scholarly journals Fibrosis-Net: A Tailored Deep Convolutional Neural Network Design For Prediction of Pulmonary Fibrosis Progression From Chest CT Images

2021 ◽  
Author(s):  
Alexander Wong ◽  
Jack Lu ◽  
Adam Dorfman ◽  
Paul McInnis ◽  
Mahmoud Famouri ◽  
...  
Keyword(s):  
Decision Making ◽  
Pulmonary Fibrosis ◽  
Network Design ◽  
Ct Images ◽  
Chest Ct ◽  
Correct Decision ◽  
Fibrosis Progression ◽  
Likelihood Score ◽  
Log Likelihood ◽  
Performance Validation

Abstract Background: Pulmonary fibrosis is a devastating chronic lung disease that causes irreparable lung tissue scarring and damage, resulting in progressive loss in lung capacity and has no known cure. A critical step in the treatment and management of pulmonary fibrosis is the assessment of lung function decline, with computed tomography (CT) imaging being a particularly effective method for determining the extent of lung damage caused by pulmonary fibrosis. Motivated by this, we introduce Fibrosis-Net, a deep convolutional neural network design tailored for the prediction of pulmonary fibrosis progression from chest CT images. More specifically, machine-driven design exploration was leveraged to determine a strong architectural design for CT lung analysis, upon which we build a customized network design tailored for predicting forced vital capacity (FVC) based on a patient's CT scan, initial spirometry measurement, and clinical metadata. Finally, we leverage an explainability-driven performance validation strategy to study the decision-making behaviour of Fibrosis-Net as to verify that predictions are based on relevant visual indicators in CT images.Results: Experiments using a patient cohort from the OSIC Pulmonary Fibrosis Progression Challenge showed that the proposed Fibrosis-Net is able to achieve a significantly higher modified Laplace Log Likelihood score than the winning solutions on the challenge. Furthermore, explainability-driven performance validation demonstrated that the proposed Fibrosis-Net exhibits correct decision-making behaviour by leveraging clinically-relevant visual indicators in CT images when making predictions on pulmonary fibrosis progress. Conclusion: Fibrosis-Net is able to achieve a significantly higher modified Laplace Log Likelihood score than the winning solutions on the OSIC Pulmonary Fibrosis Progression Challenge, and has been shown to exhibit correct decision-making behaviour when making predictions. Fibrosis-Net is available to the general public in an open-source and open access manner as part of the OpenMedAI initiative. While Fibrosis-Net is not yet a production-ready clinical assessment solution, we hope that its release will encourage researchers, clinicians, and citizen data scientists alike to leverage and build upon it.

scholarly journals Clinical characteristics of imported and second-generation COVID-19 cases outside Wuhan, China: A multicenter retrospective study

2020 ◽  
Author(s):  
Puyu Shi ◽  
Guoxia Ren ◽  
Jun Yang ◽  
Zhiqiang Li ◽  
Shujiao Deng ◽  
...  
Keyword(s):  
Pulmonary Fibrosis ◽  
Hospital Admission ◽  
Symptom Onset ◽  
Lymphocyte Count ◽  
Clinical Characteristics ◽  
Second Generation ◽  
Discharge Rate ◽  
Ct Images ◽  
Chest Ct ◽  
Multicenter Cohort Study

AbstractBackgroundThe mortality of COVID-19 differs between countries and regions. By now, reports on COVID-19 are largely focused on first-generation cases. This study aimed to clarify the clinical characteristics of imported and second-generation cases.MethodsThis retrospective, multicenter cohort study included 134 confirmed COVID-19 cases from 9 cities outside Wuhan. Epidemiological, clinical and outcome data were extracted from medical records and were compared between severe and non-severe cases. We further profiled the dynamic laboratory findings of some patients.Results34.3% of the 134 patients were severe cases, and 11.2% had complications. As of March 7, 2020, 91.8% patients were discharged and one patient (0.7%) died. The median age was 46 years. The median interval from symptom onset to hospital admission was 4.5 (IQR 3-7) days. The median lymphocyte count was 1.1×109/L. Age, lymphocyte count, CRP, ESR, DBIL, LDH, HBDH showed difference between severe and no-severe cases (all P<0.05). Baseline lymphocyte count was higher in the survived patients than in the non-survivor case, and it increased as the condition improved, but declined sharply when death occurred. The IL-6 level displayed a downtrend in survivors, but rose very high in the death case. Pulmonary fibrosis was found on later chest CT images in 51.5% of the pneumonia cases.ConclusionImported and second-generation cases outside Wuhan had a better prognosis than initial cases in Wuhan. Lymphocyte count and IL-6 level could be used for evaluating prognosis. Pulmonary fibrosis as the sequelae of COVID-19 should be taken into account.SummaryImported and second-generation cases manifested less complications, lower fatality, and higher discharge rate than initial cases, which may be related to the shorter interval from symptom onset to hospital admission, younger age, and higher lymphocyte count of the imported and second-generation patients. Lymphocyte count and IL-6 level could be used as indicators for evaluating prognosis. Pulmonary fibrosis was found in later chest CT images in more than half of the pneumonia cases and should be taken into account.

An investigation of book reading habits levels and correct decision-making styles in terms of some variables: A research on coach and referees

2020 ◽  
Vol 8 (4) ◽  
pp. 849-865
Author(s):  
Mihriay Musa ◽  
Keyword(s):  
Decision Making ◽  
Quantitative Research ◽  
Simple Random Sampling ◽  
T Test ◽  
Book Reading ◽  
Correct Decision ◽  
Reading Habits ◽  
Significance Level ◽  
Decision Making Styles ◽  
The Universe

In this study, it was aimed to examine the reading habits levels and making the correct decision styles of basketball, handball, volleyball, and football coaches and referees in terms of some variables, the research was carried out with the general survey model, one of the quantitative research designs, the active coaches and referees of basketball, football, volleyball, and handball in İzmir, Denizli and Uşak provinces constituted the universe of the study, the sample of the study, on the other hand, consisted of 98 participants, 52 of whom were coaches and 46 were referees, determined by the simple random sampling method, one sample t-test at a 0.05 significance level was conducted to determine whether the sample represented the universe equally and homogeneously. Melbourne decision making scale I-II, and book reading habits scale were used to collect data in the study. Since the data are suitable for normal distribution, the t-test in comparing the pairwise means; parametric tests such as one-way ANOVA tests were used at 0.05 significance level in comparing the mean scores of more than two groups. In terms of education levels, it has been observed that female coaches and referees studying at faculties of sports sciences have higher levels of reading habit, love of reading, and being influenced by books. In addition, it was determined that individuals who trust and respect the decisions of their families have higher reading habits and correct decision-making styles and do not panic during the decision-making process.

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.

Long-Acting Human Serum Albumin-Thioredoxin Fusion Protein Suppresses Bleomycin-Induced Pulmonary Fibrosis Progression

2013 ◽  
Vol 345 (2) ◽  
pp. 271-283 ◽  
Author(s):  
Ryota Tanaka ◽  
Hiroshi Watanabe ◽  
Azusa Kodama ◽  
Victor Tuan Giam Chuang ◽  
Yu Ishima ◽  
...  
Keyword(s):  
Human Serum Albumin ◽  
Pulmonary Fibrosis ◽  
Fusion Protein ◽  
Serum Albumin ◽  
Human Serum ◽  
Fibrosis Progression ◽  
Long Acting

Normative Weighing and Legal Guidance of Conduct

2012 ◽  
Vol 25 (2) ◽  
pp. 359-391
Author(s):  
Noam Gur
Keyword(s):  
Decision Making ◽  
Practical Reason ◽  
Social Order ◽  
Daily Activity ◽  
Empirical Work ◽  
Pivotal Role ◽  
Reasons For Action ◽  
Correct Decision ◽  
Normative Force ◽  
Frequent Presence

Contemporary legal philosophers commonly understand the normative force of law in terms of practical reason. They sharply disagree, however, on how exactly it translates into practical reason. Notably, some have argued that the directives of an authority that meets certain prerequisites of legitimacy generate reasons for action that exclude some otherwise applicable reasons, while others have insisted that such directives can only give rise to reasons that compete with opposing ones in terms of their weight (an approach I will call the weighing model). Does the weighing model provide a normative framework within which law could adequately facilitate correct decision-making? At first glance, the answer appears to be ‘yes’: there seems to be nothing about law-following values—such as coordination reasons, the desirability of social order, deferential expertise, etc.—which prevents them from being factored into our decision-making in terms of normative weight that tips the balance in favor of compliance with law inasmuch as it is worthwhile to comply with it. This impression, however, turns out to be incorrect when, drawing on a body of empirical work in psychology, I observe that many of the practical difficulties law typically addresses are difficulties that have part of their root in biases to which we are systematically susceptible in the settings of our daily activity. I argue that the frequent presence of those biases in contexts of activity which law regulates, and the pivotal role law has in counteracting them, emphatically militate against the weighing model and call for its rejection.

Sign in / Sign up

Export Citation Format

Share Document