Advancing COVID-19 Diagnosis with Privacy-Preserving Collaboration in Artificial Intelligence

@article{Bai2021AdvancingCD,
  title={Advancing COVID-19 Diagnosis with Privacy-Preserving Collaboration in Artificial Intelligence},
  author={Xiang Bai and Hanchen Wang and Liya Ma and Yongchao Xu and Jiefeng Gan and Ziwei Fan and Fan Yang and Ke Ma and Jiehua Yang and Song Bai and Chang Shu and Xinyu Zou and Renhao Huang and Changzheng Zhang and Xiaowu Liu and Dandan Tu and Chuou Xu and Wenqing Zhang and Xi Le Wang and Anguo Chen and Yu Zeng and Dehua Yang and Ming-Wei Wang and Nagaraj Setty Holalkere and Neil J. Halin and Ihab R. Kamel and Jia Wu and Xue-Hua Peng and Xiang Wang and Jianbo Shao and Pattanasak Mongkolwat and Jianjun Zhang and Weiyang Liu and Michael Roberts and Zhongzhao Teng and Lucian Beer and Lorena E. Sanchez and Evis Sala and D. Rubin and Adrian Weller and Joan Lasenby and Chuangsheng Zheng and Jianming Wang and Zhen Li and Carola-Bibiane Schonlieb and Tian Xia},
  journal={ArXiv},
  year={2021}
}
Artificial intelligence (AI) provides a promising substitution for streamlining COVID-19 diagnoses. However, concerns surrounding security and trustworthiness impede the collection of large-scale representative medical data, posing a considerable challenge for training a well-generalised model in clinical practices. To address this, we launch the Unified CT-COVID AI Diagnostic Initiative (UCADI), where the AI model can be distributedly trained and independently executed at each host institution… 
View on PubMed
nature.com
1 Citations
Background Citations
1

One Citation

Citation Type

Citation Type

Machine Learning Generalizability Across Healthcare Settings: Insights from multi-site COVID-19 screening
TLDR
Three methods for translating ready-made models for adoption in new settings by readingjusting the decision threshold on the output of a ready- made model using site-specific data via transfer learning are introduced and it is demonstrated that site- specific customization improves predictive performance when compared to otherready-made approaches.

References

SHOWING 1-10 OF 28 REFERENCES
Swarm Learning for decentralized and confidential clinical machine learning
TLDR
Swarm Learning is introduced—a decentralized machine-learning approach that unites edge computing, blockchain-based peer-to-peer networking and coordination while maintaining confidentiality without the need for a central coordinator, thereby going beyond federated learning.
End-to-end privacy preserving deep learning on multi-institutional medical imaging
TLDR
PriMIA (Privacy-preserving Medical Image Analysis), a free, open-source software framework for differentially private, securely aggregated federated learning and encrypted inference on medical imaging data, is presented.
AI for radiographic COVID-19 detection selects shortcuts over signal
TLDR
It is demonstrated that explainable AI should be seen as a prerequisite to clinical deployment of machine-learning healthcare models, because the undesired ‘shortcuts’ learned by AI systems may not impair performance in new hospitals.
Artificial intelligence for the detection of COVID-19 pneumonia on chest CT using multinational datasets
TLDR
It is shown that a series of deep learning algorithms, trained in a diverse multinational cohort of 1280 patients to localize parietal pleura/lung parenchyma followed by classification of COVID-19 pneumonia, can achieve up to 90.8% accuracy, with 84% sensitivity and 93% specificity.
Using Artificial Intelligence to Detect COVID-19 and Community-acquired Pneumonia Based on Pulmonary CT: Evaluation of the Diagnostic Accuracy
TLDR
A deep learning model was developed to extract visual features from volumetric chest CT scans for the detection of coronavirus 2019 and differentiate it from community-acquired pneumonia and other lung conditions.
Artificial intelligence–enabled rapid diagnosis of patients with COVID-19
TLDR
Artificial intelligence algorithms integrating chest CT findings with clinical symptoms, exposure history and laboratory testing to rapidly diagnose patients who are positive for COVID-19 with similar accuracy as compared to a senior radiologist.
The Algorithmic Foundations of Differential Privacy
TLDR
The preponderance of this monograph is devoted to fundamental techniques for achieving differential privacy, and application of these techniques in creative combinations, using the query-release problem as an ongoing example.
Performance of radiologists in differentiating COVID-19 from viral pneumonia on chest CT
TLDR
Radiologists in China and the United States distinguished COVID-19 from viral pneumonia on chest CT with high specificity but moderate sensitivity.
Data augmentation using generative adversarial networks (CycleGAN) to improve generalizability in CT segmentation tasks
TLDR
It is shown that in several CT segmentation tasks performance is improved significantly, especially in out-of-distribution (noncontrast CT) data, which will be valuable to medical imaging researchers to reduce manual segmentation effort and cost in CT imaging.
The role of chest radiography in confirming covid-19 pneumonia
TLDR
Advice is offered to non-radiologists on how to look for changes on chest radiograph that may be suggestive of covid-19 pneumonia, as prompt review and report from an onsite or remote radiologist is not always available.
...
1
2
3
...