Development and validation of open-source deep neural networks for comprehensive chest x-ray reading: a retrospective, multicentre study.

BACKGROUND: Artificial intelligence (AI) systems for automated chest x-ray interpretation hold promise for standardising reporting and reducing delays in health systems with shortages of trained radiologists. Yet, there are few freely accessible AI systems trained on large datasets for practitioners to use with their own data with a view to accelerating clinical deployment of AI systems in radiology. We aimed to contribute an AI system for comprehensive chest x-ray abnormality detection. METHODS: In this retrospective cohort study, we developed open-source neural networks, X-Raydar and X-Raydar-NLP, for classifying common chest x-ray findings from images and their free-text reports. Our networks were developed using data from six UK hospitals from three National Health Service (NHS) Trusts (University Hospitals Coventry and Warwickshire NHS Trust, University Hospitals Birmingham NHS Foundation Trust, and University Hospitals Leicester NHS Trust) collectively contributing 2 513 546 chest x-ray studies taken from a 13-year period (2006-19), which yielded 1 940 508 usable free-text radiological reports written by the contemporary assessing radiologist (collectively referred to as the "historic reporters") and 1 896 034 frontal images. Chest x-rays were labelled using a taxonomy of 37 findings by a custom-trained natural language processing (NLP) algorithm, X-Raydar-NLP, from the original free-text reports. X-Raydar-NLP was trained on 23 230 manually annotated reports and tested on 4551 reports from all hospitals. 1 694 921 labelled images from the training set and 89 238 from the validation set were then used to train a multi-label image classifier. Our algorithms were evaluated on three retrospective datasets: a set of exams sampled randomly from the full NHS dataset reported during clinical practice and annotated using NLP (n=103 328); a consensus set sampled from all six hospitals annotated by three expert radiologists (two independent annotators for each image and a third consultant to facilitate disagreement resolution) under research conditions (n=1427); and an independent dataset, MIMIC-CXR, consisting of NLP-annotated exams (n=252 374). FINDINGS: X-Raydar achieved a mean AUC of 0·919 (SD 0·039) on the auto-labelled set, 0·864 (0·102) on the consensus set, and 0·842 (0·074) on the MIMIC-CXR test, demonstrating similar performance to the historic clinical radiologist reporters, as assessed on the consensus set, for multiple clinically important findings, including pneumothorax, parenchymal opacification, and parenchymal mass or nodules. On the consensus set, X-Raydar outperformed historical reporter balanced accuracy with significance on 27 of 37 findings, was non-inferior on nine, and inferior on one finding, resulting in an average improvement of 13·3% (SD 13·1) to 0·763 (0·110), including a mean 5·6% (13·2) improvement in critical findings to 0·826 (0·119). INTERPRETATION: Our study shows that automated classification of chest x-rays under a comprehensive taxonomy can achieve performance levels similar to those of historical reporters and exhibit robust generalisation to external data. The open-sourced neural networks can serve as foundation models for further research and are freely available to the research community. FUNDING: Wellcome Trust.

Long-term survival in elderly patients with a do-not-intubate order treated with noninvasive mechanical ventilation.

BACKGROUND: Noninvasive mechanical ventilation (NIMV) is an effective tool in treating patients with acute respiratory failure (ARF), since it reduces both the need for endotracheal intubation and the mortality in comparison with nonventilated patients. A particular issue is represented by the outcome of NIMV in patients referred to the emergency department for ARF and with a do-not-intubate (DNI) status because of advanced age or excessively critical conditions. This study evaluated long-term survival in a group of elderly patients with acute hypercapnic ARF who had a DNI order and who were successfully treated by NIMV. METHODS: The population consisted of 54 patients with a favorable outcome after NIMV for ARF. They were followed up for 3 years by regular control visits, with at least one visit every 4 months, or as needed according to the patient's condition. Of these, 31 continued NIMV at home and 23 were on long-term oxygen therapy (LTOT) alone. RESULTS: A total of 16 of the 52 patients had not survived at the 1-year follow-up, and another eight patients died during the 3-year observation, with an overall mortality rate of 30.8% after 1 year and 46.2% after 3 years. Comparing patients who continued NIMV at home with those who were on LTOT alone, 9 of the 29 patients on home NIMV died (6 after 1 year and 3 after 3 years) and 15 of the 23 patients on LTOT alone died (10 after 1 year and 5 after 3 years). CONCLUSION: These results show that elderly patients with ARF successfully treated by NIMV following a DNI order have a satisfactory long-term survival.

Effect of noninvasive mechanical ventilation in elderly patients with hypercapnic acute-on-chronic respiratory failure and a do-not-intubate order.

Noninvasive mechanical ventilation (NIMV) is effective in the treatment of patients with acute respiratory failure (ARF). It proved to reduce the need of endotracheal intubation (ETI), the incidence of ETI-associated pneumonia, and mortality compared to nonventilated patients. A particular aspect concerns the outcome of NIMV in patients referring to an emergency room (ER) for ARF, and with a do-not-intubate (DNI) status due to advanced age or critical conditions. The aim of our study is to assess the outcome of NIMV in a group of elderly patients with acute hypercapnic ARF who had a DNI status. An overall number of 62 subjects (30 males, 32 females, mean age 81 +/- 4.8 years, range 79-91 years) referred to our semi-intensive respiratory department were enrolled in the study. The underlying diseases were severe chronic obstructive pulmonary disease (COPD) in 50/62 subjects, restrictive thoracic disorders in 7/62 subjects, and multiorgan failure in 5/62 subjects. Fifty-four/62 patients were successfully treated with NIMV while 2/62 did not respond to NIMV and were therefore submitted to ETI (one survived). Among NIMV-treated patients, death occurred in 6 patients after a mean of 9.9 days; the overall rate of NIMV failure was 12.9%. Negative prognostic factors for NIMV response proved to be: an older age, a low Glasgow Coma Score, a high APACHE score at admission, a high PaCO2 after 12 hours and a low pH both after 1 and 12 hours of NIMV. We conclude that elderly patients with acute hypercapnic ARF with a DNI status can be successfully treated by NIMV.