Comparing DeepSeek and GPT-4o in ECG interpretation: Is AI improving over time?
This first evaluation of DeepSeek in ECG interpretation reveals its performance is lower than that of GPT-4o and human experts. While GPT-4o demonstrates greater accuracy, both models fall short of expert-level performance, underscoring the need for caution and further validation before clinical integration.
Authors
- Seval Komut (ORCID: https://orcid.org/0000-0002-9558-4832)
- Anılcan Tahsin Karahan
- Serkan Günay (ORCID: https://orcid.org/0000-0002-8343-0916)
- Yavuz Yiğit (ORCID: https://orcid.org/0000-0002-7226-983X)
- Mert Barındık (ORCID: https://orcid.org/0000-0002-4104-4048)
- Ahmet Öztürk (ORCID: https://orcid.org/0000-0002-9678-2195)
Institutions
- Hitit Üniversitesi (TR)
- Hamad General Hospital (QA)
- Hamad Medical Corporation (QA)
Publication Details
- Journal
- Heart & Lung
- Published
- 2025-09-13
- DOI
- https://doi.org/10.1016/j.hrtlng.2025.08.007
- Citations
- 3
- Primary Topic
- Explainable Artificial Intelligence (XAI)
- Type
- article
- Field-Weighted Citation Impact
- 7.44