A Novel Method for COVID-19 Diagnosis Using Artificial Intelligence in Chest X-ray Images.

The Coronavirus disease 2019 (COVID-19) is an infectious disease spreading rapidly and uncontrollably throughout the world. The critical challenge is the rapid detection of Coronavirus infected people. The available techniques being utilized are body-temperature measurement, along with anterior nasal swab analysis. However, taking nasal swabs and lab testing are complex, intrusive, and require many resources. Furthermore, the lack of test kits to meet the exceeding cases is also a major limitation. The current challenge is to develop some technology to non-intrusively detect the suspected Coronavirus patients through Artificial Intelligence (AI) techniques such as deep learning (DL). Another challenge to conduct the research on this area is the difficulty of obtaining the dataset due to a limited number of patients giving their consent to participate in the research study. Looking at the efficacy of AI in healthcare systems, it is a great challenge for the researchers to develop an AI algorithm that can help health professionals and government officials automatically identify and isolate people with Coronavirus symptoms. Hence, this paper proposes a novel method CoVIRNet (COVID Inception-ResNet model), which utilizes the chest X-rays to diagnose the COVID-19 patients automatically. The proposed algorithm has different inception residual blocks that cater to information by using different depths feature maps at different scales, with the various layers. The features are concatenated at each proposed classification block, using the average-pooling layer, and concatenated features are passed to the fully connected layer. The efficient proposed deep-learning blocks used different regularization techniques to minimize the overfitting due to the small COVID-19 dataset. The multiscale features are extracted at different levels of the proposed deep-learning model and then embedded into various machine-learning models to validate the combination of deep-learning and machine-learning models. The proposed CoVIR-Net model achieved 95.7% accuracy, and the CoVIR-Net feature extractor with random-forest classifier produced 97.29% accuracy, which is the highest, as compared to existing state-of-the-art deep-learning methods. The proposed model would be an automatic solution for the assessment and classification of COVID-19. We predict that the proposed method will demonstrate an outstanding performance as compared to the state-of-the-art techniques being used currently.

Risk Factors Associated with Statin-Associated Muscle Symptoms in Patients Attending a Specialized Regional Lipid Clinic.

BACKGROUND: Statin-associated muscle symptoms (SAMS) are the major side effects reported for statins. Data from previous studies suggest that 7-29% of patients on statin had associated muscle symptoms. In the UK, there is a lack of corresponding data on SAMS and factors associated with the development of SAMS. OBJECTIVE: This analysis is aimed at establishing the prevalence of SAMS and identifying major contributory risk factors in patients attending a lipid clinic. METHODS: Clinical records of 535 consecutive patients, who visited the lipid clinic in the University Hospitals of Leicester, were studied retrospectively between 2009 and 2012. SAMS were defined by the presence of muscle symptoms with two or more different statins. Patients who reported muscle symptoms to statin with one or no rechallenge were excluded. The association of SAMS with clinical characteristics such as age and BMI, sex, smoking, excess alcohol, comorbidities, and medications was tested for statistical significance. A binomial logistic regression model was applied to adjust for risk factors significantly associated with SAMS. RESULTS: The prevalence of SAMS was found to be 11%. On unadjusted analysis, the mean age of patients who had SAMS was significantly higher than those without SAMS (59.4 ± 10.5 years vs. 50.3 ± 13.4 years, respectively, P < 0.001). Nonsmokers were more likely to develop SAMS in comparison to active smokers (P = 0.037). Patients taking antihypertensive medications were more likely to develop SAMS (P = 0.010). In binomial logistic regression analysis, only age was positively and significantly associated with SAMS after adjusting for other risk factors (ß = 0.054, P = 0.001). CONCLUSION: To the best of our knowledge, this study is the largest cohort of patients with SAMS in the United Kingdom. Our data suggest that the prevalence of SAMS is 11% and increased age is a risk factor associated with the development of SAMS in our cohort of patients.

Comparison of estimated and calculated fetal radiation dose for a pregnant woman who underwent computed tomography and conventional X-ray examinations based on a phantom study.

This study aimed to determine the mean fetal doses for patients who underwent computed tomography (CT) and/or conventional X-ray (CXR) examinations. In addition to developing an approach to estimate the fetal dose based on data registered in the picture archive and communication system (PACS), the radiation doses for pregnant women and their fetuses were estimated using the VirtualDoseCT and VirtualDoseIR softwares. To verify the data, the fetal dose was measured using thermoluminescent dosimeters (TLDs) implanted at different uterus sites of an anthropomorphic pregnant phantom. Calculated fetal dose values were estimated in relation to the dose-area product (DAP) and volume CT dose index (CTDIvol). DAP and CTDIvol were obtained from data registered in the PACS. The fetal doses varied between < 0.001 and 3.9 mGy and between 0.26 and 16.21 mGy for the CXR and CT examinations, respectively. These values were similar to those of previous studies on both imaging modalities. The conversion factors obtained to calculate fetal doses for CXR examinations were between 0.01 and 0.73 mGy/Gy cm2, whereas they varied between 0.02 and 0.61 mGy/mGy for CT examinations. Overall, the fetal dose conversion factors based on DAP and CTDIvol values can be used for fast fetal dose estimations in common CXR and CT examinations.

Assessment of computed tomography radiation doses for paediatric head and chest examinations using paediatric phantoms of three different ages.

INTRODUCTION: With the rapid development of computed tomography (CT) scanners, the assessment of the radiation dose received by the patient has become a heavily researched topic and may result in a reduction in radiation exposure risk. In this study, radiation doses were measured using three paediatric phantoms for head and chest CT examinations in Najran, Saudi Arabia. METHODS: Thirteen scanners were included in the study to estimate the CT radiation doses using three phantoms representing three age groups (1-, 5-, and 10-year-old patients). RESULTS: The volume CT dose index (CTDIvol) estimated for each phantom ranged from 6.56 to 41.12 mGy and 0.292 to 11.10 mGy for the head and chest examinations, respectively. The estimation of lifetime attributable risk (LAR) indicated that the cancer risk could reach approximately 0.02-0.16% per 500 children undergoing head and chest CT examinations. CONCLUSION: The comparison with the published data of the European Commission (EC) and countries reported in this study revealed that the mean CTDIvol for the head examinations was within the recommended dose reference levels (DRLs). Meanwhile, chest results exceeded the international DRLs for the one-year-old phantoms, suggesting that optimisation work is required at a number of sites. IMPLICATIONS FOR PRACTICE: The variation among CT doses reported in this study showed that substantial standardisation is needed.

A comparison of the CT-dosimetry software packages based on stylized and boundary representation phantoms.

INTRODUCTION: With the rapid development of computed tomography (CT) equipment, the assessment of effective and organ dose using suitable tools becomes an important issue and will provide health professionals with useful information regarding the radiation risks and the development of standard imaging protocols. Different clinical centres and/or institutions may use several software packages, each with different methods and algorithms for CT dose evaluation. Consequently, radiation doses calculated with these computer software packages might be different for the same patient and representative scanner models. METHODS: The effective and organ doses calculated by VirtualDose, CT-expo, and ImPACT software were compared for both males and females using kidney, chest, head, pelvis, abdomen, and whole-body CT protocols. The calculation of radiation dose in these software depends on the use of stylized and boundary representation (BREP) phantoms. RESULTS: In general, the results showed that there was a discrepancy between the effective dose values calculated by the three packages. The effective dose in all examinations varied by factors ranging from 1.1 to 1.5 for male and from 1.1 to 1.3 for female. For the female phantom, the VirtualDose shows the highest effective doses in kidney and abdomen examinations while CT-expo gives the highest doses for head and pelvis examinations. For the male phantom, the VirtualDose shows the highest effective doses were for chest examinations. CONCLUSION: VirtualDose approach gives the most accurate estimation, however, further work using a size-based method are necessary to improve the assessment of the effective and equivalent organ dose in CT examinations using these packages. IMPLICATIONS FOR PRACTICE: The re-evaluation dosimetry software in comparison with patient size would allow for a more accurate estimation of dose and support the optimization process.

DOSE MEASUREMENT USING GAFCHROMIC FILM FOR PATIENTS UNDERGOING INTERVENTIONAL CARDIOLOGY PROCEDURES.

The aim of this study was to evaluate dose area product (DAP) and skin dose to patients undergoing coronary angiography (CA) and percutaneous transluminal coronary angioplasty (PTCA) using GafChromic XR film at King Khalid Hospital, Najran, Saudi Arabia. The entrance skin doses (ESDs) were calculated from DAP using GafChromic XR film placed on the patients back. The mean DAP obtained for patients undergoing CA and PTCA examinations were found to be 31.4 and 74.2 Gy cm2, respectively. The mean ESD using GafChromic XR film was found to be 0.264 and 0.596 Gy for CA and PTCA examinations, respectively.

Regional survey of entrance surface dose to patients from X-ray examinations in Saudi Arabia.

The aim of this study was to evaluate the entrance surface doses (ESDs) to patients undergoing chest and lumbar spine X-ray examinations in Najran, Saudi Arabia. ESD per examination was estimated from X-ray tube output parameters in two hospitals comprising three X-ray units and a sample of 137 radiographs. Hospital mean ESDs estimated range from 0.068 to 0.34 mGy for chest posteroanterior, 0.44-3.42 mGy for lumbar spine anteroposterior and 0.96-7.98 mGy for lumbar spine lateral. The results are useful to national and professional organizations and can be used as a baseline upon which future dose measurements may be compared.

Direct air activation measurements at a 15-MV medical linear accelerator.

Direct radiometric determination of (14)N (γ, n) (13)N air activation was achieved at a 15-MV medical linear accelerator operating in a high-energy photon mode. (13)N was identified by irradiating a gas-tight Marinelli beaker filled with nitrogen gas and later observing the 10-min half-life of the 511-keV positron-electron annihilation line using high-resolution gamma spectroscopy. Quantitative evaluation of the spectral signal yielded a (13)N production rate of 836.8 ± 32 Bq Gy(-1) in air per 40 × 40 cm(2) field cross section at 100 cm source-surface distance.

Regional survey of image quality and radiation dose in computed tomography examinations in Saudi Arabia.

This study is the first regional investigation in Najran, Saudi Arabia aimed at investigating radiation dose and image quality of computed tomography (CT) examinations. The survey data was collected from five scanners in four hospitals. For all CT scanners, a correction factor was calculated to measure the weighted computed tomography dose index (CTDIw) using standard dosimetry phantoms. The CTDIw were reported in this study and compared with other countries. It was found that most CTDIw values were close to the European reference levels and in line with the results of similar surveys in the other parts of world. Concerning image quality, 80 % of the scanners were found to be in compliance with the relative international guidelines for all the examined parameters.

Paediatric dose measurements for chest X-ray examinations at Maternity and Children Hospital in Najran - Saudi Arabia.

The entrance skin dose (ESD) of chest X-ray examinations for AP and PA projections of paediatric patients at Maternity and Children Hospital in Najran, Saudi Arabia have been obtained using DoseCal software. The majority of the results obtained show low measured ESD for chest X-ray examinations. The mean of ESD for the AP projection was found to be 37.5, 40.5, 41.3, and 52.3 µ Gy for age groups 0-1, >1-5, >5-10, and >10-15 years respectively. However, the ESD for PA projection was found to be 50.7 and 56.7 µ Gy for age groups >5-10, and >10-15 years respectively.

Doses to patients from photoneutrons emitted in a medical linear accelerator.

This study of doses to patients from emitted photoneutrons in a medical linear accelerator (Varian 2100C) was carried out. Dose calculation was performed using Monte Carlo Geant4 code. The model was used to calculate the neutron fluence, as a function of the neutron energy inside the treatment room to estimate the effective dose to patients. The ambient dose equivalent versus field size for patients is reported in this study. The ambient dose equivalent using 1 x 1 cm(2) field size, at isocentre and X-ray modes of 20, 18, 15 and 10 MV, was found to be 1.85, 1.79, 0.61 and 0.06 mSv Gy(-1), respectively. The mean energies of emitted photoneutrons were 0.48, 0.44, 0.40 and 0.16 MeV at X-ray modes of 20, 18, 15 and 10 MV, respectively. The results of ambient dose equivalent from emitted photoneutrons cannot be ignored and can represent a risk for healthy tissues. This study emphasised that Geant4 Monte Carlo code is an appropriate choice for studying photoneutron production and transport.