If modeled correctly, EHRs the potential to bring more efficient allocations of health care resources, driving early intervention strategies and advancing personalized healthcare. But EHRs are difficult to model because of their realization as noisy, multi-modal data occurring at irregular time intervals. To address the temporal nature of EHR, model treat EHRs as samples generated from a Temporal Point Process (TPP). Neural network parameterisations of TPPs are proposed, collectively referred as Neural TPPs. Proposed Neural TPPs, where labels are jointly modeled with time, significantly outperforms those that treat labels and time as conditionally independent. For completeness, model is evaluated with few datasets consisting of MIMIC-II, Stack Overflow and retweet data from Twitter. Also Synthea simulator is used which generates patient level EHRs using human expert curated Markov processes.
Input variables : EHR Data
Output Variables : future health outcomes
| 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 : arxiv.org
| Model Category | : | Public |
| Date Published | : | July, 2020 |
| Healthcare Domain | : |
Payer
Provider |
| Code | : | github.com |
| Health Risk Management |
| Health Risk Prediction |