Computer-assisted image processing 12 lead ECG model to diagnose hyperkalemia

详细信息    查看全文

• 作者：Venu Velagapudi ; MD^a ; ^{velagapv@gmail.com} ; John C. O'Horo ; MD^b ; Anu Vellanki ; MD^c ; Stephen P. Baker ; MScPH^d ; Rahul Pidikiti ; MD^c ; Jeffrey S. Stoff ; MD^a ; Dennis A. Tighe ; MD^c
• 关键词：Hyperkalemia ; ECG ; T wave width ; T wave slope ; QRS prolongation
• 刊名：Journal of Electrocardiology
• 出版年：2017
• 出版时间：January-February 2017
• 年：2017
• 卷：50
• 期：1
• 页码：131-138
• 全文大小：472 K
• 卷排序：50

文摘

We evolved and validated a computer assisted ECG model to diagnose hyperkalemia. Preditected K = 5.2354 + 0.03434 × Descending T slope – 0.2329 × T width – 0.9652 (QRS not > 100) AUROC 0.798 (95% CI 0.74–0.86) training set, 0.78 (0.69–0.88) validation set. Model performance was optimal (maximal cut point analysis at potassium 5.91 mEq/L). The model outperformed traditional ECG interpretation. Improved sensitivity, specificity over prior studies at relevant hyperkalemia range.