Tuberculosis is a public health problem worldwide, including in the United States-particularly among immunocompromised patients and other high-risk groups. Tuberculosis manifests in active and latent forms. Active disease can occur as primary tuberculosis, developing shortly after infection, or postprimary tuberculosis, developing after a long period of latent infection. Primary tuberculosis occurs most commonly in children and immunocompromised patients, who present with lymphadenopathy, pulmonary consolidation, and pleural effusion. Postprimary tuberculosis may manifest with cavities, consolidations, and centrilobular nodules. Miliary tuberculosis refers to hematogenously disseminated disease that is more commonly seen in immunocompromised patients, who present with miliary lung nodules and multiorgan involvement. The principal means of testing for active tuberculosis is sputum analysis, including smear, culture, and nucleic acid amplification testing. Imaging findings, particularly the presence of cavitation, can affect treatment decisions, such as the duration of therapy. Latent tuberculosis is an asymptomatic infection that can lead to postprimary tuberculosis in the future. Patients who are suspected of having latent tuberculosis may undergo targeted testing with a tuberculin skin test or interferon-γ release assay. Chest radiographs are used to stratify for risk and to assess for asymptomatic active disease. Sequelae of previous tuberculosis that is now inactive manifest characteristically as fibronodular opacities in the apical and upper lung zones. Stability of radiographic findings for 6 months distinguishes inactive from active disease. Nontuberculous mycobacterial disease can sometimes mimic the findings of active tuberculosis, and laboratory confirmation is required to make the distinction. Familiarity with the imaging, clinical, and laboratory features of tuberculosis is important for diagnosis and management. RSNA, 2017.
There is considerable overlap in the imaging appearance of viral and bacterial respiratory infections. However, some characteristic differences can be seen, especially with RSV and adenovirus infections.
Objectives
We aimed to develop deep learning models using longitudinal chest X-rays (CXRs) and clinical data to predict in-hospital mortality of COVID-19 patients in the intensive care unit (ICU).
Methods
Six hundred fifty-four patients (212 deceased, 442 alive, 5645 total CXRs) were identified across two institutions. Imaging and clinical data from one institution were used to train five longitudinal transformer-based networks applying five-fold cross-validation. The models were tested on data from the other institution, and pairwise comparisons were used to determine the best-performing models.
Results
A higher proportion of deceased patients had elevated white blood cell count, decreased absolute lymphocyte count, elevated creatine concentration, and incidence of cardiovascular and chronic kidney disease. A model based on pre-ICU CXRs achieved an AUC of 0.632 and an accuracy of 0.593, and a model based on ICU CXRs achieved an AUC of 0.697 and an accuracy of 0.657. A model based on all longitudinal CXRs (both pre-ICU and ICU) achieved an AUC of 0.702 and an accuracy of 0.694. A model based on clinical data alone achieved an AUC of 0.653 and an accuracy of 0.657. The addition of longitudinal imaging to clinical data in a combined model significantly improved performance, reaching an AUC of 0.727 (
p
= 0.039) and an accuracy of 0.732.
Conclusions
The addition of longitudinal CXRs to clinical data significantly improves mortality prediction with deep learning for COVID-19 patients in the ICU.
Key Points
•
Deep learning was used to predict mortality in COVID-19 ICU patients.
•
Serial radiographs and clinical data were used.
•
The models could inform clinical decision-making and resource allocation.
Supplementary Information
The online version contains supplementary material available at 10.1007/s00330-022-08588-8.
There is relatively poor interobserver agreement for subsegmental and/or small pulmonary artery defects, especially in CT pulmonary angiograms degraded by technical artifacts. These factors can lead to an increased frequency of inaccurate interpretation or indeterminate diagnosis of subsegmental and/or small defects. Caution is indicated in interpreting the significance of small vascular defects in CT pulmonary angiograms.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.