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.

Development and Validation of Noninvasive MRI-Based Signature for Preoperative Prediction of Early Recurrence in Perihilar Cholangiocarcinoma.

BACKGROUND: Cholangiocarcinoma is a type of hepatobiliary tumor. For perihilar cholangiocarcinoma (pCCA), patients who experience early recurrence (ER) have a poor prognosis. Preoperative accurate prediction of postoperative ER can avoid unnecessary operation; however, prediction is challenging. PURPOSE: To develop a novel signature based on clinical and/or MRI radiomics features of pCCA to preoperatively predict ER. STUDY TYPE: Retrospective. POPULATION: One hundred eighty-four patients (median age, 61.0 years; interquartile range: 53.0-66.8 years) including 115 men and 69 women. FIELD STRENGTH/SEQUENCE: A 1.5 T; volumetric interpolated breath-hold examination (VIBE) sequence. ASSESSMENT: The models were developed from the training set (128 patients) and validated in a separate testing set (56 patients). The contrast-enhanced arterial and portal vein phase MR images of hepatobiliary system were used for extracting radiomics features. The correlation analysis, least absolute shrinkage and selection operator (LASSO) logistic regression (LR), backward stepwise LR were mainly used for radiomics feature selection and modeling (Modelradiomic ). The univariate and multivariate backward stepwise LR were used for preoperative clinical predictors selection and modeling (Modelclinic ). The radiomics and preoperative clinical predictors were combined by multivariate LR method to construct clinic-radiomics nomogram (Modelcombine ). STATISTICAL TESTS: Chi-squared (χ2 ) test or Fisher's exact test, Mann-Whitney U-test or t-test, Delong test. Two tailed P < 0.05 was considered statistically significant. RESULTS: Based on the comparison of area under the curves (AUC) using Delong test, Modelclinic and Modelcombine had significantly better performance than Modelradiomic and tumor-node-metastasis (TNM) system in training set. In the testing set, both Modelclinic and Modelcombine had significantly better performance than TNM system, whereas only Modelcombine was significantly superior to Modelradiomic . However, the AUC values were not significantly different between Modelclinic and Modelcombine (P = 0.156 for training set and P = 0.439 for testing set). DATA CONCLUSION: A noninvasive model combining the MRI-based radiomics signature and clinical variables is potential to preoperatively predict ER for pCCA. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY STAGE: 4.