this ambiguity, this paper presents a two-stage tongue's multicolour classification based on a support vector machine (SVM) whose support vectors are reduced by our proposed k-means clustering identifiers and red colour range for precise tongue colour diagnosis. In the first stage, k-means clustering is used to cluster a tongue image into four clusters of image background (black), deep red region, red/light red region, and transitional region. In the second-stage classification, red/light red tongue images are further classified into red tongue or light red tongue based on the red colour range derived in our work. Overall, true rate classification accuracy of the proposed two-stage classification to diagnose red, light red, and deep red tongue colours is 94%. The number of support vectors in SVM is improved by 41.2%, and the execution time for one image is recorded as 48 seconds.
Retinal image registration is important to assist diagnosis and monitor retinal diseases, such as diabetic retinopathy and glaucoma. However, registering retinal images for various registration applications requires the detection and distribution of feature points on the low-quality region that consists of vessels of varying contrast and sizes. A recent feature detector known as Saddle detects feature points on vessels that are poorly distributed and densely positioned on strong contrast vessels. Therefore, we propose a multiresolution difference of Gaussian pyramid with Saddle detector (D-Saddle) to detect feature points on the low-quality region that consists of vessels with varying contrast and sizes. D-Saddle is tested on Fundus Image Registration (FIRE) Dataset that consists of 134 retinal image pairs. Experimental results show that D-Saddle successfully registered 43% of retinal image pairs with average registration accuracy of 2.329 pixels while a lower success rate is observed in other four state-of-the-art retinal image registration methods GDB-ICP (28%), Harris-PIIFD (4%), H-M (16%), and Saddle (16%). Furthermore, the registration accuracy of D-Saddle has the weakest correlation (Spearman) with the intensity uniformity metric among all methods. Finally, the paired t-test shows that D-Saddle significantly improved the overall registration accuracy of the original Saddle.
This paper presents a classification system to classify the cognitive load corresponding to targets and distractors present in opposite visual hemifields. The approach includes the study of EEG (electroencephalogram) signal features acquired in a spatial attention task. The process comprises of EEG feature selection based on the feature distribution, followed by the stepwise discriminant analysis- (SDA-) based channel selection. Repeated measure analysis of variance (rANOVA) is applied to test the statistical significance of the selected features. Classifiers are developed and compared using the selected features to classify the target and distractor present in visual hemifields. The results provide a maximum classification accuracy of 87.2% and 86.1% and an average classification accuracy of 76.5 ± 4% and 76.2 ± 5.3% over the thirteen subjects corresponding to the two task conditions. These correlates present a step towards building a feature-based neurofeedback system for visual attention.
Balance in the human body's movement is generally associated with different synergistic pathologies. The trunk is supported by one's leg most of the time when walking. A person with poor balance may face limitation when performing their physical activities on a daily basis, and they may be more prone to having risk of fall. The ground reaction forces (GRFs), centre of pressure (COP), and centre of mass (COM) in quite standing posture were often measured for the evaluation of balance. Currently, there is still no experimental evidence or study on leg length discrepancy (LLD) during walking. Analysis of the stability parameters is more representative of the functional activity undergone by the person who has a LLD. Therefore, this study hopes to shed new light on the effects of LLD on the dynamic stability associated with VGRF, COP, and COM during walking. Eighteen healthy subjects were selected among the university population with normal BMIs. Each subject was asked to walk with 1.0 to 2.0 ms-1 of walking speed for three to five trials each. Insoles of 0.5 cm thickness were added, and the thickness of the insoles was subsequently raised until 4 cm and placed under the right foot as we simulated LLD. The captured data obtained from a force plate and motion analysis present Peak VGRF (single-leg stance) and WD (double-leg stance) that showed more forces exerted on the short leg rather than long leg. Obviously, changes occurred on the displacement of COM trajectories in the ML and vertical directions as LLD increased at the whole gait cycle. Displacement of COP trajectories demonstrated that more distribution was on the short leg rather than on the long leg. The root mean square (RMS) of COP-COM distance showed, obviously, changes only in ML direction with the value at 3 cm and 3.5 cm. The cutoff value via receiver operating characteristic (ROC) indicates the significant differences starting at the level 2.5 cm up to 4 cm in long and short legs for both AP and ML directions. The present study performed included all the proposed parameters on the effect of dynamic stability on LLD during walking and thus helps to determine and evaluate the balance pattern.
Compression, in general, aims to reduce file size, with or without decreasing data quality of the original file. Digital Imaging and Communication in Medicine (DICOM) is a medical imaging file standard used to store multiple information such as patient data, imaging procedures, and the image itself. With the rising usage of medical imaging in clinical diagnosis, there is a need for a fast and secure method to share large number of medical images between healthcare practitioners, and compression has always been an option. This work analyses the Huffman coding compression method, one of the lossless compression techniques, as an alternative method to compress a DICOM file in open PACS settings. The idea of the Huffman coding compression method is to provide codeword with less number of bits for the symbol that has a higher value of byte frequency distribution. Experiments using different type of DICOM images are conducted, and the analysis on the performances in terms of compression ratio and compression/decompression time, as well as security, is provided. The experimental results showed that the Huffman coding technique has the capability to compress the DICOM file up to 1 : 3.7010 ratio and up to 72.98% space savings.
The medical system is facing the transformations with augmentation in the use of medical information systems, electronic records, smart, wearable devices, and handheld. The central nervous system function is to control the activities of the mind and the human body. Modern speedy development in medical and computational growth in the field of the central nervous system enables practitioners and researchers to extract and visualize insight from these systems. The function of augmented reality is to incorporate virtual and real objects, interactively running in a real-time and real environment. The role of augmented reality in the central nervous system becomes a thought-provoking task. Gesture interaction approach-based augmented reality in the central nervous system has enormous impending for reducing the care cost, quality refining of care, and waste and error reducing. To make this process smooth, it would be effective to present a comprehensive study report of the available state-of-the-art-work for enabling doctors and practitioners to easily use it in the decision making process. This comprehensive study will finally summarise the outputs of the published materials associate to gesture interaction-based augmented reality approach in the central nervous system. This research uses the protocol of systematic literature which systematically collects, analyses, and derives facts from the collected papers. The data collected range from the published materials for 10 years. 78 papers were selected and included papers based on the predefined inclusion, exclusion, and quality criteria. The study supports to identify the studies related to augmented reality in the nervous system, application of augmented reality in the nervous system, technique of augmented reality in the nervous system, and the gesture interaction approaches in the nervous system. The derivations from the studies show that there is certain amount of rise-up in yearly wise articles, and numerous studies exist, related to augmented reality and gestures interaction approaches to different systems of the human body, specifically to the nervous system. This research organises and summarises the existing associated work, which is in the form of published materials, and are related to augmented reality. This research will help the practitioners and researchers to sight most of the existing studies subjected to augmented reality-based gestures interaction approaches for the nervous system and then can eventually be followed as support in future for complex anatomy learning.
The 3D convolutional neural network is able to make use of the full nonlinear 3D context information of lung nodule detection from the DICOM (Digital Imaging and Communications in Medicine) images, and the Gradient Class Activation has shown to be useful for tailoring classification tasks and localization interpretation for fine-grained features and visual explanation for the internal working. Gradient-weighted class activation plays a crucial role for clinicians and radiologists in terms of trusting and adopting the model. Practitioners not only rely on a model that can provide high precision but also really want to gain the respect of radiologists. So, in this paper, we explored the lung nodule classification using the improvised 3D AlexNet with lightweight architecture. Our network employed the full nature of the multiview network strategy. We have conducted the binary classification (benign and malignant) on computed tomography (CT) images from the LUNA 16 database conglomerate and database image resource initiative. The results obtained are through the 10-fold cross-validation. Experimental results have shown that the proposed lightweight architecture achieved a superior classification accuracy of 97.17% on LUNA 16 dataset when compared with existing classification algorithms and low-dose CT scan images as well.
Emotion is a crucial aspect of human health, and emotion recognition systems serve important roles in the development of neurofeedback applications. Most of the emotion recognition methods proposed in previous research take predefined EEG features as input to the classification algorithms. This paper investigates the less studied method of using plain EEG signals as the classifier input, with the residual networks (ResNet) as the classifier of interest. ResNet having excelled in the automated hierarchical feature extraction in raw data domains with vast number of samples (e.g., image processing) is potentially promising in the future as the amount of publicly available EEG databases has been increasing. Architecture of the original ResNet designed for image processing is restructured for optimal performance on EEG signals. The arrangement of convolutional kernel dimension is demonstrated to largely affect the model's performance on EEG signal processing. The study is conducted on the Shanghai Jiao Tong University Emotion EEG Dataset (SEED), with our proposed ResNet18 architecture achieving 93.42% accuracy on the 3-class emotion classification, compared to the original ResNet18 at 87.06% accuracy. Our proposed ResNet18 architecture has also achieved a model parameter reduction of 52.22% from the original ResNet18. We have also compared the importance of different subsets of EEG channels from a total of 62 channels for emotion recognition. The channels placed near the anterior pole of the temporal lobes appeared to be most emotionally relevant. This agrees with the location of emotion-processing brain structures like the insular cortex and amygdala.
Medication Errors (MEs) are still significant challenges, especially in nonautomated health systems. Qualitative studies are mostly used to identify the parameters involved in MEs. Failing to provide accurate information in expert-based decisions can provoke unrealistic results and inappropriate corrective actions eventually. However, mostly, some levels of uncertainty accompany the decisions in real practice. This study tries to present a hybrid decision-making approach to assigning different weights to risk factors and considering the uncertainty in the ranking process in the Failure Modes and Effect Analysis (FMEA) technique. Initially, significant MEs are identified by three groups of qualified experts (doctors, nurses, and pharmacists). Afterward, for assigning weights to the risk factors, Z-number couples with the Stepwise Weight Assessment Ratio Analysis (SWARA) method, named Z-SWARA, to add reliability concept in the decision-making process. Finally, the identified MEs are ranked through the developed Weighted Aggregated Sum Product Assessment (WASPAS) method, namely, Z-WASPAS. To demonstrate the applicability of the proposed approach, the ranking results compare with typical methods, such as fuzzy-WASPAS and FMEA. The findings of the present study highlight improper medication administration as the main failure mode, which can result in a fatality or patient injury. Moreover, the utilization of multiple-criteria decision-making methods in combination with Z-number can be a useful tool in the healthcare management field since it can address the problems by considering reliability and uncertainty simultaneously.
Low-dose Computed Tomography (LDCT) has gained a great deal of attention in clinical procedures due to its ability to reduce the patient's risk of exposure to the X-ray radiation. However, reducing the X-ray dose increases the quantum noise and artifacts in the acquired LDCT images. As a result, it produces visually low-quality LDCT images that adversely affect the disease diagnosing and treatment planning in clinical procedures. Deep Learning (DL) has recently become the cutting-edge technology of LDCT denoising due to its high performance and data-driven execution compared to conventional denoising approaches. Although the DL-based models perform fairly well in LDCT noise reduction, some noise components are still retained in denoised LDCT images. One reason for this noise retention is the direct transmission of feature maps through the skip connections of contraction and extraction path-based DL modes. Therefore, in this study, we propose a Generative Adversarial Network with Inception network modules (InNetGAN) as a solution for filtering the noise transmission through skip connections and preserving the texture and fine structure of LDCT images. The proposed Generator is modeled based on the U-net architecture. The skip connections in the U-net architecture are modified with three different inception network modules to filter out the noise in the feature maps passing over them. The quantitative and qualitative experimental results have shown the performance of the InNetGAN model in reducing noise and preserving the subtle structures and texture details in LDCT images compared to the other state-of-the-art denoising algorithms.
In sports, fatigue management is vital as adequate rest builds strength and enhances performance, whereas inadequate rest exposes the body to prolonged fatigue (PF) or also known as overtraining. This paper presents PF identification and classification based on surface electromyography (EMG) signals. An experiment was performed on twenty participants to investigate the behaviour of surface EMG during the inception of PF. PF symptoms were induced in accord with a five-day Bruce Protocol treadmill test on four lower extremity muscles: the biceps femoris (BF), rectus femoris (RF), vastus medialis (VM), and vastus lateralis (VL). The results demonstrate that the experiment successfully induces soreness, unexplained lethargy, and performance decrement and also indicate that the progression of PF can be observed based on changes in frequency features (ΔF med and ΔF mean) and time features (ΔRMS and ΔMAV) of surface EMG. This study also demonstrates the ability of wavelet index features in PF identification. Using a naïve Bayes (NB) classifier exhibits the highest accuracy based on time and frequency features with 98% in distinguishing PF on RF, 94% on BF, 9% on VL, and 97% on VM. Thus, this study has positively indicated that surface EMG can be used in identifying the inception of PF. The implication of the findings is significant in sports to prevent a greater risk of PF.
The primary objective of this proposed framework work is to detect and classify various lung diseases such as pneumonia, tuberculosis, and lung cancer from standard X-ray images and Computerized Tomography (CT) scan images with the help of volume datasets. We implemented three deep learning models namely Sequential, Functional & Transfer models and trained them on open-source training datasets. To augment the patient's treatment, deep learning techniques are promising and successful domains that extend the machine learning domain where CNNs are trained to extract features and offers great potential from datasets of images in biomedical application. Our primary aim is to validate our models as a new direction to address the problem on the datasets and then to compare their performance with other existing models. Our models were able to reach higher levels of accuracy for possible solutions and provide effectiveness to humankind for faster detection of diseases and serve as best performing models. The conventional networks have poor performance for tilted, rotated, and other abnormal orientation and have poor learning framework. The results demonstrated that the proposed framework with a sequential model outperforms other existing methods in terms of an F1 score of 98.55%, accuracy of 98.43%, recall of 96.33% for pneumonia and for tuberculosis F1 score of 97.99%, accuracy of 99.4%, and recall of 98.88%. In addition, the functional model for cancer outperformed with an accuracy of 99.9% and specificity of 99.89% and paves way to less number of trained parameters, leading to less computational overhead and less expensive than existing pretrained models. In our work, we implemented a state-of-the art CNN with various models to classify lung diseases accurately.
Quality of care data has gained transparency captured through various measurements and reporting. Readmission measure is especially related to unfavorable patient outcomes that directly bends the curve of healthcare cost. Under the Hospital Readmission Reduction Program, payments to hospitals were reduced for those with excessive 30-day rehospitalization rates. These penalties have intensified efforts from hospital stakeholders to implement strategies to reduce readmission rates. One of the key strategies is the deployment of predictive analytics stratified by patient population. The recent research in readmission model is focused on making its prediction more accurate. As cost-saving improvements through artificial intelligent-based health solutions are expected, the broad economic impact of such digital tool remains unknown. Meanwhile, reducing readmission rate is associated with increased operating expenses due to targeted interventions. The increase in operating margin can surpass native readmission cost. In this paper, we propose a quantized evaluation metric to provide a methodological mean in assessing whether a predictive model represents cost-effective way of delivering healthcare. Herein, we evaluate the impact machine learning has had on transitional care and readmission with proposed metric. The final model was estimated to produce net healthcare savings at over $1 million given a 50% rate of successfully preventing a readmission.
In recent times, knee joint pains have become severe enough to make daily tasks difficult. Knee osteoarthritis is a type of arthritis and a leading cause of disability worldwide. The middle of the knee contains a vital portion, the anterior cruciate ligament (ACL). It is necessary to diagnose the ACL ruptured tears early to avoid surgery. The study aimed to perform a comparative analysis of machine learning models to identify the condition of three ACL tears. In contrast to previous studies, this study also considers imbalanced data distributions as machine learning techniques struggle to deal with this problem. The paper applied and analyzed four machine learning classification models, namely, random forest (RF), categorical boosting (Cat Boost), light gradient boosting machines (LGBM), and highly randomized classifier (ETC) on the balanced, structured dataset of ACL. After oversampling a hyperparameter adjustment, the above four models have achieved an average accuracy of 95.72%, 94.98%, 94.98%, and 98.26%. There are 2070 observations and eight features in the collection of three diagnosis ACL classes after oversampling. The area under curve value was approximately 0.998, respectively. Experiments were performed using twelve machine learning algorithms with imbalanced and balanced datasets. However, the accuracy of the imbalanced dataset has remained under 76% for all twelve models. After oversampling, the proposed model may contribute to the investigation of ACL tears on magnetic resonance imaging and other knee ligaments efficiently and automatically without involving radiologists.
Coronavirus disease 2019 (COVID-19) is a novel disease that affects healthcare on a global scale and cannot be ignored because of its high fatality rate. Computed tomography (CT) images are presently being employed to assist doctors in detecting COVID-19 in its early stages. In several scenarios, a combination of epidemiological criteria (contact during the incubation period), the existence of clinical symptoms, laboratory tests (nucleic acid amplification tests), and clinical imaging-based tests are used to diagnose COVID-19. This method can miss patients and cause more complications. Deep learning is one of the techniques that has been proven to be prominent and reliable in several diagnostic domains involving medical imaging. This study utilizes a convolutional neural network (CNN), stacked autoencoder, and deep neural network to develop a COVID-19 diagnostic system. In this system, classification undergoes some modification before applying the three CT image techniques to determine normal and COVID-19 cases. A large-scale and challenging CT image dataset was used in the training process of the employed deep learning model and reporting their final performance. Experimental outcomes show that the highest accuracy rate was achieved using the CNN model with an accuracy of 88.30%, a sensitivity of 87.65%, and a specificity of 87.97%. Furthermore, the proposed system has outperformed the current existing state-of-the-art models in detecting the COVID-19 virus using CT images.
Medical device reliability is the ability of medical devices to endure functioning and is indispensable to ensure service delivery to patients. Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) technique was employed in May 2021 to evaluate existing reporting guidelines on medical device reliability. The systematic searching is conducted in eight different databases, including Web of Science, Science Direct, Scopus, IEEE Explorer, Emerald, MEDLINE Complete, Dimensions, and Springer Link, with 36 articles shortlisted from the year 2010 to May 2021. This study aims to epitomize existing literature on medical device reliability, scrutinize existing literature outcomes, investigate parameters affecting medical device reliability, and determine the scientific research gaps. The result of the systematic review listed three main topics on medical device reliability: risk management, performance prediction using Artificial Intelligence or machine learning, and management system. The medical device reliability assessment challenges are inadequate maintenance cost data, determining significant input parameter selection, difficulties accessing healthcare facilities, and limited age in service. Medical device systems are interconnected and interoperating, which increases complexity in assessing their reliability. To the best of our knowledge, although machine learning has become popular in predicting medical device performance, the existing models are only applicable to selected devices such as infant incubators, syringe pumps, and defibrillators. Despite the importance of medical device reliability assessment, there is no explicit protocol and predictive model to anticipate the situation. The problem worsens with the unavailability of a comprehensive assessment strategy for critical medical devices. Therefore, this study reviews the current state of critical device reliability in healthcare facilities. The present knowledge can be improved by adding new scientific data emphasis on critical medical devices used in healthcare services.
The coronavirus (COVID-19) pandemic has had a terrible impact on human lives globally, with far-reaching consequences for the health and well-being of many people around the world. Statistically, 305.9 million people worldwide tested positive for COVID-19, and 5.48 million people died due to COVID-19 up to 10 January 2022. CT scans can be used as an alternative to time-consuming RT-PCR testing for COVID-19. This research work proposes a segmentation approach to identifying ground glass opacity or ROI in CT images developed by coronavirus, with a modified structure of the Unet model having been used to classify the region of interest at the pixel level. The problem with segmentation is that the GGO often appears indistinguishable from a healthy lung in the initial stages of COVID-19, and so, to cope with this, the increased set of weights in contracting and expanding the Unet path and an improved convolutional module is added in order to establish the connection between the encoder and decoder pipeline. This has a major capacity to segment the GGO in the case of COVID-19, with the proposed model being referred to as "convUnet." The experiment was performed on the Medseg1 dataset, and the addition of a set of weights at each layer of the model and modification in the connected module in Unet led to an improvement in overall segmentation results. The quantitative results obtained using accuracy, recall, precision, dice-coefficient, F1score, and IOU were 93.29%, 93.01%, 93.67%, 92.46%, 93.34%, 86.96%, respectively, which is better than that obtained using Unet and other state-of-the-art models. Therefore, this segmentation approach proved to be more accurate, fast, and reliable in helping doctors to diagnose COVID-19 quickly and efficiently.
Investigating gender differences based on emotional changes becomes essential to understand various human behaviors in our daily life. Ten students from the University of Vienna have been recruited by recording the electroencephalogram (EEG) dataset while watching four short emotional video clips (anger, happiness, sadness, and neutral) of audiovisual stimuli. In this study, conventional filter and wavelet (WT) denoising techniques were applied as a preprocessing stage and Hurst exponent (Hur) and amplitude-aware permutation entropy (AAPE) features were extracted from the EEG dataset. k-nearest neighbors (kNN) and support vector machine (SVM) classification techniques were considered for automatic gender recognition from emotional-based EEGs. The main novelty of this paper is twofold: first, to investigate Hur as a complexity feature and AAPE as an irregularity parameter for the emotional-based EEGs using two-way analysis of variance (ANOVA) and then integrating these features to propose a new CompEn hybrid feature fusion method towards developing the novel WT_CompEn gender recognition framework as a core for an automated gender recognition model to be sensitive for identifying gender roles in the brain-emotion relationship for females and males. The results illustrated the effectiveness of Hur and AAPE features as remarkable indices for investigating gender-based anger, sadness, happiness, and neutral emotional state. Moreover, the proposed WT_CompEn framework achieved significant enhancement in SVM classification accuracy of 100%, indicating that the novel WT_CompEn may offer a useful way for reliable enhancement of gender recognition of different emotional states. Therefore, the novel WT_CompEn framework is a crucial goal for improving the process of automatic gender recognition from emotional-based EEG signals allowing for more comprehensive insights to understand various gender differences and human behavior effects of an intervention on the brain.