scholarly journals Diagnostic Value of Deep Learning-Based CT Feature for Severe Pulmonary Infection

2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
Tinglong Huang ◽  
Xuelan Zheng ◽  
Lisui He ◽  
Zhiliang Chen
Keyword(s):  
Image Segmentation ◽  
Bacterial Infections ◽  
Pulmonary Infection ◽  
Diagnostic Value ◽  
University Hospital ◽  
Ct Image ◽  
Training Time ◽  
Dc Algorithm ◽  
Image Diagnosis ◽  
Sensitivity Specificity

The study aimed to explore the diagnostic value of computed tomography (CT) images based on cavity convolution U-Net algorithm for patients with severe pulmonary infection. A new lung CT image segmentation algorithm (U-Net+ deep convolution (DC)) was proposed based on U-Net network and compared with convolutional neural network (CNN) algorithm. Then, it was applied to CT image diagnosis of 100 patients with severe lung infection in The Second Affiliated Hospital of Fujian Medical University hospital and compared with traditional methods, and its sensitivity, specificity, and accuracy were compared. It was found that the single training time and loss of U-Net + DC algorithm were reduced by 59.4% and 9.8%, respectively, compared with CNN algorithm, while Dice increased by 3.6%. The lung contour segmented by the proposed model was smooth, which was the closest to the gold standard. Fungal infection, bacterial infection, viral infection, tuberculosis infection, and mixed infection accounted for 28%, 18%, 7%, 7%, and 40%, respectively. 36%, 38%, 26%, 17%, and 20% of the patients had ground-glass shadow, solid shadow, nodule or mass shadow, reticular or linear shadow, and hollow shadow in CT, respectively. The incidence of various CT characteristics in patients with fungal and bacterial infections was statistically significant ( P < 0.05 ). The specificity (94.32%) and accuracy (97.22%) of CT image diagnosis based on U-Net + DC algorithm were significantly higher than traditional diagnostic method (75.74% and 74.23%), and the differences were statistically significant ( P < 0.05 ). The network of the algorithm in this study demonstrated excellent image segmentation effect. The CT image based on the U-Net + DC algorithm can be used for the diagnosis of patients with severe pulmonary infection, with high diagnostic value.

scholarly journals Machine Learning Algorithms Identify Pathogen-Specific Biomarkers of Clinical and Metabolomic Characteristics in Septic Patients with Bacterial Infections

2020 ◽  
Vol 2020 ◽  
pp. 1-11
Author(s):  
Lingling Zheng ◽  
Fangqin Lin ◽  
Changxi Zhu ◽  
Guangjian Liu ◽  
Xiaohui Wu ◽  
...  
Keyword(s):  
Machine Learning ◽  
Bacterial Infections ◽  
Learning Algorithms ◽  
Intestinal Failure ◽  
Diagnostic Value ◽  
Machine Learning Algorithms ◽  
Respiratory Disorder ◽  
Clinical Indicators ◽  
Auc Value ◽  
Sensitivity Specificity

Sepsis is a high-mortality disease that is infected by bacteria, but pathogens in individual patients are difficult to diagnosis. Metabolomic changes triggered by microbial activity provide us with the possibility of accurately identifying infection. We adopted machine learning methods for training different classifiers with a clinical-metabolomic database from sepsis cases to identify the pathogen of sepsis. Records of clinical indicators and concentration of metabolites were obtained for each patient upon their arrival at the hospital. Machine learning algorithms were used in 100 patients with clear infection and corresponding 29 controls to select specific biosignatures to discriminate microorganism in septic patients. The sensitivity, specificity, and AUC value of clinical and metabolomic characteristics in predicting diagnostic outcomes were determined at admission. Our analyses demonstrate that the biosignatures selected by machine learning algorithms could have diagnostic value on the identification of infected patients and Gram-positive from Gram-negative; related AUC values were 0.94±0.054 and 0.80±0.085, respectively. Pathway and blood disease enrichment analyses of clinical and metabolomic biomarkers among infected patients showed that sepsis disease was accompanied by abnormal nitrogen metabolism, cell respiratory disorder, and renal or intestinal failure. The panel of selected clinical and metabolomic characteristics might be powerful biomarkers to discriminate patients with sepsis.

scholarly journals Comparative Study: Statistical Approach and Deep Learning Method for Automatic Segmentation Methods for Lung CT Image Segmentation

2020 ◽  
Vol 2 (4) ◽  
pp. 187-193
Author(s):  
Dr. Akey Sungheetha ◽  
Dr. Rajesh Sharma R
Keyword(s):  
Image Segmentation ◽  
Deep Learning ◽  
Comparative Study ◽  
Performance Metrics ◽  
Automatic Segmentation ◽  
Ct Image ◽  
Learning Method ◽  
Segmentation Methods ◽  
Sensitivity Specificity ◽  
Lung Ct

Recently, deep learning technique is playing important starring role for image segmentation field in medical imaging of accurate tasks. In a critical component of diagnosis, deep learning is an organized network with homogeneous areas to provide accurate results. It is proved its superior quality with statistical model automatic segmentation methods in many critical condition environments. In this research article, we focus the improved accuracy and speed of the system process compared with conservative automatic segmentation methods. Also we compared performance metrics such as accuracy, sensitivity, specificity, precision, RMSE, Precision- Recall Curve with different algorithm in deep learning method. This comparative study covers the constructing an efficient and accurate model for Lung CT image segmentation.

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 Diagnostic value of [18F]FDG PET/MRI for staging in patients with ovarian cancer

2020 ◽  
Vol 10 (1) ◽  
Author(s):  
Hideaki Tsuyoshi ◽  
Tetsuya Tsujikawa ◽  
Shizuka Yamada ◽  
Hidehiko Okazawa ◽  
Yoshio Yoshida
Keyword(s):  
Ovarian Cancer ◽  
Fdg Pet ◽  
Diagnostic Value ◽  
Mri Contrast ◽  
Diagnostic Potential ◽  
Contrast Enhanced Ct ◽  
Contrast Enhanced ◽  
Imaging Results ◽  
N Stage ◽  
Sensitivity Specificity

Abstract Purpose To evaluate the diagnostic potential of PET/MRI with 2-[18F]fluoro-2-deoxy-d-glucose ([18F]FDG) in ovarian cancer. Materials and methods Participants comprised 103 patients with suspected ovarian cancer underwent pretreatment [18F]FDG PET/MRI, contrast-enhanced CT (ceCT) and pelvic dynamic contrast-enhanced MRI (ceMRI). Diagnostic performance of [18F]FDG PET/MRI and ceMRI for assessing the characterization and the extent of the primary tumor (T stage) and [18F]FDG PET/MRI and ceCT for assessing nodal (N stage) and distant (M stage) metastases was evaluated by two experienced readers. Histopathological and follow-up imaging results were used as the gold standard. The McNemar test was employed for statistical analysis. Results Accuracy for the characterization of suspected ovarian cancer was significantly better for [18F]FDG PET/MRI (92.5%) [95% confidence interval (CI) 0.84–0.95] than for ceMRI (80.6%) (95% CI 0.72–0.83) (p < 0.05). Accuracy for T status was 96.4% (95% CI 0.96–0.96) and 92.9% (95% CI 0.93–0.93) for [18F]FDG PET/MRI and ceMRI/ceCT, respectively. Patient-based accuracies for N and M status were 100% (95% CI 0.88–1.00) and 100% (95% CI 0.88–1.00) for [18F]FDG PET/MRI and 85.2% (95% CI 0.76–0.85) and 30.8% (95% CI 0.19–0.31) for ceCT and M staging representing significant differences (p < 0.01). Lesion-based sensitivity, specificity and accuracy for N status were 78.6% (95% CI 0.57–0.91), 95.7% (95% CI 0.93–0.97) and 93.9% (95% CI 0.89–0.97) for [18F]FDG PET/MRI and 42.9% (95% CI 0.24–0.58), 96.6% (95% CI 0.94–0.98) and 90.8% (95% CI 0.87–0.94) for ceCT. Conclusions [18F]FDG PET/MRI offers better sensitivity and specificity for the characterization and M staging than ceMRI and ceCT, and diagnostic value for T and N staging equivalent to ceMRI and ceCT, suggesting that [18F]FDG PET/MRI might represent a useful diagnostic alternative to conventional imaging modalities in ovarian cancer.

scholarly journals Discovery and function exploration of microRNA-155 as a molecular biomarker for early detection of breast cancer

Breast Cancer ◽  
2021 ◽  
Author(s):  
Xuemin Liu ◽  
Qingyu Chang ◽  
Haiqiang Wang ◽  
Hairong Qian ◽  
Yikun Jiang
Keyword(s):  
Breast Cancer ◽  
Early Detection ◽  
Protein Interactions ◽  
Meta Analysis ◽  
Enrichment Analysis ◽  
Diagnostic Value ◽  
Diagnostic Biomarker ◽  
Analysis Model ◽  
Sensitivity Specificity ◽  
Function Enrichment

Abstract Background MicroRNA-155 (miR-155) may function as a diagnostic biomarker of breast cancer (BC). Nevertheless, the available evidence is controversial. Therefore, we performed this study to summarize the global predicting role of miR-155 for early detection of BC and preliminarily explore the functional roles of miR-155 in BC. Methods We first collected published studies and applied the bivariate meta-analysis model to generate the pooled diagnostic parameters of miR-155 in diagnosing BC such as sensitivity, specificity and area under curve (AUC). Then, we applied function enrichment and protein–protein interactions (PPI) analyses to explore the potential mechanisms of miR-155. Results A total of 21 studies were finally included. The results indicated that miR-155 allowed for the discrimination between BC patients and healthy controls with a sensitivity of 0.87 (95% CI 0.78–0.93), specificity of 0.82 (0.72–0.89), and AUC of 0.91 (0.88–0.93). In addition, the overall sensitivity, specificity and AUC for circulating miR-155 were 0.88 (0.76–0.95), 0.83 (0.72–0.90), and 0.92 (0.89–0.94), respectively. Function enrichment analysis revealed several vital ontologies terms and pathways associated with BC occurrence and development. Furthermore, in the PPI network, ten hub genes and two significant modules were identified to be involved in some important pathways associated with the pathogenesis of BC. Conclusions We demonstrated that miR-155 has great potential to facilitate accurate BC detection and may serve as a promising diagnostic biomarker for BC. However, well-designed cohort studies and biological experiments should be implemented to confirm the diagnostic value of miR-155 before it can be applied to routine clinical procedures.

Attention-based Full Slice Brain CT Image Diagnosis with Explanations

2021 ◽  
Author(s):  
Guanghui Fu ◽  
Jianqiang Li ◽  
Ruiqian Wang ◽  
Yue Ma ◽  
Yueda Chen
Keyword(s):  
Ct Image ◽  
Brain Ct ◽  
Image Diagnosis

scholarly journals Causes of death and comorbidities in hospitalized patients with COVID-19

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Sefer Elezkurtaj ◽  
Selina Greuel ◽  
Jana Ihlow ◽  
Edward Georg Michaelis ◽  
Philip Bischoff ◽  
...  
Keyword(s):  
Cause Of Death ◽  
Causes Of Death ◽  
Pulmonary Infection ◽  
Diffuse Alveolar Damage ◽  
University Hospital ◽  
Health Conditions ◽  
Teaching Hospitals ◽  
Increased Risk ◽  
Clinical Records ◽  
Full Body

AbstractInfection by the new corona virus strain SARS-CoV-2 and its related syndrome COVID-19 has been associated with more than two million deaths worldwide. Patients of higher age and with preexisting chronic health conditions are at an increased risk of fatal disease outcome. However, detailed information on causes of death and the contribution of pre-existing health conditions to death yet is missing, which can be reliably established by autopsy only. We performed full body autopsies on 26 patients that had died after SARS-CoV-2 infection and COVID-19 at the Charité University Hospital Berlin, Germany, or at associated teaching hospitals. We systematically evaluated causes of death and pre-existing health conditions. Additionally, clinical records and death certificates were evaluated. We report findings on causes of death and comorbidities of 26 decedents that had clinically presented with severe COVID-19. We found that septic shock and multi organ failure was the most common immediate cause of death, often due to suppurative pulmonary infection. Respiratory failure due to diffuse alveolar damage presented as immediate cause of death in fewer cases. Several comorbidities, such as hypertension, ischemic heart disease, and obesity were present in the vast majority of patients. Our findings reveal that causes of death were directly related to COVID-19 in the majority of decedents, while they appear not to be an immediate result of preexisting health conditions and comorbidities. We therefore suggest that the majority of patients had died of COVID-19 with only contributory implications of preexisting health conditions to the mechanism of death.

Dynamic adaptive residual network for liver CT image segmentation

2021 ◽  
Vol 91 ◽  
pp. 107024 ◽  
Author(s):  
Xiwang Xie ◽  
Weidong Zhang ◽  
Huadeng Wang ◽  
Lingqiao Li ◽  
Zhengyun Feng ◽  
...  
Keyword(s):  
Image Segmentation ◽  
Ct Image ◽  
Residual Network ◽  
Liver Ct

scholarly journals Evaluation of alpha-l-fucosidase for the diagnosis of hepatocellular carcinoma based on meta-analysis

2020 ◽  
Vol 0 (0) ◽  
Author(s):  
Lei Xi ◽  
Chunqing Yang
Keyword(s):  
Hepatocellular Carcinoma ◽  
Operating Characteristic ◽  
Meta Analysis ◽  
Area Under The Curve ◽  
Diagnostic Odds Ratio ◽  
Diagnostic Value ◽  
Receiver Operating Characteristic Curves ◽  
Sensitivity Specificity ◽  
Review Manager

AbstractObjectivesThe main aim of the present study was to assess the diagnostic value of alpha-l-fucosidase (AFU) for hepatocellular carcinoma (HCC).MethodsStudies that explored the diagnostic value of AFU in HCC were searched in EMBASE, SCI, and PUBMED. The sensitivity, specificity, and DOR about the accuracy of serum AFU in the diagnosis of HCC were pooled. The methodological quality of each article was evaluated with QUADAS-2 (quality assessment for studies of diagnostic accuracy 2). Receiver operating characteristic curves (ROC) analysis was performed. Statistical analysis was conducted by using Review Manager 5 and Open Meta-analyst.ResultsEighteen studies were selected in this study. The pooled estimates for AFU vs. α-fetoprotein (AFP) in the diagnosis of HCC in 18 studies were as follows: sensitivity of 0.7352 (0.6827, 0.7818) vs. 0.7501 (0.6725, 0.8144), and specificity of 0.7681 (0.6946, 0.8283) vs. 0.8208 (0.7586, 0.8697), diagnostic odds ratio (DOR) of 7.974(5.302, 11.993) vs. 13.401 (8.359, 21.483), area under the curve (AUC) of 0.7968 vs. 0.8451, respectively.ConclusionsAFU is comparable to AFP for the diagnosis of HCC.

Sign in / Sign up

Export Citation Format

Share Document