Integration of natural and deep artificial cognitive models in medical images: BERT-based NER and relation extraction for electronic medical records

Abstract

IntroductionMedical images and signals are important data sources in the medical field, and they contain key information such as patients' physiology, pathology, and genetics. However, due to the complexity and diversity of medical images and signals, resulting in difficulties in medical knowledge acquisition and decision support.MethodsIn order to solve this problem, this paper proposes an end-to-end framework based on BERT for NER and RE tasks in electronic medical records. Our framework first integrates NER and RE tasks into a unified model, adopting an end-to-end processing manner, which removes the limitation and error propagation of multiple independent steps in traditional methods. Second, by pre-training and fine-tuning the BERT model on large-scale electronic medical record data, we enable the model to obtain rich semantic representation capabilities that adapt to the needs of medical fields and tasks. Finally, through multi-task learning, we enable the model to make full use of the correlation and complementarity between NER and RE tasks, and improve the generalization ability and effect of the model on different data sets.Results and discussionWe conduct experimental evaluation on four electronic medical record datasets, and the model significantly out performs other methods on different datasets in the NER task. In the RE task, the EMLB model also achieved advantages on different data sets, especially in the multi-task learning mode, its performance has been significantly improved, and the ETE and MTL modules performed well in terms of comprehensive precision and recall. Our research provides an innovative solution for medical image and signal data.

Deep learning-based identification of esophageal cancer subtypes through analysis of high-resolution histopathology images

“…Artificial intelligence (AI) and deep learning (DL) diagnostic systems are becoming increasingly popular, opening up new possibilities in image analysis ( Guo et al, 2023 ; Muller et al, 2023 ; Wang et al, 2023 ; Xiao et al, 2023 ). By harnessing shape and texture attributes alongside higher-order spatial features that capture intricate pixel-level relationships, these systems elevate images into high-dimensional features, vastly enhancing their capability for detection and classification ( Kumar et al, 2012 ).…”