Medkit is a new benchmarking suite designed specifically for medical sequential decision making. The models produces batch dataset, which can be used for training and evaluating methods for modelling human decision making, using a network consisting of an LSTM layer followed by two fully connected layers. The synthetic setting is via scenarios, scenarios are made up of domains, environments and policies. Medkit enables users to simulate decision making behaviours with various degrees of Markovianity, individual consistency, bounded rationality and variation in practice. The domains are based on real life datasets like data collected from general medicine floor of Ronald Reagan UCLA Medical Center in California, data for patients in the intensive care unit from Amsterdam UMC, etc, giving more realistic environments.
Input variables : Parameters, scenarios
Output Variables : Synthetic batch medical dataset
| 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 | : | June, 2021 |
| Healthcare Domain | : | Provider |
| Code | : | github.com |
| Data Privacy |
| Synthetic Data Generation |