JMIR Publications

ABSTRACT

Background:

Identifying and extracting family history information (FHI) from clinical reports is significant in recognizing disease susceptibility. However, FHI is usually described in a narrative manner within patients’ electronic health records (EHRs), which require the application of natural language processing technologies to automatically extract such information to provide a more comprehensive patient-centered information to the physicians.

Objective:

This study aims to overcome the two main challenges observed in previous researches focusing on FHI extraction. One is the requirement to develop post-processing rules to infer the member and the side information of family mentions. The other is to efficiently utilize intra- and inter-sentence information to assist FHI extraction.

Methods:

We formulated the task as a sequential labeling problem and proposed an enhanced relation-side scheme which encodes the required family member properties to not only eliminate the need of post-processing rules but also relieve the insufficient training instance issues. Moreover, an attention-based neural network structure was proposed to exploit cross-sentence information to identify FHI and its attributes requiring cross-sentence inference.

Results:

The dataset released by the 2019 n2c2/OHNLP family history extraction task was used to evaluate the performance of the proposed methods. We started by comparing the performance of the traditional neural sequence models with the ordinary scheme and the enhanced scheme. Next, we studied the effectiveness of the proposed attention-enhanced neural networks by comparing their performance with that of the traditional networks. It was observed that with the enhanced scheme, the recalls of the neural network can be improved and lead to an increase of F-score by 0.024. The proposed neural attention mechanism enhanced both the recall and the precision and resulted in an improved F-score of 0.807, which was ranked fourth in the shared task.

Conclusions:

We presented an attention-based neural network along with enhanced tag scheme, which enables the neural network model to learn and interpret the implicit relationship and side information of the recognized family members across sentences without relying on heuristic rules.