scholarly journals A deep learning algorithm using CT images to screen for Corona Virus Disease (COVID-19)

2020 ◽  
Author(s):  
Shuai Wang ◽  
Bo Kang ◽  
Jinlu Ma ◽  
Xianjun Zeng ◽  
Mingming Xiao ◽  
...  
Keyword(s):  
Deep Learning ◽  
Virus Disease ◽  
False Negative ◽  
Critical Time ◽  
Ct Images ◽  
Diagnostic Methods ◽  
Internal Validation ◽  
Negative Results ◽  
Testing Dataset ◽  
Total Accuracy

AbstractBackgroundThe outbreak of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-COV-2) has caused more than 2.5 million cases of Corona Virus Disease (COVID-19) in the world so far, with that number continuing to grow. To control the spread of the disease, screening large numbers of suspected cases for appropriate quarantine and treatment is a priority. Pathogenic laboratory testing is the gold standard but is time-consuming with significant false negative results. Therefore, alternative diagnostic methods are urgently needed to combat the disease. Based on COVID-19 radiographical changes in CT images, we hypothesized that Artificial Intelligence’s deep learning methods might be able to extract COVID-19’s specific graphical features and provide a clinical diagnosis ahead of the pathogenic test, thus saving critical time for disease control.Methods and FindingsWe collected 1,065 CT images of pathogen-confirmed COVID-19 cases (325 images) along with those previously diagnosed with typical viral pneumonia (740 images). We modified the Inception transfer-learning model to establish the algorithm, followed by internal and external validation. The internal validation achieved a total accuracy of 89.5% with specificity of 0.88 and sensitivity of 0.87. The external testing dataset showed a total accuracy of 79.3% with specificity of 0.83 and sensitivity of 0.67. In addition, in 54 COVID-19 images that first two nucleic acid test results were negative, 46 were predicted as COVID-19 positive by the algorithm, with the accuracy of 85.2%.ConclusionThese results demonstrate the proof-of-principle for using artificial intelligence to extract radiological features for timely and accurate COVID-19 diagnosis.Author summaryTo control the spread of the COVID-19, screening large numbers of suspected cases for appropriate quarantine and treatment measures is a priority. Pathogenic laboratory testing is the gold standard but is time-consuming with significant false negative results. Therefore, alternative diagnostic methods are urgently needed to combat the disease. We hypothesized that Artificial Intelligence’s deep learning methods might be able to extract COVID-19’s specific graphical features and provide a clinical diagnosis ahead of the pathogenic test, thus saving critical time. We collected 1,065 CT images of pathogen-confirmed COVID-19 cases along with those previously diagnosed with typical viral pneumonia. We modified the Inception transfer-learning model to establish the algorithm. The internal validation achieved a total accuracy of 89.5% with specificity of 0.88 and sensitivity of 0.87. The external testing dataset showed a total accuracy of 79.3% with specificity of 0.83 and sensitivity of 0.67. In addition, in 54 COVID-19 images that first two nucleic acid test results were negative, 46 were predicted as COVID-19 positive by the algorithm, with the accuracy of 85.2%. Our study represents the first study to apply artificial intelligence to CT images for effectively screening for COVID-19.

scholarly journals A deep learning algorithm using CT images to screen for Corona virus disease (COVID-19)

2021 ◽  
Author(s):  
Shuai Wang ◽  
Bo Kang ◽  
Jinlu Ma ◽  
Xianjun Zeng ◽  
Mingming Xiao ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Deep Learning ◽  
Learning Algorithm ◽  
Virus Disease ◽  
Ct Images ◽  
Viral Pneumonia ◽  
Diagnostic Methods ◽  
Deep Learning Algorithm ◽  
Corona Virus ◽  
Total Accuracy

Abstract Objective The outbreak of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-COV-2) has caused more than 26 million cases of Corona virus disease (COVID-19) in the world so far. To control the spread of the disease, screening large numbers of suspected cases for appropriate quarantine and treatment are a priority. Pathogenic laboratory testing is typically the gold standard, but it bears the burden of significant false negativity, adding to the urgent need of alternative diagnostic methods to combat the disease. Based on COVID-19 radiographic changes in CT images, this study hypothesized that artificial intelligence methods might be able to extract specific graphical features of COVID-19 and provide a clinical diagnosis ahead of the pathogenic test, thus saving critical time for disease control. Methods We collected 1065 CT images of pathogen-confirmed COVID-19 cases along with those previously diagnosed with typical viral pneumonia. We modified the inception transfer-learning model to establish the algorithm, followed by internal and external validation. Results The internal validation achieved a total accuracy of 89.5% with a specificity of 0.88 and sensitivity of 0.87. The external testing dataset showed a total accuracy of 79.3% with a specificity of 0.83 and sensitivity of 0.67. In addition, in 54 COVID-19 images, the first two nucleic acid test results were negative, and 46 were predicted as COVID-19 positive by the algorithm, with an accuracy of 85.2%. Conclusion These results demonstrate the proof-of-principle for using artificial intelligence to extract radiological features for timely and accurate COVID-19 diagnosis. Key Points • The study evaluated the diagnostic performance of a deep learning algorithm using CT images to screen for COVID-19 during the influenza season. • As a screening method, our model achieved a relatively high sensitivity on internal and external CT image datasets. • The model was used to distinguish between COVID-19 and other typical viral pneumonia, both of which have quite similar radiologic characteristics.

COVID-19 pneumonia diagnosis using a simple 2D deep learning framework with a single chest CT image (Preprint)

2020 ◽  
Author(s):  
Jinseok Lee
Keyword(s):  
Deep Learning ◽  
Diagnostic Performance ◽  
Ct Images ◽  
Chest Ct ◽  
University Hospital ◽  
Detection Accuracy ◽  
Ct Image ◽  
Test Dataset ◽  
Learning Framework ◽  
Testing Dataset

BACKGROUND The coronavirus disease (COVID-19) has explosively spread worldwide since the beginning of 2020. According to a multinational consensus statement from the Fleischner Society, computed tomography (CT) can be used as a relevant screening tool owing to its higher sensitivity for detecting early pneumonic changes. However, physicians are extremely busy fighting COVID-19 in this era of worldwide crisis. Thus, it is crucial to accelerate the development of an artificial intelligence (AI) diagnostic tool to support physicians. OBJECTIVE We aimed to quickly develop an AI technique to diagnose COVID-19 pneumonia and differentiate it from non-COVID pneumonia and non-pneumonia diseases on CT. METHODS A simple 2D deep learning framework, named fast-track COVID-19 classification network (FCONet), was developed to diagnose COVID-19 pneumonia based on a single chest CT image. FCONet was developed by transfer learning, using one of the four state-of-art pre-trained deep learning models (VGG16, ResNet50, InceptionV3, or Xception) as a backbone. For training and testing of FCONet, we collected 3,993 chest CT images of patients with COVID-19 pneumonia, other pneumonia, and non-pneumonia diseases from Wonkwang University Hospital, Chonnam National University Hospital, and the Italian Society of Medical and Interventional Radiology public database. These CT images were split into a training and a testing set at a ratio of 8:2. For the test dataset, the diagnostic performance to diagnose COVID-19 pneumonia was compared among the four pre-trained FCONet models. In addition, we tested the FCONet models on an additional external testing dataset extracted from the embedded low-quality chest CT images of COVID-19 pneumonia in recently published papers. RESULTS Of the four pre-trained models of FCONet, the ResNet50 showed excellent diagnostic performance (sensitivity 99.58%, specificity 100%, and accuracy 99.87%) and outperformed the other three pre-trained models in testing dataset. In additional external test dataset using low-quality CT images, the detection accuracy of the ResNet50 model was the highest (96.97%), followed by Xception, InceptionV3, and VGG16 (90.71%, 89.38%, and 87.12%, respectively). CONCLUSIONS The FCONet, a simple 2D deep learning framework based on a single chest CT image, provides excellent diagnostic performance in detecting COVID-19 pneumonia. Based on our testing dataset, the ResNet50-based FCONet might be the best model, as it outperformed other FCONet models based on VGG16, Xception, and InceptionV3.

scholarly journals UTILITY OF ANTIGEN DETECTION TEST AND POLYMERASE CHAIN REACTION IN THE DIFFERENTIATION OF TUBERCULOUS AND NON-TUBERCULOUS MYCOBACTERIA

2017 ◽  
Vol 10 (6) ◽  
pp. 289
Author(s):  
Yogita Singh ◽  
Raji Vasanth ◽  
Shrikala Baliga ◽  
Dhanashree B
Keyword(s):  
Polymerase Chain Reaction ◽  
False Negative ◽  
Diagnostic Methods ◽  
Clinical Samples ◽  
Chain Reaction ◽  
College Hospital ◽  
Medical College ◽  
Negative Results ◽  
Polymerase Chain ◽  
Mycobacterium Spp

Objectives: Cultivation and identification of mycobacteria to species level remains difficult and time-consuming. Hence, easy and rapid diagnostic methods are necessary for the differentiation of Mycobacterium tuberculosis (MTB) from non-tuberculous mycobacteria (NTM). The present study aims to detect and differentiate MTB from NTM isolated from clinical samples by immunochromatographic test (ICT) and polymerase chain reaction (PCR). Methods: Over a period of 1 year, clinical samples (n=496) received from suspected cases of TB, at the Department of Microbiology, Kasturba Medical College Hospital, Mangalore were cultured to isolate Mycobacterium spp. Identification of all the isolates was done by conventional biochemical technique, ICT, and PCR. Results: Among the 496 samples processed, 49 (9.87%) were acid-fast bacilli smear positive and 59 (11.89%) samples showed the growth of Mycobacterium spp. Among these, 10 were rapid growers, 49 were slow-growing mycobacteria, out of which 30 were MTB as identified by conventional biochemical reaction. Out of 59 Mycobacterial isolates subjected to ICT for the detection of MPT 64 antigen, only 28 were identified as MTB. However, all the 30 isolates were correctly identified as MTB by PCR. Conclusion: Hence, PCR is essential for rapid differentiation of non-tuberculous Mycobacterium from MTB. False negative results seen with immunochromatographic MPT 64 antigen assay could be due to mutations within the mpt64 gene. Further studies are necessary to characterize these PCR-positive and immunochromatographic assay negative MTB isolates.

scholarly journals Detection of LytA Genes in Streptococcus pneumoniae Isolated from sputum pneumonia patients

Biomedika ◽  
2020 ◽  
Vol 13 (1) ◽  
pp. 23-30
Author(s):  
Mustika Sari Hutabarat ◽  
Firdaus Hamid ◽  
Irawaty Djaharuddin ◽  
Alfian Zainuddin ◽  
Rossana Agus ◽  
...  
Keyword(s):  
Streptococcus Pneumoniae ◽  
False Negative ◽  
Pneumococcal Infection ◽  
Diagnostic Methods ◽  
Molecular Diagnostic ◽  
Clinical Samples ◽  
Biochemical Tests ◽  
Negative Results ◽  
False Negative Results ◽  
Polymerase Chain

Streptococcus pneumoniae (pneumococcus) is a Gram-positive facultative anaerobic bacterium that is a major cause of morbidity and mortality worldwide. But the lack of reporting of disease by this bacterium in Indonesia, one of the causes is because the diagnosis of pneumococcal infection is often clinically not typical and conventional methods which are still the standard gold method often give false-negative results. So the purpose of this study was to evaluate the performance of culture and molecular diagnostic methods using the Polymerase Chain Reaction (PCR) technique in detecting Streptococcus pneumoniae in sputum clinical samples using the Autolysin (LytA) gene which is a virulence factor of this bacterium. 57 isolates from 60 samples were confirmed as Streptococcus sp through microscopic identification, culture, and biochemical tests. Then the sensitivity test with an optochin test of 9 (9%) compared the results descriptively with the PCR technique using the Autolysin A (LytA) gene which was obtained more sensitive by 15 (25%).

The application of viral nucleic acid detection and lung CT in COVID-19 diagnosis

2020 ◽  
Author(s):  
Rui Hu
Keyword(s):  
Nucleic Acid ◽  
Clinical Symptoms ◽  
Virus Disease ◽  
False Negative ◽  
False Negative Rate ◽  
Diagnostic Methods ◽  
Nucleic Acid Detection ◽  
Viral Nucleic Acid ◽  
Early Medical Intervention ◽  
Lung Ct

The Corona Virus Disease 2019 (COVID-19) has the characteristics of fast propagation speed and strong pathogenicity and has attracted wide attention of people, medical workers, and researchers around the world. Accurate, rapid, and timely screening and diagnosis of COVID-19 is of great significance to control the development of the epidemic situation and save the lives of patients. Currently, the detection of viral nucleic acid and lung CT is the main screening and diagnostic methods of COVID-19. Nucleic acid detection has the advantages of fast, strong specificity and high sensitivity, but there is a certain false-negative rate. CT result of lung examination is visual, but it is not typical due to the uncertain time of clinical symptoms and the early medical intervention. Therefore, the diagnosis of COVID-19 should include a combination of epidemiological history, clinical symptoms, imaging, and laboratory tests.

scholarly journals Can We Use Targeted Next-Generation Sequencing an Alternative Method to the Conventional Tests in Haematological Malignancies?

2020 ◽  
Author(s):  
Emine Ikbal Atli ◽  
Hakan Gurkan ◽  
Engin Atli ◽  
Hakki Onur Kirkizlar ◽  
Sinem Yalcintepe ◽  
...  
Keyword(s):  
False Negative ◽  
Molecular Cytogenetics ◽  
Hematologic Malignancies ◽  
Turnaround Time ◽  
Diagnostic Methods ◽  
Myeloid Malignancies ◽  
Targeted Next Generation Sequencing ◽  
Negative Results ◽  
Golden Standard ◽  
Molecular Cytogenetic Techniques

Abstract Introduction: Advanced diagnostic methods give a huge advantage for identification of the abnormalities in myeloid malignancies. Researchers tried to show the potential importance of genetic tests both before the onset of the disease and during the remission. Large testing panels prevents false negative results in myeloid malignancies. But the important question is how can be merged with conventional cytogenetic and molecular cytogenetic techniques together with NGS technologies. Methods: In this paper, we draw an algorithm for evaluation of the malignancies. In order to evaluation of genetic abnormities we performed cytogenetics, molecular cytogenetics and NGS testing panels in hematologic malignancies. In this study, we analyzed 132 patients which are referred to Medical Genetics Laboratory within different type of hematologic malignancies. We highlighted possible algorithm for cytogenetically normal cases.Results: We analyzed cytogenetically normal patients by using NGS 141 gene panel and we detected two or more pathogenic variations in 20 out of 132 patients.Conclusions: Despite of long turnaround time conventional techniques is still golden standard for myeloid malignancies but sometimes cryptic gene fusions or complex abnormalities cannot easily be identified by conventional techniques that conditions advanced technologies are recommended.

scholarly journals Evaluation of the Allplex™ GI-Helminth(I) Assay, the first marketed multiplex PCR for helminth diagnosis

Parasite ◽  
2021 ◽  
Vol 28 ◽  
pp. 33
Author(s):  
Brice Autier ◽  
Jean-Pierre Gangneux ◽  
Florence Robert-Gangneux
Keyword(s):  
Multiplex Pcr ◽  
False Negative ◽  
Extraction Process ◽  
Trichuris Trichiura ◽  
Diagnostic Methods ◽  
Bead Beating ◽  
Negative Results ◽  
Multiplex Pcr Assay ◽  
False Negative Results ◽  
Mechanical Lysis

Molecular biology has been gaining more importance in parasitology. Recently, a commercial multiplex PCR assay detecting helminths was marketed: the Allplex™ GI-Helminth(I) Assay. It targets Ancylostoma spp., Ascaris spp., Enterobius vermicularis, Hymenolepis spp., Necator americanus, Strongyloides spp., Taenia spp. and Trichuris trichiura, but also the two most common microsporidia genera in human health, i.e. Enterocytozoon spp. and Encephalitozoon spp. This study aimed to evaluate and compare the Allplex™ GI-Helminth(I) Assay to classical diagnostic methods, based on a cohort of 110 stool samples positive for helminths (microscopy) or for microsporidia (PCR). Samples were stored at −80 °C until analysis by the Allplex™ GI-Helminth(I) Assay. False-negatives were re-tested with bead-beating pretreatment. Without mechanical lysis, concordance and agreement between microscopy and Allplex™ GI-Helminth(I) Assay ranged from 91% to 100% and from 0.15 to 1.00, respectively depending on the target. Concordance was perfect for Taenia spp. (n = 5) and microsporidia (n = 10). False-negative results were observed in 54% (6/13), 34% (4/11) and 20% (7/35) of cases, for hookworms, E. vermicularis and Strongyloides spp. detection, respectively. For these targets, pretreatment improved the results, but only slightly. Trichuris trichiura detection was critically low without pretreatment, as only 9% (1/11) of the samples were positive, but detection reached 91% (10/11) with bead-beating pretreatment. Mechanical lysis was also needed for Ascaris spp. and Hymenolepis spp. to reduce false-negative results from 1/8 to 1/21, respectively, to none for both. Overall, with an optimized extraction process, the Allplex™ GI-Helminth(I) Assay allows the detection of numerous parasites with roughly equivalent performance to that of microscopy, except for hookworms.

Corona Virus Disease (COVID-19) Detection in CT Images Using Synergic Deep Learning

2021 ◽  
pp. 809-819
Author(s):  
Yiwei Gao ◽  
Hongjie Hu ◽  
Huafeng Liu
Keyword(s):  
Deep Learning ◽  
Virus Disease ◽  
Ct Images ◽  
Corona Virus

Deep learning for automated glaucomatous optic neuropathy detection from ultra-widefield fundus images

2020 ◽  
pp. bjophthalmol-2020-317327
Author(s):  
Zhongwen Li ◽  
Chong Guo ◽  
Duoru Lin ◽  
Danyao Nie ◽  
Yi Zhu ◽  
...  
Keyword(s):  
Deep Learning ◽  
Optic Neuropathy ◽  
Network Architecture ◽  
Characteristic Curve ◽  
False Negative ◽  
Learning System ◽  
Glaucomatous Optic Neuropathy ◽  
Pathological Myopia ◽  
Data Set ◽  
Negative Results

Background/AimsTo develop a deep learning system for automated glaucomatous optic neuropathy (GON) detection using ultra-widefield fundus (UWF) images.MethodsWe trained, validated and externally evaluated a deep learning system for GON detection based on 22 972 UWF images from 10 590 subjects that were collected at 4 different institutions in China and Japan. The InceptionResNetV2 neural network architecture was used to develop the system. The area under the receiver operating characteristic curve (AUC), sensitivity and specificity were used to assess the performance of detecting GON by the system. The data set from the Zhongshan Ophthalmic Center (ZOC) was selected to compare the performance of the system to that of ophthalmologists who mainly conducted UWF image analysis in clinics.ResultsThe system for GON detection achieved AUCs of 0.983–0.999 with sensitivities of 97.5–98.2% and specificities of 94.3–98.4% in four independent data sets. The most common reasons for false-negative results were confounding optic disc characteristics caused by high myopia or pathological myopia (n=39 (53%)). The leading cause for false-positive results was having other fundus lesions (n=401 (96%)). The performance of the system in the ZOC data set was comparable to that of an experienced ophthalmologist (p>0.05).ConclusionOur deep learning system can accurately detect GON from UWF images in an automated fashion. It may be used as a screening tool to improve the accessibility of screening and promote the early diagnosis and management of glaucoma.

Sign in / Sign up

Export Citation Format

Share Document