The model assigns assertion status to clinical entities such as negation, uncertain, hypothetical, experiencer and conditional labels, extracted by NER based on their context in the text. It also has a provision for assertion type filter and is compatible with Spark NLP versions 2.7.2+. The assertion model was trained on i2b2 (sampled from MIMIC) dataset and uses BDLSTM and also has versions that use Logistic Regression, SBERT and BioBERT as base algorithms. It’s one of the state-of-the-art models in the industry to detect clinical assertions.
Input variables : Clinical text
Output Variables : Assertion status (absent, present, conditional, associated_with_someone_else, hypothetical, possible)
| 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 : nlp.johnsnowlabs.com
| Model Category | : | Commercial |
| Date Published | : | January, 2021 |
| Healthcare Domain | : |
Payer
Provider |
| Code | : | github.com |
| Clinical Information Extraction |