Predictive Modeling of Outcomes After Traumatic and Nontraumatic Spinal Cord Injury Using Machine Learning: Review of Current Progress and Future Directions

Low Back Pain Among Health Sciences Undergraduates: Results Obtained from a Machine-Learning Analysis

Background and objective. Low back pain (LBP) is considered the most common and challenging disorder in health care. Although its incidence increases with age, a student's sedentary behavior could contribute to this risk. Through machine learning (ML), advanced algorithms can analyze complex patterns in health data, enabling accurate prediction and targeted prevention of medical conditions such as LBP. This study aims to detect the factors associated with LBP among health sciences students. Methods. A self-administered modified version of the Standardized Nordic Questionnaire was completed by 222 freshman health sciences students from May to June 2022. A supervised random forest algorithm was utilized to analyze data and prioritize the importance of variables related to LBP. The model's predictive capability was further visualized using a decision tree to identify high-risk patterns and associations. Results. A total of 197/222 (88.7%) students participated in this study, most of whom (75%) were female. Their mean age and body mass index were 23 ± 3.8 and 23 ± 3.5, respectively. In this group, 46% (n = 90) of the students reported having experienced LBP in the last month, 15% (n = 30) were smokers, and 60% (n = 119) were involved in prolonged sitting (more than 3 h per day). The decision tree of ML revealed that a history of pain (score = 1), as well as disability (score= 0.34) and physical activity (score = 0.21), were significantly associated with LBP. Conclusions. Approximately 46% of the health science students reported LBP in the last month, and a machine-learning approach highlighted a history of pain as the most significant factor related to LBP.

Predicting morbidity and mortality after surgery for isolated traumatic spinal injury without spinal cord injury

BACKGROUND

Traumatic spinal injuries are associated with a high risk of morbidity and mortality. The aim of this study is to investigate which variables best predict adverse outcomes in patients who had surgery for isolated traumatic spinal injury without spinal cord injury.

METHODS

The American College of Surgeons Trauma Quality Improvement Program database was used to identify adult (18 years or older) surgically managed patients with an isolated traumatic spinal injury, without spinal cord injury admitted between 2013 and 2021. An isolated injury was defined as a spine Abbreviated Injury Scale score ≥2 and an Abbreviated Injury Scale score ≤1 in the remaining body regions, as well as corresponding International Classification of Diseases, Ninth and Tenth Revision, codes. The predictive value of demographic, clinical, and comorbidity data was evaluated using logistic regression models and ranked using the permutation importance method.

RESULTS

A total of 39,457 patients were included in the study, of whom 554 died during hospitalization. The most important variables for predicting in-hospital mortality were age, sex, Glasgow Coma Scale on admission, Orthopedic Frailty Score, and cervical spine injury. The most important variables for predicting complications were age, cervical spine injury, the need for cervical spine surgery, Revised Cardiac Risk Index, and alcohol use disorder. Finally, age, cervical spine injury, sex, Glasgow Coma Scale on admission, and Orthopedic Frailty Score had the highest relative importance when predicting failure to rescue. Models based on the five most important variables for each outcome demonstrated an excellent predictive ability for in-hospital mortality (area under the receiver operating characteristic curve [AUROC], 0.84; 95% confidence interval [CI], 0.82-0.86) and failure to rescue (AUROC [95% CI], 0.86 [0.84-0.87]) as well as an acceptable predictive ability for complications (AUROC [95% CI], 0.72 [0.71-0.73]).

CONCLUSION

The most important factors identified to predict mortality, complications, and failure to rescue in traumatic spinal injury patients without spinal cord injury who undergo surgery were patients' age, sex, frailty, cervical spine injury that necessitated surgical intervention, and cardiovascular risk.

LEVEL OF EVIDENCE

Prognostic and Epidemiological; Level III.

The diagnostic and prognostic capability of artificial intelligence in spinal cord injury: A systematic review

Background

Artificial intelligence (AI) models have shown potential for diagnosing and prognosticating traumatic spinal cord injury (tSCI), but their clinical utility remains uncertain.

Method

ology: The primary aim was to evaluate the performance of AI algorithms in diagnosing and prognosticating tSCI. Subsequent systematic searching of seven databases identified studies evaluating AI models. PROBAST and TRIPOD tools were used to assess the quality and reporting of included studies (PROSPERO: CRD42023464722). Fourteen studies, comprising 20 models and 280,817 pooled imaging datasets, were included. Analysis was conducted in line with the SWiM guidelines.

Results

For prognostication, 11 studies predicted outcomes including AIS improvement (30%), mortality and ambulatory ability (20% each), and discharge or length of stay (10%). The mean AUC was 0.770 (range: 0.682-0.902), indicating moderate predictive performance. Diagnostic models utilising DTI, CT, and T2-weighted MRI with CNN-based segmentation achieved a weighted mean accuracy of 0.898 (range: 0.813-0.938), outperforming prognostic models.

Conclusion

AI demonstrates strong diagnostic accuracy (mean accuracy: 0.898) and moderate prognostic capability (mean AUC: 0.770) for tSCI. However, the lack of standardised frameworks and external validation limits clinical applicability. Future models should integrate multimodal data, including imaging, patient characteristics, and clinician judgment, to improve utility and alignment with clinical practice.

SCIseg: Automatic Segmentation of Intramedullary Lesions in Spinal Cord Injury on T2-weighted MRI Scans

Purpose To develop a deep learning tool for the automatic segmentation of the spinal cord and intramedullary lesions in spinal cord injury (SCI) on T2-weighted MRI scans. Materials and Methods This retrospective study included MRI data acquired between July 2002 and February 2023. The data consisted of T2-weighted MRI scans acquired using different scanner manufacturers with various image resolutions (isotropic and anisotropic) and orientations (axial and sagittal). Patients had different lesion etiologies (traumatic, ischemic, and hemorrhagic) and lesion locations across the cervical, thoracic, and lumbar spine. A deep learning model, SCIseg (which is open source and accessible through the Spinal Cord Toolbox, version 6.2 and above), was trained in a three-phase process involving active learning for the automatic segmentation of intramedullary SCI lesions and the spinal cord. The segmentations from the proposed model were visually and quantitatively compared with those from three other open-source methods (PropSeg, DeepSeg, and contrast-agnostic, all part of the Spinal Cord Toolbox). The Wilcoxon signed rank test was used to compare quantitative MRI biomarkers of SCI (lesion volume, lesion length, and maximal axial damage ratio) derived from the manual reference standard lesion masks and biomarkers obtained automatically with SCIseg segmentations. Results The study included 191 patients with SCI (mean age, 48.1 years ± 17.9 [SD]; 142 [74%] male patients). SCIseg achieved a mean Dice score of 0.92 ± 0.07 and 0.61 ± 0.27 for spinal cord and SCI lesion segmentation, respectively. There was no evidence of a difference between lesion length (P = .42) and maximal axial damage ratio (P = .16) computed from manually annotated lesions and the lesion segmentations obtained using SCIseg. Conclusion SCIseg accurately segmented intramedullary lesions on a diverse dataset of T2-weighted MRI scans and automatically extracted clinically relevant lesion characteristics. Keywords: Spinal Cord, Trauma, Segmentation, MR Imaging, Supervised Learning, Convolutional Neural Network (CNN) Published under a CC BY 4.0 license.

LncRNA TSIX knockdown restores spinal cord injury repair through miR-30a/SOCS3 axis

Spinal cord injury (SCI) is a serious injury to the central nervous system. Previous studies have discovered that the development of SCI is associated with gene expression. The purpose of this study was to explore the significance of lncRNA TSIX in SCI and its underlying mechanism involved. An in vivo SCI mice model and an in vitro hypoxia-treated HT22 cells model were applied in this study. TSIX and SOCS3 expression in SCI tissues was measured by qRT-PCR, western blot and FISH assay. LV-sh-TSIX was injected into SCI mice intrathecally or subjected to HT22 cells to access the consequent alteration in inflammation response, cell apoptosis and functional recovery through ELISA, immunohistochemistry, TUNEL, flow cytometry assays and BMS scores. Then, the underlying mechanism of TSIX was analyzed by bioinformatics analysis and then confirmed by RIP, RNA pull-down and dual-luciferase reporter assay. It was identified that TSIX was up-regulated in HT22 cells under hypoxia operation and spinal cord tissues of SCI mice. TSIX knockdown improved the lesion size and BMS score and inhibited inflammation and cell apoptosis. MiR-30a was identified as a target for TSIX and SOCS3, and TSIX binds to miR-30a by competing with SOCS3, thereby counteracting miR-30a-mediated SOCS3 inhibition. In addition, LV-sh-TSIX effects were significantly overturned by miR-30a inhibition or SOCS3 over-expression. Knockdown of TSIX improved functional recovery and attenuated the inflammation response and cell apoptosis via miR-30a/SOCS3 axis. These results may provide a potential novel insight for SCI treatment.

Prediction of segmental motor outcomes in traumatic spinal cord injury: Advances beyond sum scores

BACKGROUND AND OBJECTIVES

Neurological and functional recovery after traumatic spinal cord injury (SCI) is highly challenged by the level of the lesion and the high heterogeneity in severity (different degrees of in/complete SCI) and spinal cord syndromes (hemi-, ant-, central-, and posterior cord). So far outcome predictions in clinical trials are limited in targeting sum motor scores of the upper (UEMS) and lower limb (LEMS) while neglecting that the distribution of motor function is essential for functional outcomes. The development of data-driven prediction models of detailed segmental motor recovery for all spinal segments from the level of lesion towards the lowest motor segments will improve the design of rehabilitation programs and the sensitivity of clinical trials.

METHODS

This study used acute-phase International Standards for Neurological Classification of SCI exams to forecast 6-month recovery of segmental motor scores as the primary evaluation endpoint. Secondary endpoints included severity grade improvement, independent walking, and self-care ability. Different similarity metrics were explored for k-nearest neighbor (kNN) matching within 1267 patients from the European Multicenter Study about Spinal Cord Injury before validation in 411 patients from the Sygen trial. The kNN performance was compared to linear and logistic regression models.

RESULTS

We obtained a population-wide root-mean-squared error (RMSE) in motor score sequence of 0.76(0.14, 2.77) and competitive functional score predictions (AUCwalker = 0.92, AUCself-carer = 0.83) for the kNN algorithm, improving beyond the linear regression task (RMSElinear = 0.98(0.22, 2.57)). The validation cohort showed comparable results (RMSE = 0.75(0.13, 2.57), AUCwalker = 0.92). We deploy the final historic control model as a web tool for easy user interaction (https://hicsci.ethz.ch/).

DISCUSSION

Our approach is the first to provide predictions across all motor segments independent of the level and severity of SCI. We provide a machine learning concept that is highly interpretable, i.e. the prediction formation process is transparent, that has been validated across European and American data sets, and provides reliable and validated algorithms to incorporate external control data to increase sensitivity and feasibility of multinational clinical trials.

Prediction of poststroke independent walking using machine learning: a retrospective study

BACKGROUND

Accurately predicting the walking independence of stroke patients is important. Our objective was to determine and compare the performance of logistic regression (LR) and three machine learning models (eXtreme Gradient Boosting (XGBoost), Support Vector Machines (SVM), and Random Forest (RF)) in predicting walking independence at discharge in stroke patients, as well as to explore the variables that predict prognosis.

METHODS

778 (80% for the training set and 20% for the test set) stroke patients admitted to China Rehabilitation Research Center between February 2020 and January 2023 were retrospectively included. The training set was used for training models. The test set was used to validate and compare the performance of the four models in terms of area under the curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and F1 score.

RESULTS

Among the three ML models, the AUC of the XGBoost model is significantly higher than that of the SVM and RF models (P < 0.001, P = 0.024, respectively). There was no significant difference in the AUCs between the XGBoost model and the LR model (0.891 vs. 0.880, P = 0.560). The XGBoost model demonstrated superior accuracy (87.82% vs. 86.54%), sensitivity (50.00% vs. 39.39%), PPV (73.68% vs. 73.33%), NPV (89.78% vs. 87.94%), and F1 score (59.57% vs. 51.16%), with only slightly lower specificity (96.09% vs. 96.88%). Together, the XGBoost model and the stepwise LR model identified age, FMA-LE at admission, FAC at admission, and lower limb spasticity as key factors influencing independent walking.

CONCLUSION

Overall, the XGBoost model performed best in predicting independent walking after stroke. The XGBoost and LR models together confirm that age, admission FMA-LE, admission FAC, and lower extremity spasticity are the key factors influencing independent walking in stroke patients at hospital discharge.

TRIAL REGISTRATION

Not applicable.

Machine-learning-based prediction by stacking ensemble strategy for surgical outcomes in patients with degenerative cervical myelopathy

BACKGROUND

Machine learning (ML) is extensively employed for forecasting the outcome of various illnesses. The objective of the study was to develop ML based classifiers using a stacking ensemble strategy to predict the Japanese Orthopedic Association (JOA) recovery rate for patients with degenerative cervical myelopathy (DCM).

METHODS

A total of 672 patients with DCM were included in the study and labeled with JOA recovery rate by 1-year follow-up. All data were collected during 2012-2023 and were randomly divided into training and testing (8:2) sub-datasets. A total of 91 initial ML classifiers were developed, and the top 3 initial classifiers with the best performance were further stacked into an ensemble classifier with a supported vector machine (SVM) classifier. The area under the curve (AUC) was the main indicator to assess the prediction performance of all classifiers. The primary predicted outcome was the JOA recovery rate.

RESULTS

By applying an ensemble learning strategy (e.g., stacking), the accuracy of the ML classifier improved following combining three widely used ML models (e.g., RFE-SVM, EmbeddingLR-LR, and RFE-AdaBoost). Decision curve analysis showed the merits of the ensemble classifiers, as the curves of the top 3 initial classifiers varied a lot in predicting JOA recovery rate in DCM patients.

CONCLUSIONS

The ensemble classifiers successfully predict the JOA recovery rate in DCM patients, which showed a high potential for assisting physicians in managing DCM patients and making full use of medical resources.

Relationship between spinal alignment and functional disability after thoracolumbar spinal fractures: A systematic review

Background

Thoracolumbar spinal fractures (TLSF) can cause pain, neurological deficits, and functional disability. Operative treatments aim to preserve neurological function, improve functional status, and restore spinal alignment and stability. In this review, we evaluate the relationship between spinal alignment and functional impairment in patients with TLSF.

Methods

We performed a systematic review in accordance with the PRISMA guidelines to identify full-text articles that evaluate the correlation between spinal alignment and functional outcomes of TLSF. The artificial intelligence software Rayyan assisted the screening process. Functional outcomes referred to activity/disability, quality of life, and pain scores, as well as return to work metrics. Radiological assessments included were vertebral compression angle, Cobb and Gardner angles, sagittal vertical axis, pelvic incidence, and pelvic tilt. Statistical analyses were performed for the data provided by articles using the SPSS v24.

Results

Of 1,616 articles reviewed, 6 were included for final analysis. Only 1 study primarily addressed the effects of spinopelvic parameters and functional outcomes. Four studies correlated Cobb angles with functional outcome, while 3 others compared vertebral compression angles with functional outcomes. Outcomes were assessed using work status or a combination of VAS pain and spine score, ODI, SF-36, and RMDQ-24. Neither the analysis done within the articles, nor the one made with the raw data provided by them, showed a significant correlation between the radiological measurements assessed at time of injury and final functional outcomes.

Conclusions

A correlation between the assessed spinal radiological measurements assessed with the functional outcomes of TLSF was not found in this review. Further well-designed prospective studies are necessary to evaluate spinal alignment measurements in TLSF with functional outcomes.

Developing Machine Learning-Based Predictive Models for Hallux Valgus Recurrence Based on Measurements From Radiographs

BACKGROUND

Machine learning (ML) is increasingly used to predict the prognosis of numerous diseases. This retrospective analysis aimed to develop a prediction model using ML algorithms and to identify predictors associated with the recurrence of hallux valgus (HV) following surgery.

METHODS

A total of 198 symptomatic feet that underwent chevron osteotomy combined with a distal soft tissue procedure were enrolled and analyzed from 2 independent medical centers. The feet were grouped according to nonrecurrence or recurrence based on 1-year follow-up outcomes. Preoperative weightbearing radiographs and immediate postoperative nonweightbearing radiographs were obtained for each HV foot. Radiographic measurements (eg, HV angle and intermetatarsal angle) were acquired and used for ML model training. A total of 9 commonly used ML models were trained on the data obtained from one institute (108 feet), and tested on the other data set from another independent institute (90 feet) for external validation. Optimal feature sets for each model were identified based on a 2000-resample bootstrap-based internal validation via an exhaustive search. The performance of each model was then tested on the external validation set. The area under the curve (AUC), classification accuracy, sensitivity, and specificity of each model were calculated to evaluate the performance of each model.

RESULTS

The support vector machine (SVM) model showed the highest predictive accuracy compared to other methods, with an AUC of 0.88 and an accuracy of 75.6%. Preoperative hallux valgus angle, tibial sesamoid position, postoperative intermetatarsal angle, and postoperative tibial sesamoid position were identified as the most selected features by several ML models.

CONCLUSION

ML classifiers such as SVM could predict the recurrence of HV (an HVA >20 degrees) at a 1-year follow-up while identifying associated predictors in a multivariate manner. This study holds the potential for foot and ankle surgeons to effectively identify individuals at higher risk of HV recurrence postsurgery.

Predictive modeling of ambulatory outcomes after spinal cord injury using machine learning

STUDY DESIGN

Retrospective multi-site cohort study.

OBJECTIVES

To develop an accurate machine learning predictive model using predictor variables from the acute rehabilitation period to determine ambulatory status in spinal cord injury (SCI) one year post injury.

SETTING

Model SCI System (SCIMS) database between January 2000 and May 2019.

METHODS

Retrospective cohort study using data that were previously collected as part of the SCI Model System (SCIMS) database. A total of 4523 patients were analyzed comparing traditional models (van Middendorp and Hicks) compared to machine learning algorithms including Elastic Net Penalized Logistic Regression (ENPLR), Gradient Boosted Machine (GBM), and Artificial Neural Networks (ANN).

RESULTS

Compared with GBM and ANN, ENPLR was determined to be the preferred model based on predictive accuracy metrics, calibration, and variable selection. The primary metric to judge discrimination was the area under the receiver operating characteristic curve (AUC). When compared to the van Middendorp all patients (0.916), ASIA A and D (0.951) and ASIA B and C (0.775) and Hicks all patients (0.89), ASIA A and D (0.934) and ASIA B and C (0.775), ENPLR demonstrated improved AUC for all patients (0.931), ASIA A and D (0.965) ASIA B and C (0.803).

CONCLUSIONS

Utilizing artificial intelligence and machine learning methods are feasible for accurately classifying outcomes in SCI and may provide improved sensitivity in identifying which individuals are less likely to ambulate and may benefit from augmentative strategies, such as neuromodulation. Future directions should include the use of additional variables to further refine these models.

Orthopedic Frailty Score and adverse outcomes in patients with surgically managed isolated traumatic spinal injury

Background

With an aging global population, the prevalence of frailty in patients with traumatic spinal injury (TSI) is steadily increasing. The aim of the current study is to evaluate the utility of the Orthopedic Frailty Score (OFS) in assessing the risk of adverse outcomes in patients with isolated TSI requiring surgery, with the hypothesis that frailer patients suffer from a disproportionately increased risk of these outcomes.

Methods

The Trauma Quality Improvement Program database was queried for all adult patients (18 years or older) who suffered an isolated TSI due to blunt force trauma, between 2013 and 2019, and underwent spine surgery. Patients were categorized as non-frail (OFS 0), pre-frail (OFS 1), or frail (OFS ≥2). The association between the OFS and in-hospital mortality, complications, and failure to rescue (FTR) was determined using Poisson regression models, adjusted for potential confounding.

Results

A total of 43 768 patients were included in the current investigation. After adjusting for confounding, frailty was associated with a more than doubling in the risk of in-hospital mortality (adjusted incidence rate ratio (IRR) (95% CI): 2.53 (2.04 to 3.12), p<0.001), a 25% higher overall risk of complications (adjusted IRR (95% CI): 1.25 (1.02 to 1.54), p=0.032), a doubling in the risk of FTR (adjusted IRR (95% CI): 2.00 (1.39 to 2.90), p<0.001), and a 10% increase in the risk of intensive care unit admission (adjusted IRR (95% CI): 1.10 (1.04 to 1.15), p=0.004), compared with non-frail patients.

Conclusion

The findings indicate that the OFS could be an effective method for identifying frail patients with TSIs who are at a disproportionate risk of adverse events.

Level of evidence

Level III.

Prediction of gait recovery using machine learning algorithms in patients with spinal cord injury

With advances in artificial intelligence, machine learning (ML) has been widely applied to predict functional outcomes in clinical medicine. However, there has been no attempt to predict walking ability after spinal cord injury (SCI) based on ML. In this situation, the main purpose of this study was to predict gait recovery after SCI at discharge from an acute rehabilitation facility using various ML algorithms. In addition, we explored important variables that were related to the prognosis. Finally, we attempted to suggest an ML-based decision support system (DSS) for predicting gait recovery after SCI. Data were collected retrospectively from patients with SCI admitted to an acute rehabilitation facility between June 2008 to December 2021. Linear regression analysis and ML algorithms (random forest [RF], decision tree [DT], and support vector machine) were used to predict the functional ambulation category at the time of discharge (FAC_DC) in patients with traumatic or non-traumatic SCI (n = 353). The independent variables were age, sex, duration of acute care and rehabilitation, comorbidities, neurological information entered into the International Standards for Neurological Classification of SCI worksheet, and somatosensory-evoked potentials at the time of admission to the acute rehabilitation facility. In addition, the importance of variables and DT-based DSS for FAC_DC was analyzed. As a result, RF and DT accurately predicted the FAC_DC measured by the root mean squared error. The root mean squared error of RF and the DT were 1.09 and 1.24 for all participants, 1.20 and 1.06 for those with trauma, and 1.12 and 1.03 for those with non-trauma, respectively. In the analysis of important variables, the initial FAC was found to be the most influential factor in all groups. In addition, we could provide a simple DSS based on strong predictors such as the initial FAC, American Spinal Injury Association Impairment Scale grades, and neurological level of injury. In conclusion, we provide that ML can accurately predict gait recovery after SCI for the first time. By focusing on important variables and DSS, we can guide early prognosis and establish personalized rehabilitation strategies in acute rehabilitation hospitals.

Early Prognostication of Critical Patients With Spinal Cord Injury: A Machine Learning Study With 1485 Cases

STUDY DESIGN

A retrospective case-series.

OBJECTIVE

The study aims to use machine learning to predict the discharge destination of spinal cord injury (SCI) patients in the intensive care unit.

SUMMARY OF BACKGROUND DATA

Prognostication following SCI is vital, especially for critical patients who need intensive care.

PATIENTS AND METHODS

Clinical data of patients diagnosed with SCI were extracted from a publicly available intensive care unit database. The first recorded data of the included patients were used to develop a total of 98 machine learning classifiers, seeking to predict discharge destination (eg, death, further medical care, home, etc.). The microaverage area under the curve (AUC) was the main indicator to assess discrimination. The best average-AUC classifier and the best death-sensitivity classifier were integrated into an ensemble classifier. The discrimination of the ensemble classifier was compared with top death-sensitivity classifiers and top average-AUC classifiers. In addition, prediction consistency and clinical utility were also assessed.

RESULTS

A total of 1485 SCI patients were included. The ensemble classifier had a microaverage AUC of 0.851, which was only slightly inferior to the best average-AUC classifier ( P =0.10). The best average-AUC classifier death sensitivity was much lower than that of the ensemble classifier. The ensemble classifier had a death sensitivity of 0.452, which was inferior to the top 8 death-sensitivity classifiers, whose microaverage AUC were inferior to the ensemble classifier ( P <0.05). In addition, the ensemble classifier demonstrated a comparable Brier score and superior net benefit in the DCA when compared with the performance of the origin classifiers.

CONCLUSIONS

The ensemble classifier shows an overall superior performance in predicting discharge destination, considering discrimination ability, prediction consistency, and clinical utility. This classifier system may aid in the clinical management of critical SCI patients in the early phase following injury.

LEVEL OF EVIDENCE

Level 3.

SCIseg: Automatic Segmentation of T2-weighted Intramedullary Lesions in Spinal Cord Injury

Purpose

To develop a deep learning tool for the automatic segmentation of T2-weighted intramedullary lesions in spinal cord injury (SCI).

Material and Methods

This retrospective study included a cohort of SCI patients from three sites enrolled between July 2002 and February 2023. A deep learning model, SCIseg, was trained in a three-phase process involving active learning for the automatic segmentation of intramedullary SCI lesions and the spinal cord. The data consisted of T2-weighted MRI acquired using different scanner manufacturers with heterogeneous image resolutions (isotropic/anisotropic), orientations (axial/sagittal), lesion etiologies (traumatic/ischemic/hemorrhagic) and lesions spread across the cervical, thoracic and lumbar spine. The segmentations from the proposed model were visually and quantitatively compared with other open-source baselines. Wilcoxon signed-rank test was used to compare quantitative MRI biomarkers (lesion volume, lesion length, and maximal axial damage ratio) computed from manual lesion masks and those obtained automatically with SCIseg predictions.

Results

MRI data from 191 SCI patients (mean age, 48.1 years ± 17.9 [SD]; 142 males) were used for model training and evaluation. SCIseg achieved the best segmentation performance for both the cord and lesions. There was no statistically significant difference between lesion length and maximal axial damage ratio computed from manually annotated lesions and those obtained using SCIseg.

Conclusion

Automatic segmentation of intramedullary lesions commonly seen in SCI replaces the tedious manual annotation process and enables the extraction of relevant lesion morphometrics in large cohorts. The proposed model segments lesions across different etiologies, scanner manufacturers, and heterogeneous image resolutions. SCIseg is open-source and accessible through the Spinal Cord Toolbox.

Multivariable Prediction Models for Traumatic Spinal Cord Injury: A Systematic Review

Background

Traumatic spinal cord injuries (TSCI) greatly affect the lives of patients and their families. Prognostication may improve treatment strategies, health care resource allocation, and counseling. Multivariable clinical prediction models (CPMs) for prognosis are tools that can estimate an absolute risk or probability that an outcome will occur.

Objectives

We sought to systematically review the existing literature on CPMs for TSCI and critically examine the predictor selection methods used.

Methods

We searched MEDLINE, PubMed, Embase, Scopus, and IEEE for English peer-reviewed studies and relevant references that developed multivariable CPMs to prognosticate patient-centered outcomes in adults with TSCI. Using narrative synthesis, we summarized the characteristics of the included studies and their CPMs, focusing on the predictor selection process.

Results

We screened 663 titles and abstracts; of these, 21 full-text studies (2009-2020) consisting of 33 distinct CPMs were included. The data analysis domain was most commonly at a high risk of bias when assessed for methodological quality. Model presentation formats were inconsistently included with published CPMs; only two studies followed established guidelines for transparent reporting of multivariable prediction models. Authors frequently cited previous literature for their initial selection of predictors, and stepwise selection was the most frequent predictor selection method during modelling.

Conclusion

Prediction modelling studies for TSCI serve clinicians who counsel patients, researchers aiming to risk-stratify participants for clinical trials, and patients coping with their injury. Poor methodological rigor in data analysis, inconsistent transparent reporting, and a lack of model presentation formats are vital areas for improvement in TSCI CPM research.

Functional Outcome Prediction After Spinal Cord Injury Using Ensemble Machine Learning

OBJECTIVES

To establish a machine learning model to predict functional outcomes after SCI with Spinal Cord Independence Measure (SCIM) using features present at the time of rehabilitation admission.

STUDY DESIGN

A retrospective, single-center study. The following data were collected from the medical charts: age, sex, acute length of stay (LOS), level of injury, American Spinal Injury Association Impairment Scale (AIS), motor scores of each key muscle, Upper Extremity Motor Score (UEMS), Lower Extremity Motor Score (LEMS), SCIM total scores, and subtotal scores on admission and discharge. Based on the multivariate linear regression analysis, age, acute LOS, UEMS, LEMS, and SCIM subtotal scores were selected as features for machine learning algorithms. Random forest, support vector machine, neural network, and gradient boosting were used as the base models and combined using ridge regression as a metamodel.

SETTING

A spinal center in Tokyo, Japan.

PARTICIPANTS

Participants were individuals with SCI admitted to our hospital from March 2016 to October 2021 for the first rehabilitation after the injury. They were divided into 2 groups: training (n=140) and testing (n=70).

INTERVENTIONS

Not applicable.

MAIN OUTCOME MEASURES

The root-mean-square error (RMSE), R2, and Mean Absolute Error (MAE) were used as accuracy measures.

RESULTS

RMSE, R2, and MAE of the meta-model using the testing group were 9.7453, 0.8835, and 7.4743, respectively, outperforming any other single base model.

CONCLUSIONS

Our study revealed that functional prognostication could be achieved using machine-learning methods with features present at the time of rehabilitation admission. Goals can be set at the beginning of rehabilitation. Moreover, our model can be used to evaluate advanced medical treatments, such as regenerative medicine.

Decision Tree Analysis Accurately Predicts Discharge Destination After Spinal Cord Injury Rehabilitation

OBJECTIVES

To predict discharge destination after spinal cord injury (SCI) rehabilitation.

STUDY DESIGN

A retrospective, single-center study. We collected the following data from medical charts: age, sex, living arrangement before injury, acute length of stay (LOS), level of injury on admission, American Spinal Injury Association Impairment Scale (AIS) on admission, Upper Extremity Motor Score (UEMS) on admission, Lower Extremity Motor Score on admission (LEMS), Spinal Cord Independence Measure (SCIM) scores on admission and discharge, and discharge destination. A decision tree algorithm was used to establish prediction models in a train-test split manner using features on admission or discharge.

SETTING

A spinal center in Tokyo, Japan.

PARTICIPANTS

Participants were individuals with SCI admitted to our hospital from March 2016 to October 2021 for the first rehabilitation after the injury. The study included 210 participants divided into 2 groups: training (n=140) and testing (n=70). Random sampling without replacement was used.

INTERVENTIONS

Not applicable.

MAIN OUTCOME MEASURES

Prediction accuracy was evaluated with sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area under the receiver operating curve (AUC).

RESULTS

AIS was significantly different between the groups. The prediction model using total SCIM scores on discharge (D-Classification and Regression Tree [CART]) revealed that a cut-off value of 40 accurately predicted the discharge destination. In contrast, the prediction model using features on admission (A-CART) revealed that subtotal SCIM mobility scores of 5, age of 74 years, and UEMS of 23 were significant predictors. Sensitivity, specificity, PPV, NPV, and AUC of D-CART and A-CART were 0.837, 0.810, 0.911, 0.680, and 0.832 and 0.857, 0.810, 0.913, 0.708, and 0.869, respectively.

CONCLUSIONS

D-CART and A-CART showed comparable prediction accuracies. This suggests that, even during the early stages of rehabilitation, it is possible to predict the discharge destination.

Developing predictive models for surgical outcomes in patients with degenerative cervical myelopathy: a comparison of statistical and machine learning approaches

BACKGROUND CONTEXT

Machine learning (ML) is widely used to predict the prognosis of numerous diseases.

PURPOSE

This retrospective analysis aimed to develop a prognostic prediction model using ML algorithms and identify predictors associated with poor surgical outcomes in patients with degenerative cervical myelopathy (DCM).

STUDY DESIGN

A retrospective study.

PATIENT SAMPLE

A total of 406 symptomatic DCM patients who underwent surgical decompression were enrolled and analyzed from three independent medical centers.

OUTCOME MEASURES

We calculated the area under the curve (AUC), classification accuracy, sensitivity, and specificity of each model.

METHODS

The Japanese Orthopedic Association (JOA) score was obtained before and 1 year following decompression surgery, and patients were grouped into good and poor outcome groups based on a cut-off value of 60% based on a previous study. Two datasets were fused for training, 1 dataset was held out as an external validation set. Optimal feature-subset and hyperparameters for each model were adjusted based on a 2,000-resample bootstrap-based internal validation via exhaustive search and grid search. The performance of each model was then tested on the external validation set.

RESULTS

The Support Vector Machine (SVM) model showed the highest predictive accuracy compared to other methods, with an AUC of 0.82 and an accuracy of 75.7%. Age, sex, disease duration, and preoperative JOA score were identified as the most commonly selected features by both the ML and statistical models. Grid search optimization for hyperparameters successfully enhanced the predictive performance of each ML model, and the SVM model still had the best performance with an AUC of 0.93 and an accuracy of 86.4%.

CONCLUSIONS

Overall, the study demonstrated that ML classifiers such as SVM can effectively predict surgical outcomes for patients with DCM while identifying associated predictors in a multivariate manner.

Enabling knowledge translation: implementation of a web-based tool for independent walking prediction after traumatic spinal cord injury

Introduction

Several clinical prediction rules (CPRs) have been published, but few are easily accessible or convenient for clinicians to use in practice. We aimed to develop, implement, and describe the process of building a web-based CPR for predicting independent walking 1-year after a traumatic spinal cord injury (TSCI).

Methods

Using the published and validated CPR, a front-end web application called "Ambulation" was built using HyperText Markup Language (HTML), Cascading Style Sheets (CSS), and JavaScript. A survey was created using QualtricsXM Software to gather insights on the application's usability and user experience. Website activity was monitored using Google Analytics. Ambulation was developed with a core team of seven clinicians and researchers. To refine the app's content, website design, and utility, 20 professionals from different disciplines, including persons with lived experience, were consulted.

Results

After 11 revisions, Ambulation was uploaded onto a unique web domain and launched (www.ambulation.ca) as a pilot with 30 clinicians (surgeons, physiatrists, and physiotherapists). The website consists of five web pages: Home, Calculation, Team, Contact, and Privacy Policy. Responses from the user survey (n = 6) were positive and provided insight into the usability of the tool and its clinical utility (e.g., helpful in discharge planning and rehabilitation), and the overall face validity of the CPR. Since its public release on February 7, 2022, to February 28, 2023, Ambulation had 594 total users, 565 (95.1%) new users, 26 (4.4%) returning users, 363 (61.1%) engaged sessions (i.e., the number of sessions that lasted 10 seconds/longer, had one/more conversion events e.g., performing the calculation, or two/more page or screen views), and the majority of the users originating from the United States (39.9%) and Canada (38.2%).

Discussion

Ambulation is a CPR for predicting independent walking 1-year after TSCI and it can assist frontline clinicians with clinical decision-making (e.g., time to surgery or rehabilitation plan), patient education and goal setting soon after injury. This tool is an example of adapting a validated CPR for independent walking into an easily accessible and usable web-based tool for use in clinical practice. This study may help inform how other CPRs can be adopted into clinical practice.

Emerging Technologies within Spine Surgery

New innovations within spine surgery continue to propel the field forward. These technologies improve surgeons' understanding of their patients and allow them to optimize treatment planning both in the operating room and clinic. Additionally, changes in the implants and surgeon practice habits continue to evolve secondary to emerging biomaterials and device design. With ongoing advancements, patients can expect enhanced preoperative decision-making, improved patient outcomes, and better intraoperative execution. Additionally, these changes may decrease many of the most common complications following spine surgery in order to reduce morbidity, mortality, and the need for reoperation. This article reviews some of these technological advancements and how they are projected to impact the field. As the field continues to advance, it is vital that practitioners remain knowledgeable of these changes in order to provide the most effective treatment possible.

Identification of injured elements in computational models of spinal cord injury using machine learning. ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY. IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY. ANNUAL INTERNATIONAL CONFERENCE 2023;2023:1-4. [PMID: 38082848 DOI: 10.1109/embc40787.2023.10340243] [Citation(s) in RCA: 0] [Impact Index Per Article: 0]

The purpose of this study was to use machine learning (ML) algorithms to identify tissue damage based on the mechanical outputs of computational models of spinal cord injury (SCI). Three datasets corresponding to gray matter, white matter, and the combination of gray and white matter tissues were used to train the models. These datasets were built from the comparison of histological images taken from SCI experiments in non-human primates and corresponding subject-specific finite element (FE) models. Four ML algorithms were evaluated and compared using cross-validation and the area under the receiver operating characteristic curve (AUC). After hyperparameter tuning, the AUC mean values for the algorithms ranged between 0.79 and 0.82, with a standard deviation no greater than 0.02. The findings of this study also showed that k-nearest neighbors and logistic regression algorithms were better at identifying injured elements than support vector machines and decision trees. Additionally, depending on the evaluated dataset, the mean values of other performance metrics, such as precision and recall, varied between algorithms. These initial results suggest that different algorithms might be more sensitive to the skewed distribution of classes in the studied datasets, and that identifying damage independently or simultaneously in the gray and white matter tissues might require a better definition of relevant features and the use of different ML algorithms. These approaches will contribute to improving the current understanding of the relationship between mechanical loading and tissue damage during SCI and will have implications for the development of prevention strategies for this condition.Clinical Relevance- Linking FE model predictions of mechanical loading to tissue damage is an essential step for FE models to provide clinically relevant information. Combined with imaging technologies, these models can provide useful insights to predict the extent of damage in animal subjects and guide the decision-making process during treatment planning.

Diffusion basis spectrum imaging predicts long-term clinical outcomes following surgery in cervical spondylotic myelopathy

BACKGROUND CONTEXT

A major shortcoming in improving care for cervical spondylotic myelopathy (CSM) patients is the lack of robust quantitative imaging tools to guide surgical decision-making. Diffusion basis spectrum imaging (DBSI), an advanced diffusion-weighted MRI technique, provides objective assessments of white matter tract integrity that may help prognosticate outcomes in patients undergoing surgery for CSM.

PURPOSE

To examine the ability of DBSI to predict clinically important CSM outcome measures at 2-years follow-up.

STUDY DESIGN/SETTING

Prospective cohort study.

PATIENT SAMPLE

Patients undergoing decompressive cervical surgery for CSM.

OUTCOME MEASURES

Neurofunctional status was assessed by the mJOA, MDI, and DASH. Quality-of-life was measured by the SF-36 PCS and SF-36 MCS. The NDI evaluated self-reported neck pain, and patient satisfaction was assessed by the NASS satisfaction index.

METHODS

Fifty CSM patients who underwent cervical decompressive surgery were enrolled. Preoperative DBSI metrics assessed white matter tract integrity through fractional anisotropy, fiber fraction, axial diffusivity, and radial diffusivity. To evaluate extra-axonal diffusion, DBSI measures restricted and nonrestricted fractions. Patient-reported outcome measures were evaluated preoperatively and up to 2-years follow-up. Support vector machine classification algorithms were used to predict surgical outcomes at 2-years follow-up. Specifically, three feature sets were built for each of the seven clinical outcome measures (eg, mJOA), including clinical only, DBSI only, and combined feature sets.

RESULTS

Twenty-seven mild (mJOA 15-17), 12 moderate (12-14) and 11 severe (0-11) CSM patients were enrolled. Twenty-four (60%) patients underwent anterior decompressive surgery compared with 16 (40%) posterior approaches. The mean (SD) follow-up was 23.2 (5.6, range 6.1-32.8) months. Feature sets built on combined data (ie, clinical+DBSI metrics) performed significantly better for all outcome measures compared with those only including clinical or DBSI data. When predicting improvement in the mJOA, the clinically driven feature set had an accuracy of 61.9 [61.6, 62.5], compared with 78.6 [78.4, 79.2] in the DBSI feature set, and 90.5 [90.2, 90.8] in the combined feature set.

CONCLUSIONS

When combined with key clinical covariates, preoperative DBSI metrics predicted improvement after surgical decompression for CSM with high accuracy for multiple outcome measures. These results suggest that DBSI may serve as a noninvasive imaging biomarker for CSM valuable in guiding patient selection and informing preoperative counseling.

LEVEL OF EVIDENCE

II.

The role of Artificial intelligence in the assessment of the spine and spinal cord

Artificial intelligence (AI) application development is underway in all areas of radiology where many promising tools are focused on the spine and spinal cord. In the past decade, multiple spine AI algorithms have been created based on radiographs, computed tomography, and magnetic resonance imaging. These algorithms have wide-ranging purposes including automatic labeling of vertebral levels, automated description of disc degenerative changes, detection and classification of spine trauma, identification of osseous lesions, and the assessment of cord pathology. The overarching goals for these algorithms include improved patient throughput, reducing radiologist workload burden, and improving diagnostic accuracy. There are several pre-requisite tasks required in order to achieve these goals, such as automatic image segmentation, facilitating image acquisition and postprocessing. In this narrative review, we discuss some of the important imaging AI solutions that have been developed for the assessment of the spine and spinal cord. We focus on their practical applications and briefly discuss some key requirements for the successful integration of these tools into practice. The potential impact of AI in the imaging assessment of the spine and cord is vast and promises to provide broad reaching improvements for clinicians, radiologists, and patients alike.

Predictive factors for degenerative lumbar spinal stenosis: a model obtained from a machine learning algorithm technique

BACKGROUND

Degenerative lumbar spinal stenosis (DLSS) is the most common spine disease in the elderly population. It is usually associated with lumbar spine joints/or ligaments degeneration. Machine learning technique is an exclusive method for handling big data analysis; however, the development of this method for spine pathology is rare. This study aims to detect the essential variables that predict the development of symptomatic DLSS using the random forest of machine learning (ML) algorithms technique.

METHODS

A retrospective study with two groups of individuals. The first included 165 with symptomatic DLSS (sex ratio 80 M/85F), and the second included 180 individuals from the general population (sex ratio: 90 M/90F) without lumbar spinal stenosis symptoms. Lumbar spine measurements such as vertebral or spinal canal diameters from L1 to S1 were conducted on computerized tomography (CT) images. Demographic and health data of all the participants (e.g., body mass index and diabetes mellitus) were also recorded.

RESULTS

The decision tree model of ML demonstrate that the anteroposterior diameter of the bony canal at L5 (males) and L4 (females) levels have the greatest stimulus for symptomatic DLSS (scores of 1 and 0.938). In addition, combination of these variables with other lumbar spine features is mandatory for developing the DLSS.

CONCLUSIONS

Our results indicate that combination of lumbar spine characteristics such as bony canal and vertebral body dimensions rather than the presence of a sole variable is highly associated with symptomatic DLSS onset.

Risk Analysis Index and Its Recalibrated Version Predict Postoperative Outcomes Better Than 5-Factor Modified Frailty Index in Traumatic Spinal Injury

OBJECTIVE

To assess the discriminative ability of the Risk Analysis Index-administrative (RAI-A) and its recalibrated version (RAI-Rev), compared to the 5-factor modified frailty index (mFI-5), in predicting postoperative outcomes in patients undergoing surgical intervention for traumatic spine injuries (TSIs).

METHODS

The Current Procedural Terminology (CPT) and International Classification of Disease-9 (ICD-9) and ICD-10 codes were used to identify patients ≥ 18 years who underwent surgical intervention for TSI from National Surgical Quality Improvement Program (ACS-NSQIP) database 2015-2019 (n = 6,571). Multivariate analysis and receiver operating characteristic (ROC) curve analysis were conducted to evaluate the comparative discriminative ability of RAI-Rev, RAI-A, and mFI-5 for 30-day postoperative outcomes.

RESULTS

Multivariate regression analysis showed that with all 3 frailty scores, increasing frailty tiers resulted in worse postoperative outcomes, and patients identified as frail and severely frail using RAI-Rev and RAI-A had the highest odds of poor outcomes. In the ROC curve/C-statistics analysis for prediction of 30-day mortality and morbidity, both RAI-Rev and RAI-A outperformed mFI-5, and for many outcomes, RAI-Rev showed better discriminative performance compared to RAI-A, including mortality (p = 0.0043, DeLong test), extended length of stay (p = 0.0042), readmission (p < 0.0001), reoperation (p = 0.0175), and nonhome discharge (p < 0.0001).

CONCLUSION

Both RAI-Rev and RAI-A performed better than mFI-5, and RAI-Rev was superior to RAI-A in predicting postoperative mortality and morbidity in TSI patients. RAI-based frailty indices can be used in preoperative risk assessment of spinal trauma patients.

The application of artificial intelligence in spine surgery

Due to its obvious advantages in processing big data and image information, the combination of artificial intelligence and medical care may profoundly change medical practice and promote the gradual transition from traditional clinical care to precision medicine mode. In this artical, we reviewed the relevant literatures and found that artificial intelligence was widely used in spine surgery. The application scenarios included etiology, diagnosis, treatment, postoperative prognosis and decision support systems of spinal diseases. The shift to artificial intelligence model in medicine constantly improved the level of doctors' diagnosis and treatment and the development of orthopedics.

Advances in monitoring for acute spinal cord injury: a narrative review of current literature

Spinal cord injury (SCI) is a devastating condition that affects about 17,000 individuals every year in the United States, with approximately 294,000 people living with the ramifications of the initial injury. After the initial primary injury, SCI has a secondary phase during which the spinal cord sustains further injury due to ischemia, excitotoxicity, immune-mediated damage, mitochondrial dysfunction, apoptosis, and oxidative stress. The multifaceted injury progression process requires a sophisticated injury-monitoring technique for an accurate assessment of SCI patients. In this narrative review, we discuss SCI monitoring modalities, including pressure probes and catheters, micro dialysis, electrophysiologic measures, biomarkers, and imaging studies. The optimal next-generation injury monitoring setup should include multiple modalities and should integrate the data to produce a final simplified assessment of the injury and determine markers of intervention to improve patient outcomes.

Machine Learning-based Prediction of Prolonged Intensive Care Unit Stay for Critical Patients with Spinal Cord Injury

STUDY DESIGN

A retrospective cohort study.

OBJECTIVE

The objective of the study was to develop machine-learning (ML) classifiers for predicting prolonged intensive care unit (ICU)-stay and prolonged hospital-stay for critical patients with spinal cord injury (SCI).

SUMMARY OF BACKGROUND DATA

Critical patients with SCI in ICU need more attention. SCI patients with prolonged stay in ICU usually occupy vast medical resources and hinder the rehabilitation deployment.

METHODS

A total of 1599 critical patients with SCI were included in the study and labeled with prolonged stay or normal stay. All data were extracted from the eICU Collaborative Research Database and the Medical Information Mart for Intensive Care III-IV Database. The extracted data were randomly divided into training, validation and testing (6:2:2) subdatasets. A total of 91 initial ML classifiers were developed, and the top three initial classifiers with the best performance were further stacked into an ensemble classifier with logistic regressor. The area under the curve (AUC) was the main indicator to assess the prediction performance of all classifiers. The primary predicting outcome was prolonged ICU-stay, while the secondary predicting outcome was prolonged hospital-stay.

RESULTS

In predicting prolonged ICU-stay, the AUC of the ensemble classifier was 0.864 ± 0.021 in the three-time five-fold cross-validation and 0.802 in the independent testing. In predicting prolonged hospital-stay, the AUC of the ensemble classifier was 0.815 ± 0.037 in the three-time five-fold cross-validation and 0.799 in the independent testing. Decision curve analysis showed the merits of the ensemble classifiers, as the curves of the top three initial classifiers varied a lot in either predicting prolonged ICU-stay or discriminating prolonged hospital-stay.

CONCLUSION

The ensemble classifiers successfully predict the prolonged ICU-stay and the prolonged hospital-stay, which showed a high potential of assisting physicians in managing SCI patients in ICU and make full use of medical resources.Level of Evidence: 3.

Expert-augmented automated machine learning optimizes hemodynamic predictors of spinal cord injury outcome

Artificial intelligence and machine learning (AI/ML) is becoming increasingly more accessible to biomedical researchers with significant potential to transform biomedicine through optimization of highly-accurate predictive models and enabling better understanding of disease biology. Automated machine learning (AutoML) in particular is positioned to democratize artificial intelligence (AI) by reducing the amount of human input and ML expertise needed. However, successful translation of AI/ML in biomedicine requires moving beyond optimizing only for prediction accuracy and towards establishing reproducible clinical and biological inferences. This is especially challenging for clinical studies on rare disorders where the smaller patient cohorts and corresponding sample size is an obstacle for reproducible modeling results. Here, we present a model-agnostic framework to reinforce AutoML using strategies and tools of explainable and reproducible AI, including novel metrics to assess model reproducibility. The framework enables clinicians to interpret AutoML-generated models for clinical and biological verifiability and consequently integrate domain expertise during model development. We applied the framework towards spinal cord injury prognostication to optimize the intraoperative hemodynamic range during injury-related surgery and additionally identified a strong detrimental relationship between intraoperative hypertension and patient outcome. Furthermore, our analysis captured how evolving clinical practices such as faster time-to-surgery and blood pressure management affect clinical model development. Altogether, we illustrate how expert-augmented AutoML improves inferential reproducibility for biomedical discovery and can ultimately build trust in AI processes towards effective clinical integration.

Automated prediction of the Thoracolumbar Injury Classification and Severity Score from CT using a novel deep learning algorithm

OBJECTIVE

Damage to the thoracolumbar spine can confer significant morbidity and mortality. The Thoracolumbar Injury Classification and Severity Score (TLICS) is used to categorize injuries and determine patients at risk of spinal instability for whom surgical intervention is warranted. However, calculating this score can constitute a bottleneck in triaging and treating patients, as it relies on multiple imaging studies and a neurological examination. Therefore, the authors sought to develop and validate a deep learning model that can automatically categorize vertebral morphology and determine posterior ligamentous complex (PLC) integrity, two critical features of TLICS, using only CT scans.

METHODS

All patients who underwent neurosurgical consultation for traumatic spine injury or degenerative pathology resulting in spine injury at a single tertiary center from January 2018 to December 2019 were retrospectively evaluated for inclusion. The morphology of injury and integrity of the PLC were categorized on CT scans. A state-of-the-art object detection region-based convolutional neural network (R-CNN), Faster R-CNN, was leveraged to predict both vertebral locations and the corresponding TLICS. The network was trained with patient CT scans, manually labeled vertebral bounding boxes, TLICS morphology, and PLC annotations, thus allowing the model to output the location of vertebrae, categorize their morphology, and determine the status of PLC integrity.

RESULTS

A total of 111 patients were included (mean ± SD age 62 ± 20 years) with a total of 129 separate injury classifications. Vertebral localization and PLC integrity classification achieved Dice scores of 0.92 and 0.88, respectively. Binary classification between noninjured and injured morphological scores demonstrated 95.1% accuracy. TLICS morphology accuracy, the true positive rate, and positive injury mismatch classification rate were 86.3%, 76.2%, and 22.7%, respectively. Classification accuracy between no injury and suspected PLC injury was 86.8%, while true positive, false negative, and false positive rates were 90.0%, 10.0%, and 21.8%, respectively.

CONCLUSIONS

In this study, the authors demonstrate a novel deep learning method to automatically predict injury morphology and PLC disruption with high accuracy. This model may streamline and improve diagnostic decision support for patients with thoracolumbar spinal trauma.

Development of a clinical prediction rule for patients with cervical spinal cord injury who have difficulty in obtaining independent living

BACKGROUND CONTEXT

A simple and easy to use clinical prediction rule (CPR) to detect patients with a cervical spinal cord injury (SCI) who would have difficulty in obtaining independent living status is vital for providing the optimal rehabilitation and education in both care recipients and caregivers. A machine learning approach was recently applied to the field of rehabilitation and has the possibility to develop an accurate and useful CPR.

PURPOSE

The aim of this study was to develop and assess a CPR using a decision tree algorithm for predicting which patients with a cervical SCI would have difficulty in obtaining an independent living.

STUDY DESIGN

The present study was a cohort study.

PATIENT SAMPLE

In the present study, the data was obtained from the nationwide Japan Rehabilitation Database (JRD). The data on the SCIs was collected from 10 hospitals and the data was collected from the registries obtained between 2005 and 2015. The severity of SCI can vary, and patient prognosis differs depending on the damage site. In this study, the patients with cervical SCI were included.

OUTCOME MEASURES

In this study, the degree of the independent living at discharge was investigated. The degree of the independent living was classified and listed as below: independent in social, independent at home, need care at home, independent at facility, need care at facility. In this study, the independent in social and independent at home were defined as "independent," and the other situations were defined as "non-independent."

METHODS

We performed a classification and regression tree (CART) analysis to develop the CPR to predict whether the cervical SCI patients obtain an independent living at discharge. The area under the curve, the classification accuracy, sensitivity, specificity, and positive predictive value were used for model evaluation.

RESULTS

A total of 4181 patients with SCI were registered in the JRD and the CART analysis was performed for 1282 patients with the cervical SCI. The Functional Independence Measure (FIM) total score and the American Spinal Injury Association impairment scale were identified as the first and second discriminators for predicting the degree of the independence, respectively. Subsequently, the CART model identified FIM eating, the residual function level, and the FIM bed to chair transfer as next discriminators. Each parameter for evaluating the CART model were the area under the curve 0.813, the classification accuracy 78.6%, the sensitivity 80.7%, the specificity 75.1%, and the positive predictive value 84.5%.

CONCLUSIONS

In this study, we developed a clinically useful CPR with moderate accuracy to predict whether the cervical SCI patients obtain independent living at the discharge.

Development of a machine learning algorithm for predicting in-hospital and 1-year mortality after traumatic spinal cord injury

BACKGROUND CONTEXT

Current prognostic tools such as the Injury Severity Score (ISS) that predict mortality following trauma do not adequately consider the unique characteristics of traumatic spinal cord injury (tSCI).

PURPOSE

Our aim was to develop and validate a prognostic tool that can predict mortality following tSCI.

STUDY DESIGN

Retrospective review of a prospective cohort study.

PATIENT SAMPLE

Data was collected from 1245 persons with acute tSCI who were enrolled in the Rick Hansen Spinal Cord Injury Registry between 2004 and 2016.

OUTCOME MEASURES

In-hospital and 1-year mortality following tSCI.

METHODS

Machine learning techniques were used on patient-level data (n=849) to develop the Spinal Cord Injury Risk Score (SCIRS) that can predict mortality based on age, neurological level and completeness of injury, AOSpine classification of spinal column injury morphology, and Abbreviated Injury Scale scores. Validation of the SCIRS was performed by testing its accuracy in an independent validation cohort (n=396) and comparing its performance to the ISS, a measure which is used to predict mortality following general trauma.

RESULTS

For 1-year mortality prediction, the values for the Area Under the Receiver Operating Characteristic Curve (AUC) for the development cohort were 0.84 (standard deviation=0.029) for the SCIRS and 0.55 (0.041) for the ISS. For the validation cohort, AUC values were 0.86 (0.051) for the SCIRS and 0.71 (0.074) for the ISS. For in-hospital mortality, AUC values for the development cohort were 0.87 (0.028) and 0.60 (0.050) for the SCIRS and ISS, respectively. For the validation cohort, AUC values were 0.85 (0.054) for the SCIRS and 0.70 (0.079) for the ISS.

CONCLUSIONS

The SCIRS can predict in-hospital and 1-year mortality following tSCI more accurately than the ISS. The SCIRS can be used in research to reduce bias in estimating parameters and can help adjust for coefficients during model development. Further validation using larger sample sizes and independent datasets is needed to assess its reliability and to evaluate using it as an assessment tool to guide clinical decision-making and discussions with patients and families.

Preoperative Radiological Parameters to Predict Clinical and Radiological Outcomes after Laminoplasty

Many studies have focused on pre-operative sagittal alignment parameters which could predict poor clinical or radiological outcomes after laminoplasty. However, the influx of too many new factors causes confusion. This study reviewed sagittal alignment parameters, predictive of clinical or radiological outcomes, in the literature. Preoperative kyphotic alignment was initially proposed as a predictor of clinical outcomes. The clinical significance of the K-line and K-line variants also has been studied. Sagittal vertical axis, T1 slope (T1s), T1s-cervical lordosis (CL), anterolisthesis, local kyphosis, the longitudinal distance index, and range of motion were proposed to have relationships with clinical outcomes. The relationship between loss of cervical lordosis (LCL) and T1s has been widely studied, but controversy remains. Extension function, the ratio of CL to T1s (CL/T1s), and Sharma classification were recently proposed as LCL predictors. In predicting postoperative kyphosis, T1s cannot predict postoperative kyphosis, but a low CL/T1s ratio was associated with postoperative kyphosis.

Prediction of Worse Functional Status After Surgery for Degenerative Cervical Myelopathy: A Machine Learning Approach

BACKGROUND

Surgical decompression for degenerative cervical myelopathy (DCM) is one of the mainstays of treatment, with generally positive outcomes. However, some patients who undergo surgery for DCM continue to show functional decline.

OBJECTIVE

To use machine learning (ML) algorithms to determine predictors of worsening functional status after surgical intervention for DCM.

METHODS

This is a retrospective analysis of prospectively collected data. A total of 757 patients enrolled in 2 prospective AO Spine clinical studies, who underwent surgical decompression for DCM, were analyzed. The modified Japanese Orthopedic Association (mJOA) score, a marker of functional status, was obtained before and 1 yr postsurgery. The primary outcome measure was the dichotomized change in mJOA at 1 yr according to whether it was negative (worse functional status) or non-negative. After applying an 80:20 training-testing split of the dataset, we trained, optimized, and tested multiple ML algorithms to evaluate algorithm performance and determine predictors of worse mJOA at 1 yr.

RESULTS

The highest-performing ML algorithm was a polynomial support vector machine. This model showed good calibration and discrimination on the testing data, with an area under the receiver operating characteristic curve of 0.834 (accuracy: 74.3%, sensitivity: 88.2%, specificity: 72.4%). Important predictors of functional decline at 1 yr included initial mJOA, male gender, duration of myelopathy, and the presence of comorbidities.

CONCLUSION

The reasons for worse mJOA are frequently multifactorial (eg, adjacent segment degeneration, tandem lumbar stenosis, ongoing neuroinflammatory processes in the cord). This study successfully used ML to predict worse functional status after surgery for DCM and to determine associated predictors.

Toward Improving the Prediction of Functional Ambulation After Spinal Cord Injury Though the Inclusion of Limb Accelerations During Sleep and Personal Factors

OBJECTIVE

To determine if functional measures of ambulation can be accurately classified using clinical measures; demographics; personal, psychosocial, and environmental factors; and limb accelerations (LAs) obtained during sleep among individuals with chronic, motor incomplete spinal cord injury (SCI) in an effort to guide future, longitudinal predictions models.

DESIGN

Cross-sectional, 1-5 days of data collection.

SETTING

Community-based data collection.

PARTICIPANTS

Adults with chronic (>1 year), motor incomplete SCI (N=27).

INTERVENTIONS

Not applicable.

MAIN OUTCOME MEASURES

Ambulatory ability based on the 10-m walk test (10MWT) or 6-minute walk test (6MWT) categorized as nonambulatory, household ambulator (0.01-0.44 m/s, 1-204 m), or community ambulator (>0.44 m/s, >204 m). A random forest model classified ambulatory ability using input features including clinical measures of strength, sensation, and spasticity; demographics; personal, psychosocial, and environmental factors including pain, environmental factors, health, social support, self-efficacy, resilience, and sleep quality; and LAs measured during sleep. Machine learning methods were used explicitly to avoid overfitting and minimize the possibility of biased results.

RESULTS

The combination of LA, clinical, and demographic features resulted in the highest classification accuracies for both functional ambulation outcomes (10MWT=70.4%, 6MWT=81.5%). Adding LAs, personal, psychosocial, and environmental factors, or both increased the accuracy of classification compared with the clinical/demographic features alone. Clinical measures of strength and sensation (especially knee flexion strength), LA measures of movement smoothness, and presence of pain and comorbidities were among the most important features selected for the models.

CONCLUSIONS

The addition of LA and personal, psychosocial, and environmental features increased functional ambulation classification accuracy in a population with incomplete SCI for whom improved prognosis for mobility outcomes is needed. These findings provide support for future longitudinal studies that use LA; personal, psychosocial, and environmental factors; and advanced analyses to improve clinical prediction rules for functional mobility outcomes.

Use of Machine Learning and Artificial Intelligence to Drive Personalized Medicine Approaches for Spine Care

Personalized medicine is a new paradigm of healthcare in which interventions are based on individual patient characteristics rather than on "one-size-fits-all" guidelines. As epidemiological datasets continue to burgeon in size and complexity, powerful methods such as statistical machine learning and artificial intelligence (AI) become necessary to interpret and develop prognostic models from underlying data. Through such analysis, machine learning can be used to facilitate personalized medicine via its precise predictions. Additionally, other AI tools, such as natural language processing and computer vision, can play an instrumental part in personalizing the care provided to patients with spine disease. In the present report, we discuss the current strides made in incorporating AI into research on spine disease, especially traumatic spinal cord injury and degenerative spine disease. We describe studies using AI to build accurate prognostic models, extract important information from medical reports via natural language processing, and evaluate functional status in a granular manner using computer vision. Through a case illustration, we have demonstrated how these breakthroughs can facilitate an increased role for more personalized medicine and, thus, change the landscape of spine care.

XGBoost, a Machine Learning Method, Predicts Neurological Recovery in Patients with Cervical Spinal Cord Injury

The accurate prediction of neurological outcomes in patients with cervical spinal cord injury (SCI) is difficult because of heterogeneity in patient characteristics, treatment strategies, and radiographic findings. Although machine learning algorithms may increase the accuracy of outcome predictions in various fields, limited information is available on their efficacy in the management of SCI. We analyzed data from 165 patients with cervical SCI, and extracted important factors for predicting prognoses. Extreme gradient boosting (XGBoost) as a machine learning model was applied to assess the reliability of a machine learning algorithm to predict neurological outcomes compared with that of conventional methodology, such as a logistic regression or decision tree. We used regularly obtainable data as predictors, such as demographics, magnetic resonance variables, and treatment strategies. Predictive tools, including XGBoost, a logistic regression, and a decision tree, were applied to predict neurological improvements in the functional motor status (ASIA [American Spinal Injury Association] Impairment Scale [AIS] D and E) 6 months after injury. We evaluated predictive performance, including accuracy and the area under the receiver operating characteristic curve (AUC). Regarding predictions of neurological improvements in patients with cervical SCI, XGBoost had the highest accuracy (81.1%), followed by the logistic regression (80.6%) and the decision tree (78.8%). Regarding AUC, the logistic regression showed 0.877, followed by XGBoost (0.867) and the decision tree (0.753). XGBoost reliably predicted neurological alterations in patients with cervical SCI. The utilization of predictive machine learning algorithms may enhance personalized management choices through pre-treatment categorization of patients.