# Natural language processing in the classification of radiology reports in benign gallbladder diseases

## TL;DR

This paper presents a natural language processing system that classifies radiology reports to determine if benign gallbladder diseases require surgery.

## Contribution

The novel contribution is a deep learning-based text classifier for surgical decision-making in benign gallbladder diseases using radiology reports.

## Key findings

- Both CNN and BiLSTM models achieved high F1-scores (up to 0.967) in classifying gallbladder disease reports.
- Performance was consistent across 300- and 1,000-dimensional Word2Vec representations.
- The system can reliably distinguish between surgical and non-surgical cases in benign gallbladder diseases.

## Abstract

To develop a natural language processing application capable of automatically
identifying benign gallbladder diseases that require surgery, from radiology
reports.

We developed a text classifier to classify reports as describing benign
diseases of the gallbladder that do or do not require surgery. We randomly
selected 1,200 reports describing the gallbladder from our database,
including different modalities. Four radiologists classified the reports as
describing benign disease that should or should not be treated surgically.
Two deep learning architectures were trained for classification: a
convolutional neural network (CNN) and a bidirectional long short-term
memory (BiLSTM) network. In order to represent words in vector form, the
models included a Word2Vec representation, with dimensions of 300 or 1,000.
The models were trained and evaluated by dividing the dataset into training,
validation, and subsets (80/10/10).

The CNN and BiLSTM performed well in both dimensional spaces. For the 300-
and 1,000-dimensional spaces, respectively, the F1-scores were 0.95945 and
0.95302 for the CNN model, compared with 0.96732 and 0.96732 for the BiLSTM
model.

Our models achieved high performance, regardless of the architecture and
dimensional space employed.