MetaPred is a meta-learning framework for low-resource predictive modeling with patient EHRs. The idea behind is to employ labeled patients from high-resource domains and design a learning to transfer framework with sources and a simulated target in meta-learning. This is achieved using four steps: constructing episodes by sampling from the source domains and the simulated target domain; learn the parameters of predictors in an episode-by-episode manner; fine-tuning the model parameters on the genuine target domain and predicting the target clinical risk with Convolutional Neural Network (CNN) and Long-Shot Term Memory (LSTM) Network as base predictors.
Input variables : Features from EHR, Patient diagnoses history, ICD codes
Output Variables : Patients with a disease
| 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 | : | Population at High Risk of Disease | Risk by Geography | Risk by Demographics | Risk by Clinical Parameters | Optimized Hospital Resource Utilization | Decreased Cost of Care | Decreased Patient Visits |
| 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 : dl.acm.org
| Model Category | : | Public |
| Date Published | : | August, 2019 |
| Healthcare Domain | : | Provider |
| Code | : | github.com |
| Health Risk Management |
| Health Risk Prediction |