meghutch / traumascanner

Algorithm for identifying radiology reports of patients with post-traumatic hemorrhage

traumascanner

Identify radiology text-reports of patients with post-traumatic hemorrhage

Installation

$ pip install traumascanner

Usage

Use traumaScanner() to identify patients with post-traumatic hemorrhage from radiology text reports.

traumaScanner() is a key-word matching/regular expression (regex) and rules-based algorithm for identifying patients with post-traumatic hemorrhage after a traumatic brain injury (TBI).

In brief, the algorithm functions as follows:

1. Identifies radiology reports with at least one of the provided set of trauma-related keywords
2. Considers negation to remove false positive trauma related reports
3. Identifies and removes reports without hemorrhage

The function will output the following csv files:

• 01_potential_trauma_reports.csv - all radiology reports which matched at least one keyword
• 02_false_postive_trauma_reports.csv - the subset of potential_trauma_reports identified as being likely false positive for trauma
• 03_trauma_no_hemorrhage_reports.csv - the subset of potential_trauma_reports which had no hemorrhage
• 04_resucued_reports.csv - the subset of trauma_no_hemorrhage_reports which were likely false negatives
• 05_post_traumatic_hemorrhage_reports - the complete set of post-traumatic hemorrhage reports identified via the traumaScanner() algorithm.

Please see the demo to learn how to use traumaScanner() on your own radiology reports.

Note: Reports in example_reports/ do not include real patient information and were created with the help of ChatGTP. Please see the notebook with the code generated by ChatGPT for creating this example dataset.

License

traumascanner was created by Meghan Hutch. It is licensed under the terms of the MIT license.

Credits

traumascanner was created with cookiecutter and the py-pkgs-cookiecutter template.