-
r37980778c78--b63a2e02318bcd8572c5ad41eae40d97
Characteristics of patient groups: Pre and post-treatment. -
r37980778c78--cc3b79340140a23d8c104fd74f7a1f2b
Summary on precision and recall for 21 different random data elements validated on multiple data sets of echocardiographic reports. -
Performance comparison according to top returned results sample.
Manual evaluation of 20 abstracts retrieved from PubMed using a search query VS automatic classification into the corresponding node in our system. -
r37980778c78--42d6c4c7018d7a39ebae4da110bb5f38
Performance of our system in finding the association between genomic anomaly and drug responses in PharmGKB dataset. -
r37980778c78--fb7dcb9856e193a3049527caf7e824f0
The patient distribution across the ten selected OCSG. -
r37980778c78--1c14df7c2c5f9b31077f6bd73981e481
Main discussion topics and sub-topics discussed in the identified groups. -
eGARD: Extracting associations between genomic anomalies and drug responses f...
Tumor molecular profiling plays an integral role in identifying genomic anomalies which may help in personalizing cancer treatments, improving patient outcomes and minimizing... -
r37980778c78--9b6c4ca72d080190448bc8b083036c31
Examples of information considered relevant for different nodes in the exposure taxonomy. -
r37980778c78--13ab640b2dc4fbc9ea4a17fc5c997f98
Comparison between manual and automatic classification of articles describing measurements of nine chemicals/chemical groups in human blood and milk. -
r37980778c78--17061dbdd3bb12d1d693cfe8cbc5caea
Performance of NLP-based†† Algorithm by Patient Characteristic. -
r37980778c78--1a4bc347df4943c6cd1ccb9c7be292dc
Examples of EchoInfer’s identification of data element and corresponding value structured output. -
List of EchoInfer data elements targeted for extraction.
List of EchoInfer data elements targeted for extraction. -
The number of abstracts retrieved by PubMed using a search query VS the numbe...
The number of abstracts retrieved by PubMed using a search query VS the number of abstracts classified into the corresponding node in our system. -
r37980778c78--c4897453be496fd19e6651012e275f01
Large volumes of data are continuously generated from clinical notes and diagnostic studies catalogued in electronic health records (EHRs). Echocardiography is one of the most... -
Performance of our system in finding the association of a genomic anomaly wit...
Performance of our system in finding the association of a genomic anomaly with drug responses () in InHouseSet1. -
Analysis of the influence of each feature type on the classification accuracy.
The classification accuracy is described as the F-score for each node after removal of respective feature type. The column “all” describes the F-scores when all feature types... -
r37980778c78--c6bd1de40382efabd53d5dc78d2553b1
The number of features for each node in the taxonomy after the feature selection step. -
r37980778c78--ed1a910d462849ad9d384aa59e81d813
Final model results on test data, features from all death certificate sections. -
Confusion matrix, SVM model, features from cause-of-death and description of ...
Confusion matrix, SVM model, features from cause-of-death and description of injury fields.
