Large Language Models in Cancer Imaging: Applications and Future Perspectives

Abstract

Recently, there has been tremendous interest on the use of large language models (LLMs) in radiology. LLMs have been employed for various applications in cancer imaging, including improving reporting speed and accuracy via generation of standardized reports, automating the classification and staging of abnormal findings in reports, incorporating appropriate guidelines, and calculating individualized risk scores. Another use of LLMs is their ability to improve patient comprehension of imaging reports with simplification of the medical terms and possible translations to multiple languages. Additional future applications of LLMs include multidisciplinary tumor board standardizations, aiding patient management, and preventing and predicting adverse events (contrast allergies, MRI contraindications) and cancer imaging research. However, limitations such as hallucinations and variable performances could present obstacles to widespread clinical implementation. Herein, we present a review of the current and future applications of LLMs in cancer imaging, as well as pitfalls and limitations.

Keywords: artificial intelligence; cancer; imaging; large language model.

Figure 1

The architecture of a general LLM encoder–decoder framework. The input is encoded into tokens via the encoder for the decoder to generate the output. Both encoder and decoder modules were built by stacks of transformer layers.

Figure 2

Despite advances in radiology, challenges in cancer imaging remain on both the physician side and the patient side. This includes an ever-increasing patient load, information overload from newer modalities and techniques, extensive relevant patient histories, poor standardization of assessment strategies, variability in diagnoses, and difficulty in patient communication and understanding. LLMs are one potential way of supporting radiologists in these challenges.

Figure 3

Current and future applications of large language models in cancer imaging. The Chinese Wén (文) character can be translated to mean “language” or “writing”. Figure created using BioRender, version 04.