Publisher Correction: Analysis of the Human Protein Atlas Image Classification competition.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Author Correction: Analysis of the Human Protein Atlas Image Classification competition.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Publisher Correction: Analysis of the Human Protein Atlas Image Classification competition.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Multi-center validation of an artificial intelligence system for detection of COVID-19 on chest radiographs in symptomatic patients.

OBJECTIVES: While chest radiograph (CXR) is the first-line imaging investigation in patients with respiratory symptoms, differentiating COVID-19 from other respiratory infections on CXR remains challenging. We developed and validated an AI system for COVID-19 detection on presenting CXR. METHODS: A deep learning model (RadGenX), trained on 168,850 CXRs, was validated on a large international test set of presenting CXRs of symptomatic patients from 9 study sites (US, Italy, and Hong Kong SAR) and 2 public datasets from the US and Europe. Performance was measured by area under the receiver operator characteristic curve (AUC). Bootstrapped simulations were performed to assess performance across a range of potential COVID-19 disease prevalence values (3.33 to 33.3%). Comparison against international radiologists was performed on an independent test set of 852 cases. RESULTS: RadGenX achieved an AUC of 0.89 on 4-fold cross-validation and an AUC of 0.79 (95%CI 0.78-0.80) on an independent test cohort of 5,894 patients. Delong's test showed statistical differences in model performance across patients from different regions (p < 0.01), disease severity (p < 0.001), gender (p < 0.001), and age (p = 0.03). Prevalence simulations showed the negative predictive value increases from 86.1% at 33.3% prevalence, to greater than 98.5% at any prevalence below 4.5%. Compared with radiologists, McNemar's test showed the model has higher sensitivity (p < 0.001) but lower specificity (p < 0.001). CONCLUSION: An AI model that predicts COVID-19 infection on CXR in symptomatic patients was validated on a large international cohort providing valuable context on testing and performance expectations for AI systems that perform COVID-19 prediction on CXR. KEY POINTS: • An AI model developed using CXRs to detect COVID-19 was validated in a large multi-center cohort of 5,894 patients from 9 prospectively recruited sites and 2 public datasets. • Differences in AI model performance were seen across region, disease severity, gender, and age. • Prevalence simulations on the international test set demonstrate the model's NPV is greater than 98.5% at any prevalence below 4.5%.

Analysis of the Human Protein Atlas Image Classification competition.

Pinpointing subcellular protein localizations from microscopy images is easy to the trained eye, but challenging to automate. Based on the Human Protein Atlas image collection, we held a competition to identify deep learning solutions to solve this task. Challenges included training on highly imbalanced classes and predicting multiple labels per image. Over 3 months, 2,172 teams participated. Despite convergence on popular networks and training techniques, there was considerable variety among the solutions. Participants applied strategies for modifying neural networks and loss functions, augmenting data and using pretrained networks. The winning models far outperformed our previous effort at multi-label classification of protein localization patterns by ~20%. These models can be used as classifiers to annotate new images, feature extractors to measure pattern similarity or pretrained networks for a wide range of biological applications.

Detection of COVID-19 Using Deep Learning Algorithms on Chest Radiographs.

PURPOSE: To evaluate the performance of a deep learning (DL) algorithm for the detection of COVID-19 on chest radiographs (CXR). MATERIALS AND METHODS: In this retrospective study, a DL model was trained on 112,120 CXR images with 14 labeled classifiers (ChestX-ray14) and fine-tuned using initial CXR on hospital admission of 509 patients, who had undergone COVID-19 reverse transcriptase-polymerase chain reaction (RT-PCR). The test set consisted of a CXR on presentation of 248 individuals suspected of COVID-19 pneumonia between February 16 and March 3, 2020 from 4 centers (72 RT-PCR positives and 176 RT-PCR negatives). The CXR were independently reviewed by 3 radiologists and using the DL algorithm. Diagnostic performance was compared with radiologists' performance and was assessed by area under the receiver operating characteristics (AUC). RESULTS: The median age of the subjects in the test set was 61 (interquartile range: 39 to 79) years (51% male). The DL algorithm achieved an AUC of 0.81, sensitivity of 0.85, and specificity of 0.72 in detecting COVID-19 using RT-PCR as the reference standard. On subgroup analyses, the model achieved an AUC of 0.79, sensitivity of 0.80, and specificity of 0.74 in detecting COVID-19 in patients presented with fever or respiratory systems and an AUC of 0.87, sensitivity of 0.85, and specificity of 0.81 in distinguishing COVID-19 from other forms of pneumonia. The algorithm significantly outperforms human readers (P<0.001 using DeLong test) with higher sensitivity (P=0.01 using McNemar test). CONCLUSIONS: A DL algorithm (COV19NET) for the detection of COVID-19 on chest radiographs can potentially be an effective tool in triaging patients, particularly in resource-stretched health-care systems.