Summary of Classifying Cancer Stage with Open-source Clinical Large Language Models, by Chia-hsuan Chang et al.
Classifying Cancer Stage with Open-Source Clinical Large Language Models
by Chia-Hsuan Chang, Mary M. Lucas, Grace Lu-Yao, Christopher C. Yang
First submitted to arxiv on: 2 Apr 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A novel approach to extract pathologic tumor-node-metastasis (pTNM) staging information from unstructured clinical reports in electronic health records is proposed. This method leverages open-source clinical large language models (LLMs), eliminating the need for labor-intensive labeled training datasets. The study compares LLMs with a BERT-based model fine-tuned using labeled data, demonstrating that while LLMs struggle with Tumor (T) classification, they can achieve comparable performance on Metastasis (M) classification and improved performance on Node (N) classification with suitable prompting strategies. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Doctors need to know the stage of a patient’s cancer to decide the best treatment. Right now, this information is often buried in unstructured reports from different healthcare providers. To help, researchers developed a way for computers to extract this important information without needing special training data. Instead, they used large language models that can learn from general medical texts. The results show that these models are good at identifying whether cancer has spread to lymph nodes or other parts of the body, but still need some improvement in understanding how far the cancer has grown. |
Keywords
» Artificial intelligence » Bert » Classification » Prompting
