scholarly journals Using Artificial Intelligence for COVID-19 Chest X-ray Diagnosis

2020 ◽  
Author(s):  
Andrew A Borkowski ◽  
Narayan A Viswanadham ◽  
L Brannon Thomas ◽  
Rodney D Guzman ◽  
Lauren A Deland ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Positive Predictive Value ◽  
Predictive Value ◽  
Detection System ◽  
Diagnostic Methods ◽  
Normal Lung ◽  
High Morbidity ◽  
X Ray ◽  
Chest X Ray ◽  
Time Required

Coronavirus disease-19 (COVID-19), caused by a novel member of the coronavirus family, is a respiratory disease that rapidly reached pandemic proportions with high morbidity and mortality. It has had a dramatic impact on society and world economies in only a few months. COVID-19 presents numerous challenges to all aspects of healthcare, including reliable methods for diagnosis, treatment, and prevention. Initial efforts to contain the spread of the virus were hampered by the time required to develop reliable diagnostic methods. Artificial intelligence (AI) is a rapidly growing field of computer science with many applications to healthcare. Machine learning is a subset of AI that employs deep learning with neural network algorithms. It can recognize patterns and achieve complex computational tasks often far quicker and with increased precision than humans. In this manuscript, we explore the potential for a simple and widely available test as a chest x-ray (CXR) to be utilized with AI to diagnose COVID-19 reliably. Microsoft CustomVision is an automated image classification and object detection system that is a part of Microsoft Azure Cognitive Services. We utilized publicly available CXR images for patients with COVID-19 pneumonia, pneumonia from other etiologies, and normal CXRs as a dataset to train Microsoft CustomVision. Our trained model overall demonstrated 92.9% sensitivity (recall) and positive predictive value (precision), with results for each label showing sensitivity and positive predictive value at 94.8% and 98.9% for COVID-19 pneumonia, 89% and 91.8% for non-COVID-19 pneumonia, 95% and 88.8% for normal lung. We then validated the program using CXRs of patients from our institution with confirmed COVID-19 diagnoses along with non-COVID-19 pneumonia and normal CXRs. Our model performed with 100% sensitivity, 95% specificity, 97% accuracy, 91% positive predictive value, and 100% negative predictive value. Finally, we developed and described a publicly available website to demonstrate how this technology can be made readily available in the future.

scholarly journals Comparison of Lung Ultrasound to X-ray for Diagnosis of Pulmonary Complications after Cardiac Surgery in Children

2020 ◽  
Vol 18 (1) ◽  
pp. 47-51
Author(s):  
Smriti Mahaju Bajracharya ◽  
Pragati Shrestha ◽  
Apurb Sharma
Keyword(s):  
Cardiac Surgery ◽  
Positive Predictive Value ◽  
Diagnostic Accuracy ◽  
Negative Predictive Value ◽  
Predictive Value ◽  
Lung Ultrasound ◽  
Pulmonary Complications ◽  
Lung Ultrasonography ◽  
X Ray ◽  
Chest X Ray

Background: The purpose of this study was to compare diagnostic performance of lung ultrasound in comparison to chest X-ray to detect pulmonary complication after cardiac surgery in children.Methods: A prospective observational study was conducted in tertiary center of Nepal. 141 consecutive paediatric patients aged less than 14 years scheduled for cardiac surgery were enrolled during the 6 months period. Ultrasound was done on the first post-operative day of cardiac surgery and compared to chest X-ray done on the same day to detect pleural effusion, consolidation, atelectasis and pneumothorax.Results: Sensitivity, specificity, positive and negative predictive values and diagnostic accuracy were calculated using standard formulas. lung ultrasonography had overall sensitivity of 60 %, specificity of 72.4%, positive predictive value of 31.9% and negative predictive value of 89.3% and diagnostic accuracy of 70.2% for diagnosing consolidation. Similarly, lung ultrasonography had overall sensitivity of 90%, specificity of 82.6%, positive predictive value of 46.1% and negative predictive value of 98% and diagnostic accuracy of 83.6 % for diagnosing pleural effusion. For atelectasis, ultrasonography had sensitivity of 50%, specificity of 76.9%, positive predictive value of 30.7% and negative predictive value of 88.2% and diagnostic accuracy of 72.3%. No pneumothoraxes were detected during our study period. Conclusions: Lung ultrasound is an alternative non-invasive technique which is able to diagnose pulmonary complications after cardiac surgery with acceptable diagnostic accuracy with no proven complications but with decreasing exposure to ionizing radiation and possibly cost.Keywords: Cardiac surgery; children; lung ultrasound; pulmonary complications

Pleural Ultrasonography versus Chest Radiography for the Diagnosis of Pneumothorax

QJM ◽  
2021 ◽  
Vol 114 (Supplement_1) ◽  
Author(s):  
Amir Ibrahim Salah ◽  
Heba Bahaa el-Dien El- Serwi ◽  
Amin Mohammad Al Ansary ◽  
Ahmed Badawy Ibrahim Houssien
Keyword(s):  
Positive Predictive Value ◽  
Negative Predictive Value ◽  
Predictive Value ◽  
High Reliability ◽  
Venous Catheter ◽  
Central Line ◽  
X Ray ◽  
Central Line Insertion ◽  
Line Insertion ◽  
Chest X Ray

Abstract Background Pneumothorax (PTX) is an emergency that requires urgent management to avoid catastrophic consequences. PTX is also an important cause of respiratory failure in the emergency department, and may occur frequently as a complication of central venous catheter insertion. Aim The aim of this study is to compare the diagnostic accuracy of bedside lung US with those for anteroposterior (AP) chest X ray (CXR) for the detection of PTX in critically ill patients. Methods This study was conducted on fifty adult patients from both sexes, mechanically ventilated at least 48 hours and planned for central line insertion. We excluded overt pneumothorax, patients requiring immediate invasive intervention, pregnancy and lactation. Lung ultrasound was done to all patients after 30 minutes from central line insertion followed by CXR to confirm the diagnosis of pneumothorax. Pneumothorax was confirmed using CT chest. Results Results showed that ultrasound is superior to chest X Ray in detection of PTX.Ultrasound showed sensitivity of 94.87%, specificity of 81.82%, positive predictive value of 94.87%, negative predictive value of 21.82% and accuracy of 92.0% in detection of PTX, while Chest X Ray showed sensitivity of 76.92%, specificity of 63.64%, positive predictive value of 88.24%, negative predictive value of 43.75% and accuracy of 74.0% in detection of PTX. Conclusions In conclusion, US represent a good approach for the evaluation of PTX, with advantages of timeliness, high accuracy and high reliability.

scholarly journals Diagnosa COVID-19 Chest X-Ray Menggunakan Arsitektur Inception Resnet

2021 ◽  
Vol 2 (1) ◽  
pp. 57-66
Author(s):  
Adhitio Satyo Bayangkari Karno Satyo ◽  
Dodi Arif ◽  
Indra Sari Kusuma Wardhana ◽  
Eka Sally Moreta
Keyword(s):  
Neural Network ◽  
Artificial Intelligence ◽  
Deep Learning ◽  
Computer Tomography ◽  
Viral Pneumonia ◽  
Normal Lung ◽  
X Ray ◽  
Image Dataset ◽  
Chest X Ray ◽  
Medical Aids

The availability of medical aids in adequate quantities is very much needed to assist the work of the medical staff in dealing with the very large number of Covid patients. Artificial Intelligence (AI) with the Deep Learning (DL) method, especially the Convolution Neural Network (CNN), is able to diagnose Chest X-ray images generated by the Computer Tomography Scanner (C.T. Scan) against certain diseases (Covid). Inception Resnet Version 2 architecture was used in this study to train a dataset of 4000 images, consisting of 4 classifications namely covid, normal, lung opacity and viral pneumonia with 1,000 images each. The results of the study with 50 epoch training obtained very good values for the accuracy of training and validation of 95.5% and 91.8%, respectively. The test with 4000 image dataset obtained 98% accuracy testing, with the precision of each class being Covid (99%), Lung_Opacity (97%), Normal (99%) and Viral pneumonia (99%).

scholarly journals Diagnosa COVID-19 Chest X-Ray Dengan Convolution Neural Network Arsitektur Resnet-152

2021 ◽  
Vol 2 (1) ◽  
pp. 26-33
Author(s):  
Widi Hastomo
Keyword(s):  
Neural Network ◽  
Artificial Intelligence ◽  
Deep Learning ◽  
Viral Pneumonia ◽  
Convolution Neural Network ◽  
Normal Lung ◽  
X Ray ◽  
Image Dataset ◽  
Chest X Ray ◽  
Medical Aids

The availability of medical aids in adequate quantities is very much needed to assist the work of the medical staff in dealing with the very large number of Covid patients. Artificial Intelligence (AI) with the Deep Learning (DL) method, especially the Convolution Neural Network (CNN), is able to diagnose Chest X-ray images generated by the Computer Tomography Scanner (C.T. Scan) against certain diseases (Covid). Resnet Version-152 architecture was used in this study to train a dataset of 10.300 images, consisting of 4 classifications namely covid, normal, lung opacity with 3,000 images each and viral pneumonia 1,000 images. The results of the study with 50 epoch training obtained very good values for the accuracy of training and validation of 95.5% and 91.8%, respectively. The test with 10.300 image dataset obtained 98% accuracy testing, with the precision of each class being Covid (99%), Lung_Opacity (99%), Normal (98%) and Viral pneumonia (98%). 

scholarly journals Training Artificial Intelligence Neural Networks to Detect Devices on a Chest X-ray, a Pathway to COVID-19 Testing

2020 ◽  
Vol 112 (5) ◽  
pp. S50
Author(s):  
Zachary Eller ◽  
Michelle Chen ◽  
Jermaine Heath ◽  
Uzma Hussain ◽  
Thomas Obisean ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Neural Networks ◽  
X Ray ◽  
Chest X Ray

scholarly journals Current limitations to identify COVID-19 using artificial intelligence with chest X-ray imaging

2021 ◽  
Vol 11 (2) ◽  
pp. 411-424 ◽  
Author(s):  
José Daniel López-Cabrera ◽  
Rubén Orozco-Morales ◽  
Jorge Armando Portal-Diaz ◽  
Orlando Lovelle-Enríquez ◽  
Marlén Pérez-Díaz
Keyword(s):  
Artificial Intelligence ◽  
X Ray ◽  
X Ray Imaging ◽  
Chest X Ray

Przewlekły kaszel jako główna manifestacja chłoniaka Hodgkina – opis przypadku

2019 ◽  
Vol 23 (3) ◽  
Author(s):  
Katarzyna Wójcicka ◽  
Andrzej Pogorzelski
Keyword(s):  
Chronic Cough ◽  
Correct Diagnosis ◽  
Diagnostic Methods ◽  
Radiographic Examination ◽  
Older Children ◽  
X Ray ◽  
Volume Curve ◽  
Flow Volume Curve ◽  
Mediastinal Widening ◽  
Chest X Ray

A cough lasting longer than 4-8 weeks, defined as chronic cough, always requires thorough diagnostic evaluation. In addition to detailed history-taking and physical examination, simple and available diagnostic methods, such as chest x-ray and spirometry, should be performed. They may be helpful tool to establish the underlying cause of cough. Many younger children may have difficulties in performing the forced expiratory maneuvers and fulfilling repeatability criteria for spirometry. The disturbances resulting from insufficient cooperation should be considered in interpratation of the obtained results. The shape of the flow-volume curve, which suggests upper or central airways obstruction, can not be ignored and always requires further investigation for diagnosis of respiratory pathology. The chest x-ray is the most frequently performed radiographic examination in children. Accurate interpretation is essential in reaching a correct diagnosis. Mediastinal widening on the chest x-ray in children can occur due to a large variety of causes. The normal thymus can take on a variety of sizes and shapes and still be considered normal in the first few years of life. In older children mediastinal widening should be differentiated from mediastinal masses. Lymph node enlargement represents a frequent cause, usually as a result of infection or malignancy. The article reports a case of a 12-year-old boy with chronic cough, mediastinal widening on the chest X-ray and abnormal spirometry results, who was finally diagnosed with stage III Hodgkin’s lymphoma.

Development of laboratory x-ray fluorescence holography equipment

2003 ◽  
Vol 18 (6) ◽  
pp. 1471-1473 ◽  
Author(s):  
Yukio Takahashi ◽  
Kouichi Hayashi ◽  
Kimio Wakoh ◽  
Naomi Nishiki ◽  
Eiichiro Matsubara
Keyword(s):  
Single Crystal ◽  
Count Rate ◽  
Detection System ◽  
High Count ◽  
Synchrotron Source ◽  
High Count Rate ◽  
X Ray ◽  
Graphite Monochromator ◽  
Time Required

Laboratory x-ray fluorescence holography equipment was developed. A single-bent graphite monochromator with a large curvature and a high-count-rate x-ray detection system were applied in this equipment. To evaluate the performance of this equipment, a hologram pattern of a gold single crystal was measured. It took two days, which was about one-third the time required for the previous measurements using the conventional x-ray source and several times that using the synchrotron source. The quality of the hologram pattern is as good as that obtained using the synchrotrons. Clear atomic images on (002) are reconstructed.

scholarly journals Histopathological Changes in Helicobacter pylori Associated Gastritis and Scope of Special Stain and Immunohistochemistry as Diagnostic Aids

2020 ◽  
Vol 7 (50) ◽  
pp. 3027-3032
Author(s):  
Ruby Elizabeth Elias ◽  
Bindiya Gisuthan ◽  
Sreeganesh A.S
Keyword(s):  
Helicobacter Pylori ◽  
Positive Predictive Value ◽  
Negative Predictive Value ◽  
Chronic Gastritis ◽  
Predictive Value ◽  
Diagnostic Methods ◽  
P Value ◽  
Special Stain ◽  
Frequent Finding ◽  
H Pylori

BACKGROUND Helicobacter pylori associated chronic gastritis plays a vital role in the development of majority of gastric adenocarcinomas and most gastric MALT (Mucosa Associated Lymphoid Tissue) lymphomas. Many diagnostic methods are available for the identification of this organism. However, in gastroenterology practice, histopathological examination of biopsy samples provides visual identification of the pathogen and the associated mucosal changes with special stains like Giemsa. The aim of this study was to evaluate the efficacy of three stains H & E- (Haematoxylin and Eosin), Giemsa and IHC (Immunohistochemistry) in the identification of H. pylori. Associated histologic changes were noted and the relationship between the degree of colonisation and the activity and chronicity of gastritis were analysed. METHODS 585 gastric biopsies taken from dyspeptic patients were evaluated for gastritis, based on updated Sydney System. In 250 randomly selected cases, three staining methods were used. RESULTS Out of 585 cases, 413 (70.60 %) had features of chronic gastritis. Mild chronic gastritis was the commonest finding and is seen in most cases of mild H. pylori colonisation. When activity was monitored, mild activity was the most frequent finding [225 (38.46 %)]. Majority of the severe activity cases showed severe H. pylori colonisation. 13.16 %, 4.79 % and 7.35 % showed intestinal metaplasia, atrophy and dysplastic changes respectively. Out of 250 cases, H & E and Giemsa stains showed 45.6 % and 57.2 % positivity while IHC demonstrated maximum number of positivity (156 cases - 62.4 %). Sensitivity and specificity of H & E was found to be 77.90 % and 98.95 %, positive predictive value was 99.13 % and negative predictive value was 69.18 %. For Giemsa stain, sensitivity was 91.67 %, specificity was 100 %, positive predictive value was 100 % and negative predictive value was 87.85 %. DISCUSSION H. pylori gastritis was a frequent finding in dyspeptic patients in southern part of India. When chi-square test was done, a significant statistical relationship between the severity of H. pylori colonisation, activity and chronicity of gastritis was noted. P value was < 0.001. With the use of special stain, Giemsa and ancillary techniques like IHC, the detection rate of H. pylori was enhanced considerably. CONCLUSIONS With increasing number of H. pylori in the mucosa, there was increase in the chronicity and activity of gastritis. Although immunohistochemistry revealed more cases of H. pylori, Giemsa can be a cost-effective substitute, because of its high specificity and positive predictive value. KEYWORDS H. pylori Gastritis, Giemsa, Haematoxylin and Eosin Stain, Immunohistochemistry

scholarly journals Artificial Intelligence Framework for Efficient Detection and Classification of Pneumonia Using Chest Radiography Images

2021 ◽  
Author(s):  
Ali Mohammad Alqudah ◽  
Shoroq Qazan ◽  
Ihssan S. Masad
Keyword(s):  
Artificial Intelligence ◽  
Input Image ◽  
Image Size ◽  
X Ray ◽  
Softmax Classifier ◽  
Non Invasive ◽  
Efficient Detection ◽  
X Ray Imaging ◽  
Chest X Ray

Abstract BackgroundChest diseases are serious health problems that threaten the lives of people. The early and accurate diagnosis of such diseases is very crucial in the success of their treatment and cure. Pneumonia is one of the most widely occurred chest diseases responsible for a high percentage of deaths especially among children. So, detection and classification of pneumonia using the non-invasive chest x-ray imaging would have a great advantage of reducing the mortality rates.ResultsThe results showed that the best input image size in this framework was 64 64 based on comparison between different sizes. Using CNN as a deep features extractor and utilizing the 10-fold methodology the propose artificial intelligence framework achieved an accuracy of 94% for SVM and 93.9% for KNN, a sensitivity of 93.33% for SVM and 93.19% for KNN and a specificity of 96.68% for SVM and 96.60% for KNN.ConclusionsIn this study, an artificial intelligence framework has been proposed for the detection and classification of pneumonia based on chest x-ray imaging with different sizes of input images. The proposed methodology used CNN for features extraction that were fed to two different types of classifiers, namely, SVM and KNN; in addition to the SoftMax classifier which is the default CNN classifier. The proposed CNN has been trained, validated, and tested using a large dataset of chest x-ray images contains in total 5852 images.

Sign in / Sign up

Export Citation Format

Share Document