Diabetes is a condition that impairs the body's ability to process blood glucose. Early detection of diabetes can make patients life more healthy and happy. Machine learning techniques are applied on EHR data for patients, also an ensemble model based on this techniques is used to perform diabetes prediction. Type-2 diabetes is usually diagnosed later on in life for most of the patients. Data for about 10,000 patient records between 2009 and 2012 was obtained form an EHR company called Practice Fusion. The data has important biometrics like age, gender, height, weight, systolic and diastolic blood pressure. Some popular machine learning techniques like KNN, SVM, Random Forest, Gradient Boosting etc. are applied on this EHR data for patients, also an ensemble model based on this techniques is used to perform better classification. In this ensemble model weighted average of outcomes of all popular machine learning algorithms is accounted for final prediction. This different approach increases the accuracy for the classification problem for type-2 diabetes and can be used as an alert for patients.
Input variables : Weight, Height, Gender, Age, Diastolic and Systolic blood pressure, BMI
Output Variables : Diabetes Status
| 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 : github.com
Additional links : arxiv.org
| Model Category | : | Public |
| Date Published | : | January, 2019 |
| Healthcare Domain | : | Provider |
| Code | : | Not available |
| Health Risk Management |
| Chronic Care Management |