computerized ecg machine (GE Healthcare)
Structured Review

Computerized Ecg Machine, supplied by GE Healthcare, used in various techniques. Bioz Stars score: 92/100, based on 1006 PubMed citations. ZERO BIAS - scores, article reviews, protocol conditions and more
https://www.bioz.com/result/computerized ecg machine/product/GE Healthcare
Average 92 stars, based on 1006 article reviews
Images
1) Product Images from "Usefulness of multi-labelling artificial intelligence in detecting rhythm disorders and acute ST-elevation myocardial infarction on 12-lead electrocardiogram"
Article Title: Usefulness of multi-labelling artificial intelligence in detecting rhythm disorders and acute ST-elevation myocardial infarction on 12-lead electrocardiogram
Journal: European Heart Journal. Digital Health
doi: 10.1093/ehjdh/ztab029
Figure Legend Snippet: Data collection and labelling. In total, 72 647 12-lead electrocardiograms were retrospectively retrieved. Electrocardiograms with duplicate data ( n = 3111), incomplete information of age or age less than 18 years old ( n = 1673), absence of definite diagnosis ( n = 6759), and those not performed at China Medical University Hospital ( n = 567) were excluded. The remaining 60 537 electrocardiogram signals from 35 981 patients were included in this study.
Techniques Used:
Figure 2 . " title="... physicians, and the commercial algorithm correctly classified the electrocardiogram as second degree AV block and acute STEMI, ..." property="contentUrl" width="100%" height="100%"/>
Figure Legend Snippet: Two representative electrocardiograms in the external testing. ( A ) The long short-term memory model, all of the four cardiologists, one of the three emergency physicians, and the commercial algorithm correctly classified the electrocardiogram as second degree AV block and acute STEMI, whereas two emergency physicians and all of the three internists annotated either second degree AV block or ST-elevation myocardial infarction but not both for this electrocardiogram. ( B ) The long short-term memory model correctly classified the electrocardiogram as BIGEMINY and first degree AV block, while most doctors (8 of the 10 physicians) and the commercial algorithm only annotated BIGEMINY but not first degree AV block. Abbreviations for the electrocardiogram diagnoses as in
Techniques Used: Blocking Assay
Figure 2 . Only the four important classes, discussed in the main text are shown here, the rest was presented in Supplementary material online, Figure S1 . " title="... long short-term memory; MR, internists; abbreviations for the electrocardiogram diagnoses are as in Figure Legend Snippet: Performance of the long short-term memory model and different groups of board-certified doctors in detecting acute ST-elevation myocardial infarction and different heart rhythms. These are the accuracies and receiver operating characteristic curves in detecting ( A ) ST-elevation myocardial infarction ( B ) atrial fibrillation ( C ) complete heart block ( D ) paroxysmal supraventricular tachycardia of our artificial intelligence model and the results of a commercial algorithm and different groups of doctors in the comparative external tests. The orange line was the receiver operating characteristic curve of the long short-term memory model. The different colour points represent different groups of board-certified doctors. AI, artificial intelligence; CV, cardiologists; ER, emergency physicians; LSTM, long short-term memory; MR, internists; abbreviations for the electrocardiogram diagnoses are as in
Techniques Used: Blocking Assay
Figure Legend Snippet: The representative ST-elevation myocardial infarction electrocardiogram images of false negative cases missed by humans and the computer in the external test. ( A ) The artificial intelligence model correctly annotated ST-elevation myocardial infarction, whereas one of the four cardiologists labelled ‘Not STEMI’ resulting in a false negative annotation. ( B ) All the four cardiologists correctly diagnosed ST-elevation myocardial infarction, while the artificial intelligence model annotated ‘Not STEMI’ and it was counted as a false negative.
Techniques Used:
