Key result
A novel batch-wise sequential learning scheme performed best among decentralized methods, resulting in a minor loss of classification performance (-0.036 AUROC) compared to a centralized baseline model.
Why the study?
Applying AI and machine learning in medicine can be challenging due to privacy concerns and strict legal regulations, which could be addressed by methods that centralize knowledge instead of data.
Do decentralized machine learning algorithms provide comparable 12-lead ECG classification performance to centralized machine learning while preserving data privacy?
Do decentralized machine learning algorithms provide comparable 12-lead ECG classification performance to centralized machine learning while preserving data privacy?
Effect estimate: -0.036 AUROC
Absolute Event Rate: 0.836% vs 0.872%
Decentralized machine learning algorithms, particularly batch-wise sequential learning, can achieve ECG classification performance comparable to centralized models while preserving data privacy.
May enable privacy-preserving multi-institutional ECG classification; leaves open optimal decentralized strategies and generalizability.
Purpose: Artificial intelligence and machine learning have led to prominent and spectacular innovations in various scenarios. Application in medicine, however, can be challenging due to privacy concerns and strict legal regulations. Methods that centralize knowledge instead of data could address this issue. Methods: In this work, 6 different decentralized machine learning algorithms are applied to 12-lead ECG classification and compared to conventional, centralized machine learning. Results: The results show that state-of-the-art federated learning leads to reasonable losses of classification performance compared to a standard, central model (-0.054 AUROC) while providing a significantly higher level of privacy. A proposed weighted variant of federated learning (-0.049 AUROC) and an ensemble (-0.035 AUROC) outperformed the standard federated learning algorithm. Overall, considering multiple metrics, the novel batch-wise sequential learning scheme performed best (-0.036 AUROC to baseline). Conclusion: Although, the technical aspects of implementing them in a real-world application are to be carefully considered, the described algorithms constitute a way forward towards preserving-preserving AI in medicine.
No takes yet. Share an insight, caveat, or question.
Baumgartner et al. (2022) studied 12-lead ECG classification (n=43,101). Decentralized machine learning algorithms (e.g., Batch-wise Sequential Learning) vs. Centralized machine learning (Baseline model) was evaluated on Classification performance (AUROC) (-0.036 AUROC). A novel batch-wise sequential learning scheme performed best among decentralized methods, resulting in a minor loss of classification performance (-0.036 AUROC) compared to a centralized baseline model.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: