Clinical, Cultural, Computational, and Regulatory Considerations to Deploy AI in Radiology: Perspectives of RSNA and MICCAI Experts

At the beginning of the artificial intelligence (AI)/machine learning (ML) era, the expectations are high, and experts foresee that AI/ML shows potential for diagnosing, managing and treating a wide variety of medical conditions. However, the obstacles for implementation of AI/ML in daily clinical practice are numerous, especially regarding the regulation of these technologies. Therefore, we provide an insight into the currently available AI/ML-based medical devices and algorithms that have been approved by the US Food & Drugs Administration (FDA). We aimed to raise awareness of the importance of regulatory bodies, clearly stating whether a medical device is AI/ML based or not. Cross-checking and validating all approvals, we identified 64 AI/ML based, FDA approved medical devices and algorithms. Out of those, only 29 (45%) mentioned any AI/ML-related expressions in the official FDA announcement. The majority (85.9%) was approved by the FDA with a 510(k) clearance, while 8 (12.5%) received de novo pathway clearance and one (1.6%) premarket approval (PMA) clearance. Most of these technologies, notably 30 (46.9%), 16 (25.0%), and 10 (15.6%) were developed for the fields of Radiology, Cardiology and Internal Medicine/General Practice respectively. We have launched the first comprehensive and open access database of strictly AI/ML-based medical technologies that have been approved by the FDA. The database will be constantly updated.

Artificial intelligence (AI) continues to garner substantial interest in medical imaging. The potential applications are vast and include the entirety of the medical imaging life cycle from image creation to diagnosis to outcome prediction. The chief obstacles to development and clinical implementation of AI algorithms include availability of sufficiently large, curated, and representative training data that includes expert labeling (eg, annotations). Current supervised AI methods require a curation process for data to optimally train, validate, and test algorithms. Currently, most research groups and industry have limited data access based on small sample sizes from small geographic areas. In addition, the preparation of data is a costly and time-intensive process, the results of which are algorithms with limited utility and poor generalization. In this article, the authors describe fundamental steps for preparing medical imaging data in AI algorithm development, explain current limitations to data curation, and explore new approaches to address the problem of data availability. Supervised artificial intelligence (AI) methods for evaluation of medical images require a curation process for data to optimally train, validate, and test algorithms. The chief obstacles to development and clinical implementation of AI algorithms include availability of sufficiently large, curated, and representative training data that includes expert labeling (eg, annotations).

Abstract Background COVID-19 and pneumonia of other etiology share similar CT characteristics, contributing to the challenges in differentiating them with high accuracy. Purpose To establish and evaluate an artificial intelligence (AI) system in differentiating COVID-19 and other pneumonia on chest CT and assess radiologist performance without and with AI assistance. Methods 521 patients with positive RT-PCR for COVID-19 and abnormal chest CT findings were retrospectively identified from ten hospitals from January 2020 to April 2020. 665 patients with non-COVID-19 pneumonia and definite evidence of pneumonia on chest CT were retrospectively selected from three hospitals between 2017 and 2019. To classify COVID-19 versus other pneumonia for each patient, abnormal CT slices were input into the EfficientNet B4 deep neural network architecture after lung segmentation, followed by two-layer fully-connected neural network to pool slices together. Our final cohort of 1,186 patients (132,583 CT slices) was divided into training, validation and test sets in a 7:2:1 and equal ratio. Independent testing was performed by evaluating model performance on separate hospitals. Studies were blindly reviewed by six radiologists without and then with AI assistance. Results Our final model achieved a test accuracy of 96% (95% CI: 90-98%), sensitivity 95% (95% CI: 83-100%) and specificity of 96% (95% CI: 88-99%) with Receiver Operating Characteristic (ROC) AUC of 0.95 and Precision-Recall (PR) AUC of 0.90. On independent testing, our model achieved an accuracy of 87% (95% CI: 82-90%), sensitivity of 89% (95% CI: 81-94%) and specificity of 86% (95% CI: 80-90%) with ROC AUC of 0.90 and PR AUC of 0.87. Assisted by the models’ probabilities, the radiologists achieved a higher average test accuracy (90% vs. 85%, Δ=5, p<0.001), sensitivity (88% vs. 79%, Δ=9, p<0.001) and specificity (91% vs. 88%, Δ=3, p=0.001). Conclusion AI assistance improved radiologists' performance in distinguishing COVID-19 from non-COVID-19 pneumonia on chest CT.