LKAN: LLM-Based Knowledge-Aware Attention Network for Clinical Staging of Liver Cancer.

Abstract

Clinical staging of liver cancer (CSoLC), an important indicator for evaluating the degree of deterioration of primary liver cancer cells (PLCCs), is key in the diagnosis, treatment, and rehabilitation of liver cancer. In China, the current CSoLC adopts the China liver cancer (CNLC) staging, which is usually evaluated by clinicians based on the patient's radiology reports. Therefore, inferring clinical information from unstructured radiology reports can provide auxiliary decision support for clinicians. The key to solving the challenging task is to guide the model to pay attention to the staging-related words or sentences, and the following issues may occur: 1) Imbalanced categories: The symptoms of liver cancer in the early- or mid-stage are not obvious, resulting in more data in the end-stage. 2) Domain sensitivity of liver cancer data: The liver cancer dataset contains a large amount of domain knowledge, and the conventional methods can exacerbate out-of-vocabulary, which greatly affects the accuracy of classification. 3) Free-text and lengthy report: The radiology report of liver cancer sparsely describes various lesions with domain-specific terms, which poses difficulties in mining key information related to staging. To tackle these challenges, this article proposes a large language model (LLM)-based Knowledge-aware Attention Network (LKAN) for CSoLC. First, for maintaining semantic consistency, LLM and a rule-based algorithm are integrated to generate more diverse and reasonable data. Second, unlabeled radiology corpus of liver cancer are pre-trained to introduce domain knowledge for subsequent representation learning. Third, attention is improved by incorporating both global and local features, which can provide professional guidance for the classifier to focus on the important information. Compared with the baseline models, the classification accuracy of LKAN has achieved the best results with 90.3% Accuracy, 90.0% Macro_F1 score, and 90.0% Macro_Recall. The code is available at https://github.com/xczhh/Supplemental-Material.