Overview
A block diagram of the Phase 01 models is provided below. The demo notebook provides an annotated version of the source code for one of the models: the Findings vs No Findings model. Since many of the models in Phase 01 use this same format, this should provide a good example for understanding how the models during this phase of the project were trained. To view the actual source code (*.py
files) used for training and classifying, please refer to the train and classify documentation.
In the table below, we provide a description of the different tasks and architectures of the models.
Model Name |
Task |
Architecture |
Source Code Function |
---|---|---|---|
Findings or No Findings |
Identify if the report has (i) lung or adrneal findings or (ii) no findings |
Stacked BiLSTM (Tensorflow / Keras) |
|
Lung or Adrenal Findings |
Identify if the report has (i) lung or (ii) adrneal findings |
Stacked BiLSTM (Tensorflow / Keras) |
|
Lung Recommended Procedure (i.e., Chest CT or Ambiguous Follow- up Recommended) |
Identify if a report with lung has findings recommends (i) a Chest CT follow-up or (ii) some other ambiguous procedure |
Stacked BiLSTM (Tensorflow / Keras) |
|
Comment Extraction |
Identify the portion of the report that contains the findings text |
XGBoost |