🤖 AI Summary
The lack of standardized annotation schemas and general-purpose extraction methods for unstructured medical text hinders structured data mining. Method: This paper proposes an open-domain, end-to-end framework for numerical triplet (value–measure–unit) extraction. It introduces the first template-free, general numerical association annotation scheme and jointly models medical text preprocessing, sequence labeling, and relation extraction to simultaneously identify and align numerical values, clinical measures, and associated units. Results: Evaluated on a thrombectomy literature dataset, the framework achieves a Dice coefficient of 0.82; random sampling validation confirms high accuracy in value–entity relation matching. This work establishes a transferable methodology and practical paradigm for structuring template-free medical texts.
📝 Abstract
Many analysis and prediction tasks require the extraction of structured data from unstructured texts. However, an annotation scheme and a training dataset have not been available for training machine learning models to mine structured data from text without special templates and patterns. To solve it, this paper presents an end-to-end machine learning pipeline, Text2Struct, including a text annotation scheme, training data processing, and machine learning implementation. We formulated the mining problem as the extraction of metrics and units associated with numerals in the text. Text2Struct was trained and evaluated using an annotated text dataset collected from abstracts of medical publications regarding thrombectomy. In terms of prediction performance, a dice coefficient of 0.82 was achieved on the test dataset. By random sampling, most predicted relations between numerals and entities were well matched to the ground-truth annotations. These results show that Text2Struct is viable for the mining of structured data from text without special templates or patterns. It is anticipated to further improve the pipeline by expanding the dataset and investigating other machine learning models. A code demonstration can be found at: https://github.com/zcc861007/CourseProject