This model is an example of Natural language Processing Technique implemented on Health Records to help Radiologists in diagnosis. A Radiologist has few minutes to analyze and interpret multiple images of a case,this leaves practically no time to thoroughly consult patient history. This models will extract compact summaries of textual diagnostic clinical information from patient EHR. The textual data can be in form of Discharge summaries, Radiology reports or visit notes as well as medical codes on which the model is trained. As per the query of interest for a Physician the model can generate summary from ranked sentences in order of importance. This ranking is possible because of Attention mechanism along with the Clinical BERT architecture.
Input variables : Sentences, ICD codes from patient's clinical records
Output Variables : Summarizer EHR information
| 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 | : | September, 2020 |
| Healthcare Domain | : |
Medical Technology
Provider |
| Code | : | github.com |
| Clinical Information Extraction |