A dynamic nomogram for predicting the probability of irreversible neurological dysfunction after cervical spinal cord injury: research based on clinical features and MRI data

Spinal Cord Injury (SCI) presents a significant challenge in rehabilitation medicine, with recovery outcomes varying widely among individuals. Machine learning (ML) is a promising approach to enhance the prediction of recovery trajectories, but its integration into clinical practice requires a thorough understanding of its efficacy and applicability. We systematically reviewed the current literature on data-driven models of SCI recovery prediction. The included studies were evaluated based on a range of criteria assessing the approach, implementation, input data preferences, and the clinical outcomes aimed to forecast. We observe a tendency to utilize routinely acquired data, such as International Standards for Neurological Classification of SCI (ISNCSCI), imaging, and demographics, for the prediction of functional outcomes derived from the Spinal Cord Independence Measure (SCIM) III and Functional Independence Measure (FIM) scores with a focus on motor ability. Although there has been an increasing interest in data-driven studies over time, traditional machine learning architectures, such as linear regression and tree-based approaches, remained the overwhelmingly popular choices for implementation. This implies ample opportunities for exploring architectures addressing the challenges of predicting SCI recovery, including techniques for learning from limited longitudinal data, improving generalizability, and enhancing reproducibility. We conclude with a perspective, highlighting possible future directions for data-driven SCI recovery prediction and drawing parallels to other application fields in terms of diverse data types (imaging, tabular, sequential, multimodal), data challenges (limited, missing, longitudinal data), and algorithmic needs (causal inference, robustness)

show abstract

MRI variables and peripheral inflammatory response biomarkers predict severity and prognosis in patients with acute cervical traumatic spinal cord injury

Ding,

Zhou,

Wang

et al. 2024

BMC Musculoskelet Disord

View full text Add to dashboard Cite

Objective Traumatic spinal cord injury (TSCI) stands as one of the most profoundly damaging and debilitating conditions. This study aims to explore the potential of magnetic resonance imaging (MRI) variables and peripheral inflammatory indicators as promising biomarkers. It aims to understand their significance in evaluating the severity and predicting the prognosis of TSCI. Furthermore, the study aims to ascertain whether combining these indicators could enhance the accuracy of injury assessment and predictive prognostic ability. Methods A multicentre retrospective cohort study was conducted to assess the severity and prognostic value of MRI variables and peripheral inflammatory response biomarkers in patients with acute cervical TSCI. The study involved 374 patients with acute cervical TSCI drawn from the First Affiliated Hospital of Nanchang University and the Second Affiliated Hospital of Nanchang University. The severity and prognosis of patients with acute cervical TSCI were assessed using the American Spinal Injury Association Impairment Scale (AIS). The correlation between MRI variables, peripheral inflammatory response biomarkers, admission severity, and the 1-year follow-up prognosis was analysed. Results After the initial assessment using the AIS grade system, 169 (49.2%) patients fell into the severe category for cervical TSCI (AIS A–B), while 205 (50.8%) patients were classified as non-severe cases (AIS C–E). The MRI variables (intramedullary lesion length [IMLL], Brain and Spinal Injury Centre [BASIC], maximum spinal cord compression [MSCC], and maximum canal compromise [MCC]) and inflammatory response biomarkers (white blood cells [WBCs], neutrophils, and C-reactive protein [CRP]) exhibited a consistent decrease correlating with the severity grades noted upon admission. Among the 374 patients assessed, 147 (39.3%) experienced a poor prognosis, as indicated by the AIS grade during the 1-year follow-up. MRI variables and peripheral inflammatory response biomarkers declined in correspondence with the follow-up AIS grades. Sex ( p < 0.001), IMLL ( p < 0.001), MSCC ( p < 0.001), MCC ( p < 0.001), BASIC ( p < 0.001), WBC ( p < 0.001), neutrophils ( p < 0.001), and CRP ( p < 0.001) were statistically significant in predicting poor outcomes. Through multivariate logistic regression analysis, BASIC score and CRP emerged as independent predictors of poor prognosis. Notably, the model combining the BASIC score and CRP yielded a larger area under the curve compared to models using only the BASIC score or CRP individually. Conclusions The BASIC score and CRP are crucial biomarkers for evaluating the severity of cervical TSCI and predicting prognosis. Their combination proved to be a more robust d...

show abstract

A dynamic nomogram for predicting the probability of irreversible neurological dysfunction after cervical spinal cord injury: research based on clinical features and MRI data

Cited by 4 publications

References 42 publications

Data-driven prediction of spinal cord injury recovery: An exploration of current status and future perspectives

Data-driven prediction of spinal cord injury recovery: An exploration of current status and future perspectives

Data-driven prediction of spinal cord injury recovery: an exploration of current status and future perspectives

MRI variables and peripheral inflammatory response biomarkers predict severity and prognosis in patients with acute cervical traumatic spinal cord injury

Contact Info

Product

Resources

About