Patient-Specific Readmission Prediction and Intervention for Health Care is one of the applications of Population Health Management for Healthcare - A Cortana Intelligence Solution. The product aims at improving clinical and health outcomes while managing and reducing cost with help of Machine Learning techniques. A classifier is constructed based on patients’ demographic information, medical record history, glucose readings and other data obtained from UCI Data Repository - Diabetes 130-US hospitals for years 1999-2008 Data Set. It is trained offline and used for scoring. The predictions from the classifier are used in decision system to guide decisions about post-discharge interventions. It also creates CMIO report which aims to serve the management team for high level performance of this patient-specific prediction and interventions assignment system.
Input variables : Patients’ demographic information (age, gender, zip code), Medical record history
(e.g. present illness, physical exam results, and medication), Glucose readings
Output Variables : Scores and predictions
| 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 |
| Business | : | Readmission Rate | Avg Hospital LOS |
| 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 : github.com
| Model Category | : | Commercial |
| Date Published | : | August, 2017 |
| Healthcare Domain | : | Provider |
| Code | : | github.com |
| Utilization Management |
| Readmission |