基于人工智能的心电图图像自动解读:ECG-GPT的开发和多国验证
Khunte, A., Sangha, V., Oikonomou, E.K., Dhingra, L.S., Aminorroaya, A., Coppi, A., Shankar, S.V., Rockers, E., Mortazavi, B.J., Bhatt, D.L., Krumholz, H.M., Al-Kindi, S., Nadkarni, G.N., Vaid, A., & Khera, R. (2026). Artificial intelligence-based automated interpretation of images of electrocardiograms: development and multinational validation of ECG-GPT. European Heart Journal – Digital Health. https://doi.org/10.1093/ehjdh/ztag031
Khunte A, Sangha V, Oikonomou EK, Dhingra LS, Aminorroaya A, Coppi A, et al. Artificial intelligence-based automated interpretation of images of electrocardiograms: development and multinational validation of ECG-GPT. European Heart Journal – Digital Health. 2026; doi: 10.1093/ehjdh/ztag031
Khunte A, Sangha V, Oikonomou EK, et al. Artificial intelligence-based automated interpretation of images of electrocardiograms: development and multinational validation of ECG-GPT[J]. European Heart Journal – Digital Health. 2026. DOI: 10.1093/ehjdh/ztag031.
@article{khunte2026,
author = {Ankit Khunte and Vikramjeet Sangha and Emmanouil K Oikonomou and Lavi S Dhingra and Aminorroaya Aminorroaya and Alessio Coppi and Sravanthi V Shankar and Erin Rockers and Behnam J Mortazavi and Deepak L Bhatt and Harlan M Krumholz and Sinan Al-Kindi and Girish N Nadkarni and Akhil Vaid and Rohan Khera},
title = {Artificial intelligence-based automated interpretation of images of electrocardiograms: development and multinational validation of ECG-GPT},
journal = {European Heart Journal – Digital Health},
year = {2026},
doi = {10.1093/ehjdh/ztag031},
note = {PMID: 41853639},
}
TY - JOUR AU - Ankit Khunte AU - Vikramjeet Sangha AU - Emmanouil K Oikonomou AU - Lavi S Dhingra AU - Aminorroaya Aminorroaya AU - Alessio Coppi AU - Sravanthi V Shankar AU - Erin Rockers AU - Behnam J Mortazavi AU - Deepak L Bhatt AU - Harlan M Krumholz AU - Sinan Al-Kindi AU - Girish N Nadkarni AU - Akhil Vaid AU - Rohan Khera TI - Artificial intelligence-based automated interpretation of images of electrocardiograms: development and multinational validation of ECG-GPT T2 - European Heart Journal – Digital Health PY - 2026 DO - 10.1093/ehjdh/ztag031 AN - PMID:41853639 ER -
心电图(ECG)是最常用的诊断工具之一,但其准确解读需要专业的医学知识。本研究开发了ECG-GPT,一种基于人工智能的心电图图像自动解读系统,并进行了多国验证。该系统利用深度学习技术,能够直接从心电图图像中提取诊断信息,识别多种心脏异常。研究使用了来自多个国家和医疗中心的心电图数据进行训练和验证,涵盖了多种心律失常、心肌缺血、传导阻滞等心血管疾病的表现。验证结果显示,ECG-GPT在多项心电图诊断任务中表现优异,其诊断准确率可与经验丰富的心脏病专家媲美,在部分指标上甚至超过了人类判读。该系统具有实时处理能力,可广泛应用于初级医疗保健和资源有限地区,有助于提高心血管疾病的诊断效率和可及性,具有重要的临床应用价值。