Arrhythmia is a condition of improper heart beating, which is diagnosed with help of electrocardiograms(ECG ). In this model an algorithm is applied which can detect different types of arrhythmia from ECG data of a single-lead wearable monitor. The model shows good performance in practical evaluations, thus can help in reducing diagnosis errors.The model takes raw ECG data as input and generates the prediction every 1.28 which is the output interval and segments into one of the twelve arrhythmia types. A Convolutional Neural Network architecture is used for in this implementation and it produces high performance on validation parameters. The model was also compared to a group cardiologists and it was able to perform better in diagnosing arrhythmia.
Input variables : Electrocardiogram (ECG) Data
Output Variables : Type of heart arrhythmias
| Statistical | : | Somers D | Accuracy | Precision and Recall | Confusion Matrix | F1 Score | Roc and Auc | Prevalence | Detection Rate | Balanced Accuracy | Cohen's Kappa | Concordance | Gini Coefficent | KS Statistic | Youden's J Index |
| Infrastructure | : | Log Bytes | Logging/User/IAMPolicy | Logging/User/VPN | CPU Utilization | Memory Usage | Error Count | Prediction Count | Prediction Latencies | Private Endpoint Prediction Latencies | Private Endpoint Response Count |
Visit Model : stanfordmlgroup.github.io
Additional links : stanfordmlgroup.github.io
| Model Category | : | Public |
| Date Published | : | July, 2017 |
| Healthcare Domain | : |
Medical Technology
Provider |
| Code | : | github.com |
| Health Risk Management |
| Disease Detection |