A retrospective study conducted in the United Arab Emirates evaluated the post-deployment performance of qXR (Qure.ai) for classifying chest X-rays (CXRs) as normal or abnormal in a high-volume visa screening setting. Conducted between January 2021 and June 2022 across 33 screening centers, the study analyzed 1,309,443 CXRs (median age 35 years; 78.7% male) to assess the algorithm’s agreement with radiologists and impact on workflow efficiency.
qXR achieved a high Negative Predictive Value (NPV) of 99.92% and a Positive Predictive Value (PPV) of 5.06%, influenced by the low prevalence of abnormal findings and the algorithm’s tendency to flag some clinically irrelevant abnormalities. The overall agreement with radiologists was 72.90%.
A survey of 20 healthcare professionals, including 17 radiologists, found that 88.2% reported reduced turnaround times, and 82% noted improved diagnostic accuracy with AI integration. Radiologists highlighted workload reduction and suggested improvements in lesion detection and customization.
The study concluded that qXR is effective in high-volume, post-deployment settings, enhancing efficiency while maintaining high diagnostic accuracy for normal CXRs. Limitations included reliance on a single-radiologist ground truth, exclusion of 93,730 CXRs due to image quality issues, and lack of clinical or microbiological confirmation of abnormal findings. Refinements are recommended to improve specificity and reduce clinically irrelevant flags.
Read full study
Post-deployment performance of a deep learning algorithm for normal and abnormal chest X-ray classification: A study at visa screening centers in the United Arab Emirates.
European Journal of Radiology Open, 2024
Abstract
Background
Chest radiographs (CXRs) are widely used to screen for infectious diseases like tuberculosis and COVID-19 among migrants. At such high-volume settings, manual CXR reporting is challenging and integrating artificial intelligence (AI) algorithms into the workflow help to rule out normal findings in minutes, allowing radiologists to focus on abnormal cases.
Methods
In this post-deployment study, all the CXRs acquired during the visa screening process across 33 centers in United Arab Emirates from January 2021 to June 2022 (18 months) were included. The qXR v2.1 chest X-ray interpretation software was used to classify the scans into normal and abnormal, and its agreement against radiologist was evaluated. Additionally, a digital survey was conducted among 20 healthcare professionals with prior AI experience to understand real-world implementation challenges and impact.
Results
The analysis of 1309,443 CXRs from 1309,431 patients (median age: 35 years; IQR [29-42]; 1030,071 males [78.7 %]) in this study revealed a Negative Predictive Value (NPV) of 99.92 % (95 % CI: 99.92, 99.93), Positive Predictive Value (PPV) of 5.06 % (95 % CI: 4.99, 5.13) and overall percent agreement of the AI with radiologists of 72.90 % (95 % CI: 72.82, 72.98). In the survey, majority (88.2 %) of the radiologists agreed to turnaround time reduction after AI integration, while 82 % suggested that the AI improved their diagnostic accuracy.
Discussion
In contrast with the existing studies, this research uses a substantially large data. A high NPV and satisfactory agreement with human readers indicate that AI can reliably identify normal CXRs, making it suitable for routine applications.