Advanced models for pre-operative risk stratification in adult spinal deformity surgery: evaluating computer vision, large language, and genome-wide association approaches in a limited sized cohort

Introduction: Adult Spinal Deformities (ASD) are varied spinal abnormalities affecting the lumbar or thoracolumbar regions, often necessitating surgical intervention when associated with pain or worsening deformity. Predicting post-operative complications and revision surgery is critical for surgical planning and patient counseling. Due to the rarity of ASD, machine learning applications have been limited to basic models (e.g.: logistic regression, standard neural networks) and coarse clinical variables. We present the novel application of advanced models (CNN, LLM, GWAS) using complex data types (radiographs, clinical notes, genomics) for ASD outcome prediction.

Methods: We developed a CNN trained on 209 ASD patients (1549 radiographs) from the Stanford Research Repository, a CNN pre-trained on VinDR-SpineXR (10,468 spine radiographs), and a LLM using free-text clinical notes from the same 209 patients, trained via Gatortron. Additionally, we conducted a GWAS using the UK Biobank, contrasting 540 surgical ASD patients with 7,355 non-surgical ASD patients.

Results: The LLM notably outperformed the CNN in predicting pulmonary complications (F1: 0.545 vs. 0.2881), neurological complications (F1: 0.250 vs. 0.224), and sepsis (F1: 0.382 vs. 0.132). Pre-trained CNN showed improved sepsis prediction (AUC: 0.638 vs. 0.534) but reduced performance for neurological complication prediction (AUC: 0.545 vs. 0.619). The LLM demonstrated high specificity (0.946) and positive predictive value (0.467) for neurological complications. The GWAS identified 21 significant (p < 10-5) SNPs associated with ASD surgery risk (OR: Mean: 3.17, SD: 1.92, Median: 2.78), with the highest odds ratio (8.06) for the LDB2 gene, implicated in ectoderm differentiation.

Conclusion : This study exemplifies the innovative application of cutting-edge models to forecast outcomes in ASD, underscoring the utility of complex data in outcome prediction for neurosurgical conditions. It evidences the promise of genetic models in identifying surgical risks and supports the integration of complex machine learning tools for informed surgical decision-making in ASD.