Classification of lumbar spine disorders using large language models and MRI segmentation.

Background: MRI is critical for diagnosing lumbar spine disorders but its complexity challenges diagnostic accuracy. This study proposes a BERT-based large language model (LLM) to enhance precision in classifying lumbar spine disorders through the integration of MRI data, textual reports, and numerical measurements.
Methods: The segmentation quality of MRI data is evaluated using dice coefficients (cut-off: 0.92) and intersection over union (IoU) metrics (cut-off: 0.88) to ensure precise anatomical feature extraction. The CNN extracts key lumbar spine features, such as lumbar lordotic angle (LLA) and disc heights, which are tokenized as direct scalar values representing positional relationships. A data source of 28,065 patients with various disorders, including degenerative disc disease, spinal stenosis, and spondylolisthesis, is used to establish diagnostic standards. These standards are refined through post-CNN processing of MRI texture features. The BERT-based spinal LLM model integrates these CNN-extracted MRI features and numerical values through early fusion layers.
Results: Segmentation analysis illustrate various lumbar spine disorders and their anatomical changes. The model achieved high performance, with all key metrics nearing 0.9, demonstrating its effectiveness in classifying conditions like spondylolisthesis, herniated disc, and spinal stenosis. External validation further confirmed the model's generalizability across different populations. External validation on 514 expert-validated MRI cases further confirms the model's clinical relevance and generalizability. The BERT-based model classifies 61 combinations of lumbar spine disorders.
Conclusions: The BERT-based spinal LLM significantly improves the precision of lumbar spine disorder classification, supporting accurate diagnosis and treatment planning.
Competing Interests: Declarations Ethics approval and consent to participate The composite dataset of lumbar spine MRI images, accessed on March 2, 2021, is publicly available at https://data.mendeley.com/datasets/k3b363f3vz/2. Since the dataset does not include any identifiable personal information or sensitive medical data that could compromise patient confidentiality, it falls outside the scope of ethical review requirements typically mandated for studies involving human subjects. The focus of this dataset is strictly on clinically relevant spinal measurements and imaging correlations, without involving patient-specific details or interventions that would necessitate ethical oversight. Therefore, the use of this dataset for research purposes is considered exempt from formal ethical approval processes. Consent for publication Not applicable. Competing interests The authors declare no competing interests.
(© 2024. The Author(s).)