This study proposed a registration framework to fuse 2D echocardiography images of the aortic valve with preoperative cardiac CT volume. The registration facilitates the fusion of CT and echocardiography to aid the diagnosis of aortic valve diseases and provide surgical guidance during transcatheter aortic valve replacement and implantation. The image registration framework consists of two major steps: temporal synchronization and spatial registration. Temporal synchronization allows time stamping of echocardiography time series data to identify frames that are at similar cardiac phase as the CT volume. Spatial registration is an intensity-based normalized mutual information method applied with pattern search optimization algorithm to produce an interpolated cardiac CT image that matches the echocardiography image. Our proposed registration method has been applied on the short-axis “Mercedes Benz” sign view of the aortic valve and long-axis parasternal view of echocardiography images from ten patients. The accuracy of our fully automated registration method was 0.81 ± 0.08 and 1.30 ± 0.13 mm in terms of Dice coefficient and Hausdorff distance for short-axis aortic valve view registration, whereas for long-axis parasternal view registration it was 0.79 ± 0.02 and 1.19 ± 0.11 mm, respectively. This accuracy is comparable to gold standard manual registration by expert. There was no significant difference in aortic annulus diameter measurement between the automatically and manually registered CT images. Without the use of optical tracking, we have shown the applicability of this technique for effective fusion of echocardiography with preoperative CT volume to potentially facilitate catheter-based surgery.

Background. The environment has been significantly impacted by rapid urbaniza- tion, leading to a need for changes in climate change and pollution indicators. The 4IR offers a potential solution to efficiently manage these impacts. Smart city ecosys- tems can provide well-designed, sustainable, and safe cities that enable holistic climate change and global warming solutions through various community-centred initiatives. These include smart planning techniques, smart environment monitoring, and smart governance. An air quality intelligence platform, which operates as a complete mea- surement site for monitoring and governing air quality, has shown promising results in providing actionable insights. This article aims to highlight the potential of ma- chine learning models in predicting air quality, providing data-driven strategic and sustainable solutions for smart cities. Methods. This study proposed an end-to-end air quality predictive model for smart city applications, utilizing four machine learning techniques and two deep learning techniques. These include Ada Boost, SVR, RF, KNN, MLP regressor and LSTM. The study was conducted in four different urban cities in Selangor, Malaysia, including Petaling Jaya, Banting, Klang, and Shah Alam. The model considered the air qual- ity data of various pollution markers such as PM2.5, PM10, O3, and CO. Additionally, meteorological data including wind speed and wind direction were also considered, and their interactions with the pollutant markers were quantified. The study aimed to determine the correlation variance of the dependent variable in predicting air pol- lution and proposed a feature optimization process to reduce dimensionality and re- move irrelevant features to enhance the prediction of PM2.5, improving the existing LSTM model. The study estimates the concentration of pollutants in the air based on training and highlights the contribution of feature optimization in air quality predic- tions through feature dimension reductions. Results. In this section, the results of predicting the concentration of pollutants (PM2.5, PM10, O3, and CO) in the air are presented in R2 and RMSE. In predicting the PM10 and PM2.5 concentration, LSTM performed the best overall high R2 values in the four study areas with the R2 values of 0.998, 0.995, 0.918, and 0.993 in Banting, How to cite this article Neo EX, Hasikin K, Lai KW, Mokhtar MI, Azizan MM, Hizaddin HF, Razak SA, Yanto . 2023. Artificial intelligence-assisted air quality monitoring for smart city management. PeerJ Comput. Sci. 9:e1306 http://doi.org/10.7717/peerj-cs.1306 Petaling, Klang and Shah Alam stations, respectively. The study indicated that among the studied pollution markers, PM2.5, PM10, NO2, wind speed and humidity are the most important elements to monitor. By reducing the number of features used in the model the proposed feature optimization process can make the model more interpretable and provide insights into the most critical factor affecting air quality. Findings from this study can aid policymakers in understanding the underlying causes of air pollution and develop more effective smart strategies for reducing pollution levels.

A process that involves the registration of two brain Magnetic Resonance Imaging (MRI) acquisitions is proposed for the subtraction between previous and current images at two different follow-up (FU) time points. Brain tumours can be non-cancerous (benign) or cancerous (malignant). Treatment choices for these conditions rely on the type of brain tumour as well as its size and location. Brain cancer is a fast-spreading tumour that must be treated in time. MRI is commonly used in the detection of early signs of abnormality in the brain area because it provides clear details. Abnormalities include the presence of cysts, haematomas or tumour cells. A sequence of images can be used to detect the progression of such abnormalities. A previous study on conventional (CONV) visual reading reported low accuracy and speed in the early detection of abnormalities, specifically in brain images. It can affect the proper diagnosis and treatment of the patient. A digital subtraction technique that involves two images acquired at two interval time points and their subtraction for the detection of the progression of abnormalities in the brain image was proposed in this study. MRI datasets of five patients, including a series of brain images, were retrieved retrospectively in this study. All methods were carried out using the MATLAB programming platform. ROI volume and diameter for both regions were recorded to analyse progression details, location, shape variations and size alteration of tumours. This study promotes the use of digital subtraction techniques on brain MRIs to track any abnormality and achieve early diagnosis and accuracy whilst reducing reading time. Thus, improving the diagnostic information for physicians can enhance the treatment plan for patients.

Abstract: Background: Early diagnosis of liver cancer may increase life expectancy. Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) play a vital role in diagnosing liver cancer. Together, both modalities offer significant individual and specific diagnosis data to physicians; however, they lack the integration of both types of information. To address this concern, a registration process has to be utilized for the purpose, as multimodal details are crucial in providing the physician with complete information. Objective: The aim was to present a model of CT-MRI registration used to diagnose liver cancer, specifically for improving the quality of the liver images and provide all the required information for earlier detection of the tumors. This method should concurrently address the issues of imaging procedures for liver cancer to fasten the detection of the tumor from both modalities. Methods: In this work, a registration scheme for fusing the CT and MRI liver images is studied. A feature point-based method with normalized cross-correlation has been utilized to aid in the diagnosis of liver cancer and provide multimodal information to physicians. Data on ten patients from an online database were obtained. For each dataset, three planar views from both modalities were interpolated and registered using feature point-based methods. The registration of algorithms was carried out by MATLAB (vR2019b, Mathworks, Natick, USA) on an Intel (R) Core (TM) i5-5200U CPU @ 2.20 GHz computer. The accuracy of the registered image is being validated qualitatively and quantitatively. Results: The results show that an accurate registration is obtained with minimal distance errors by which CT and MRI were accurately registered based on the validation of the experts. The RMSE ranges from 0.02 to 1.01 for translation, which is equivalent in magnitude to approximately 0 to 5 pixels for CT and registered image resolution. Conclusion: The CT-MRI registration scheme can provide complementary information on liver cancer to physicians, thus improving the diagnosis and treatment planning process.

The segmentation of the left ventricle (LV) is one of the fundamental procedures that must be performed to obtain quantitative measures of the heart, such as its volume, area, and ejection fraction. In clinical practice, the delineation of LV is still often conducted semi-automatically, leaving it open to operator subjectivity. The automatic LV segmentation from echocardiography images is a challenging task due to poorly defined boundaries and operator dependency. Recent research has demonstrated that deep learning has the capability to employ the segmentation process automatically. However, the well-known state-of-the-art segmentation models still lack in terms of accuracy and speed. This study aims to develop a single-stage lightweight segmentation model that precisely and rapidly segments the LV from 2D echocardiography images. In this research, a backbone network is used to acquire both low-level and high-level features. Two parallel blocks, known as the spatial feature unit and the channel feature unit, are employed for the enhancement and improvement of these features. The refined features are merged by an integrated unit to segment the LV. The performance of the model and the time taken to segment the LV are compared to other established segmentation models, DeepLab, FCN, and Mask RCNN. The model achieved the highest values of the dice similarity index (0.9446), intersection over union (0.8445), and accuracy (0.9742). The evaluation metrics and processing time demonstrate that the proposed model not only provides superior quantitative results but also trains and segments the LV in less time, indicating its improved performance over competing segmentation models.

Breast cancer is among the most common types of cancer in women and under the cases of misdiagnosed, or delayed in treatment, the mortality risk is high. The existence of breast microcalcifications is common in breast cancer patients and they are an effective indicator for early sign of breast cancer. However, microcalcifications are often missed and wrongly classified during screening due to their small sizes and indirect scattering in mammogram images. Motivated by this issue, this project proposes an adaptive transfer learning deep convolutional neural network in segmenting breast mammogram images with calcifications cases for early breast cancer diagnosis and intervention. Mammogram images of breast microcalcifications are utilized to train several deep neural network models and their performance is compared. Image filtering of the region of interest images was conducted to remove possible artifacts and noises to enhance the quality of the images before the training. Different hyperparameters such as epoch, batch size, etc were tuned to obtain the best possible result. In addition, the performance of the proposed fine-tuned hyperparameter of ResNet50 is compared with another state-of-the-art machine learning network such as ResNet34, VGG16, and AlexNet. Confusion matrices were utilized for comparison. The result from this study shows that the proposed ResNet50 achieves the highest accuracy with a value of 97.58%, followed by ResNet34 of 97.35%, VGG16 96.97%, and finally AlexNet of 83.06%.

In the late December of 2019, a novel coronavirus was discovered in Wuhan, China. In March 2020, WHO announced this epidemic had become a global pandemic and that the novel coronavirus may be mild to most people. However, some people may experience a severe illness that results in hospitalization or maybe death. COVID-19 classification remains challenging due to the ambiguity and similarity with other known respiratory diseases such as SARS, MERS, and other viral pneumonia. The typical symptoms of COVID-19 are fever, cough, chills, shortness of breath, loss of smell and taste, headache, sore throat, chest pains, confusion, and diarrhoea. This research paper suggests the concept of transfer learning using the deterministic algorithm in all binary classification models and evaluates the performance of various CNN architectures. The datasets of 746 CT images of COVID-19 and non-COVID-19 were divided for training, validation, and testing. Various augmentation techniques were applied to increase the number of datasets except for testing images. The images were then pretrained using CNN to obtain a binary class. ResNeXt101 and ResNet152 have the best F1 score of 0.978 and 0.938, whereas GoogleNet has an F1 score of 0.762. ResNeXt101 and ResNet152 have an accuracy of 97.81% and 93.80%. ResNeXt101, DenseNet201, and ResNet152 have 95.71%, 93.81%, and 90% sensitivity, whereas ResNeXt101, ResNet101, and ResNet152 have 100%, 99.58%, and 98.33 specificity, respectively.

Cardiac health diseases are one of the key causes of death around the globe. The number of heart patients has considerably increased during the pandemic. Therefore, it is crucial to assess and analyze the medical and cardiac images. Deep learning architectures, specifically convolutional neural networks have profoundly become the primary choice for the assessment of cardiac medical images. The left ventricle is a vital part of the cardiovascular system where the boundary and size perform a significant role in the evaluation of cardiac function. Due to automatic segmentation and good promising results, the left ventricle segmentation using deep learning has attracted a lot of attention. This article presents a critical review of deep learning methods used for the left ventricle segmentation from frequently used imaging modalities including magnetic resonance images, ultrasound, and computer tomography. This study also demonstrates the details of the network architecture, software, and hardware used for training along with publicly available cardiac image datasets and self-prepared dataset details incorporated. The summary of the evaluation matrices with results used by different researchers is also presented in this study. Finally, all this information is summarized and comprehended in order to assist the readers to understand the motivation and methodology of various deep learning models, as well as exploring potential solutions to future challenges in LV segmentation.

Pneumonia is an infamous life-threatening lung bacterial or viral infection. The latest viral infection endangering the lives of many people worldwide is the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which causes COVID-19. This paper is aimed at detecting and differentiating viral pneumonia and COVID-19 disease using digital X-ray images. The current practices include tedious conventional processes that solely rely on the radiologist or medical consultant’s technical expertise that are limited, time-consuming, inefficient, and outdated. The implementation is easily prone to human errors of being misdiagnosed. The development of deep learning and technology improvement allows medical scientists and researchers to venture into various neural networks and algorithms to develop applications, tools, and instruments that can further support medical radiologists. This paper presents an overview of deep learning techniques made in the chest radiography on COVID-19 and pneumonia cases.

Atherosclerotic plaque deposit in the carotid artery is used as an early estimate to identify the presence of cardiovascular diseases. Ultrasound images of the carotid artery are used to provide the extent of stenosis by examining the intima-media thickness and plaque diameter. A total of 361 images were classified using machine learning and deep learning approaches to recognize whether the person is symptomatic or asymptomatic. CART decision tree, random forest, and logistic regression machine learning algorithms, convolutional neural network (CNN), Mobilenet, and Capsulenet deep learning algorithms were applied in 202 normal images and 159 images with carotid plaque. Random forest provided a competitive accuracy of 91.41% and Capsulenet transfer learning approach gave 96.7% accuracy in classifying the carotid artery ultrasound image database.

Workplace accidents can cause a catastrophic loss to the company including human injuries and fatalities. Occupational injury reports may provide a detailed description of how the incidents occurred. Thus, the narrative is a useful information to extract, classify and analyze occupational injury. This study provides a systematic review of text mining and Natural Language Processing (NLP) applications to extract text narratives from occupational injury reports. A systematic search was conducted through multiple databases including Scopus, PubMed, and Science Direct. Only original studies that examined the application of machine and deep learning-based Natural Language Processing models for occupational injury analysis were incorporated in this study. A total of 27, out of 210 articles were reviewed in this study by adopting the Preferred Reporting Items for Systematic Review (PRISMA). This review highlighted that various machine and deep learning-based NLP models such as K-means, Naïve Bayes, Support Vector Machine, Decision Tree, and K-Nearest Neighbors were applied to predict occupational injury. On top of these models, deep neural networks are also included in classifying the type of accidents and identifying the causal factors. However, there is a paucity in using the deep learning models in extracting the occupational injury reports. This is due to these techniques are pretty much very recent and making inroads into decision-making in occupational safety and health as a whole. Despite that, this paper believed that there is a huge and promising potential to explore the application of NLP and text-based analytics in this occupational injury research field. Therefore, the improvement of data balancing techniques and the development of an automated decision-making support system for occupational injury by applying the deep learning-based NLP models are the recommendations given for future research.

The advancement of technology in medical equipment has significantly improved healthcare services. However, failures in upkeeping reliability, availability, and safety affect the healthcare services quality and significant impact can be observed in operations’ expenses. The effective and comprehensive medical equipment assessment and monitoring throughout the maintenance phase of the asset life cycle can enhance the equipment reliability, availability, and safety. The study aims to develop the prioritisation assessment and predictive systems that measure the priority of medical equipment’s preventive maintenance, corrective maintenance, and replacement programmes. The proposed predictive model is constructed by analysing features of 13,352 medical equipment used in public healthcare clinics in Malaysia. The proposed system comprises three stages: prioritisation analysis, model training, and predictive model development. In this study, we proposed 16 combinations of novel features to be used for prioritisation assessment and prediction of preventive maintenance, corrective maintenance, and replacement programme. The modified k-Means algorithm is proposed during the prioritisation analysis to automatically distinguish raw data into three main clusters of prioritisation assessment. Subsequently, these clusters are fed into and tested with six machine learning algorithms for the predictive prioritisation system. The best predictive models for medical equipment’s preventive maintenance, corrective maintenance, and replacement programmes are selected among the tested machine learning algorithms. Findings indicate that the Support Vector Machine performs the best in preventive maintenance and replacement programme prioritisation predictive systems with the highest accuracy of 99.42 and 99.80%, respectively. Meanwhile, K-Nearest Neighbour yielded the highest accuracy in corrective maintenance prioritisation predictive systems with 98.93%. Based on the promising results, clinical engineers and healthcare providers can widely adopt the proposed prioritisation assessment and predictive systems in managing expenses, reporting, scheduling, materials, and workforce.

Image registration approaches based on standard information criteria were widely employed, showing promising results in the registration of medical monomodal images. Feature-based registering is an effective clinical application technique since computational costs can be reduced significantly. The following four steps are generally used for most registration methods: detection of features, extraction of features, matching features and transformation determination. The accuracy of the registration procedure is dependent on matching a feature and detecting control points (CP). Thus, this paper supports this feature for monomodal registration of magnetic resonance imaging (MRI). MRI is gold-standard imaging to detect prostate cancer progression as it provides better information for soft tissue visualization. In this study, registration of MRI images taken from different time frames is developed to ease the physician in integrating the information about the evolution of the tumour. The accuracy of the registration process depends on matching features and CP detection by calculating the iterative closest point (ICP). The prostate volumes are calculated, and the result shows minimal errors. This registration method has been applied in coronal, sagittal and axial views from five patient datasets. The accuracy of automatic registration results is 0.1mm (axial), 1.1mm (coronal) and 0.2mm (sagittal). These accuracies are comparable to gold-standard manual registration by the experts. There was no significant difference between the automatically and manually registered MRI monomodal. Thus, the proposed method enables the physician to diagnose prostate cancer as it can provide important information about the disease progression and decide the necessary therapies regarding the patient’s condition

Medical equipment highly contributes to the effectiveness of healthcare services quality. Generally, healthcare institutions experience malfunctioning and unavailability of medical equipment that affects the healthcare services delivery to the public. The problems are frequently due to a deficiency in managing and maintaining the medical equipment condition by the responsible party. The assessment of the medical equipment condition is an important activity during the maintenance and management of the equipment life cycle to increase availability, performance, and safety. The study aimed to perform a systematic review in extracting and categorising the input parameters applied in assessing the medical equipment condition. A systematic searching was undertaken in several databases, including Web of Science, Scopus, PubMed, Science Direct, IEEE Xplore, Emerald, Springer, Medline, and Dimensions, from 2000 to 2020. The searching processes were conducted in January 2020. A total of 16 articles were included in this study by adopting Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA). The review managed to classify eight categories of medical equipment reliability attributes, namely equipment features, function, maintenance requirement, performance, risk and safety, availability and readiness, utilisation, and cost. Applying the eight attributes extracted from computerised asset maintenance management system will assist the clinical engineers in assessing the reliability of medical equipment utilised in healthcare institution. The reliability assessment done in these eight attributes will aid clinical engineers in executing a strategic maintenance action, which can increase the equipment’s availability, upkeep the performance, optimise the resources, and eventually contributes in providing effective healthcare service to the community. Finally, the recommendations for future works are presented at the end of this study.

In this review, current studies on hospital readmission due to infection of COVID-19 were discussed, compared, and further evaluated in order to understand the current trends and progress in mitigation of hospital readmissions due to COVID-19. Boolean expression of (“COVID-19” OR “covid19” OR “covid” OR “coronavirus” OR “Sars-CoV-2”) AND (“readmission” OR “re-admission” OR “rehospitalization” OR “rehospitalization”) were used in five databases, namely Web of Science, Medline, Science Direct, Google Scholar and Scopus. From the search, a total of 253 articles were screened down to 26 articles. In overall, most of the research focus on readmission rates than mortality rate. On the readmission rate, the lowest is 4.2% by Ramos-Martínez et al. from Spain, and the highest is 19.9% by Donnelly et al. from the United States. Most of the research (n = 13) uses an inferential statistical approach in their studies, while only one uses a machine learning approach. The data size ranges from 79 to 126,137. However, there is no specific guide to set the most suitable data size for one research, and all results cannot be compared in terms of accuracy, as all research is regional studies and do not involve data from the multi region. The logistic regression is prevalent in the research on risk factors of readmission post-COVID-19 admission, despite each of the research coming out with different outcomes. From the word cloud, age is the most dominant risk factor of readmission, followed by diabetes, high length of stay, COPD, CKD, liver disease, metastatic disease, and CAD. A few future research directions has been proposed, including the utilization of machine learning in statistical analysis, investigation on dominant risk factors, experimental design on interventions to curb dominant risk factors and increase the scale of data collection from single centered to multi centered.

Environmental issues such as environmental pollutions and climate change are the impacts of globalization and become debatable issues among academics and industry key players. One of the environmental issues which is air pollution has been catching attention among industrialists, researchers, and communities around the world. However, it has always neglected until the impacts on human health become worse, and at times, irreversible. Human exposure to air pollutant such as particulate matters, sulfur dioxide, ozone and carbon monoxide contributed to adverse health hazards which result in respiratory diseases, cardiorespiratory diseases, cancers, and worst, can lead to death. This has led to a spike increase of hospitalization and emergency department visits especially at areas with worse pollution cases that seriously impacting human life and health. To address this alarming issue, a predictive model of air pollution is crucial in assessing the impacts of health due to air pollution. It is also critical in predicting the air quality index when assessing the risk contributed by air pollutant exposure. Hence, this systemic review explores the existing studies on anticipating air quality impact to human health using the advancement of Artificial Intelligence (AI). From the extensive review, we highlighted research gaps in this field that are worth to inquire. Our study proposes to develop an AI-based integrated environmental and health impact assessment system using federated learning. This is specifically aims to identify the association of health impact and pollution based on socio-economic activities and predict the Air Quality Index (AQI) for impact assessment. The output of the system will be utilized for hospitals and healthcare services management and planning. The proposed solution is expected to accommodate the needs of the critical and prioritization of sensitive group of publics during pollution seasons. Our finding will bring positive impacts to the society in terms of improved healthcare services quality, environmental and health sustainability. The findings are beneficial to local authorities either in healthcare or environmental monitoring institutions especially in the developing countries.

Background This study presents a registration system that integrates preoperative cardiac Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) volume data with 2D Ultrasound (US) images of the aortic valve. The registration process aims to combine three different imaging modalities (US-CT-MRI) to improve the accuracy of diagnosing aortic valve disorders and provide surgical guidance during the implantation and replacement of the transcatheter aortic valve. Methods The registration framework involves two key components: temporal synchronization and spatial registration. Temporal synchronization allows the identification of frames in the CT and MRI volume that correspond to the same cardiac phase as the US time-series data. For spatial registration, an intensity-based normalized mutual information method combined with a pattern search optimization algorithm is used to produce interpolated cardiac CT and MRI images that align with the US image. Results The accuracy of the trimodality registration method is evaluated using the Dice similarity coefficient. The obtained coefficients are 0.92±0.05 and 0.92±0.04 for comparisons between US-CT and US-MRI, respectively, in short-axis "Mercedes Benz" sign views. The Hausdorff distance, which measures the dissimilarity between two sets of points, was found to be 1.49±0.20 and 1.49±0.19 for both US-CT and US-MRI pairings, respectively. Notably, these values are comparable to the precision achieved when an expert manually registers each image. Conclusions The proposed registration technique demonstrates excellent accuracy in enhancing image-guided systems for aortic valve surgical guidance. It shows promise in the context of Transcatheter Aortic Valve Implantation (TAVI) and Transcatheter Aortic Valve Replacement (TAVR) procedures. The successful integration of US-CT-MRI imaging modalities enables better diagnosis and surgical planning for aortic valve disorders, potentially leading to improved patient outcomes in these procedures.

A computerized bone age assessment requires segmentation of the X-ray carpal bones from other undesired tissue regions. This paper presents segmentation and area measurement of carpal bones in X-ray images. The locally weighted K-means variational level set was applied in segmenting 67 X-ray carpal bone datasets. Dice coefficient and Hausdorff distance measures show mean values above 0.7 and around 3 pixels, respectively. These satisfying segmentation outcomes enable the carpal bone areas to be measured on the segmented images. The carpal bone area measurement ranged from 4.24 mm to 48.96 mm with a mean value of 20.70 ± 10.51 mm and various values of the Pearson’s correlation coefficient implies that the segmentation method is insensitive to different carpal bone areas and locations. These results suggest that the methods can be applied in the bone age assessment by quantifying changes in the carpal bone area over certain time intervals.