Developing Artificial Neural Network Models to Predict Functioning One Year After Traumatic Spinal Cord Injury

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

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).

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.

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.

Predicting the Outcome and Survival of Patients with Spinal Cord Injury Using Machine Learning Algorithms: A Systematic Review

BACKGROUND

Spinal cord injury (SCI) is a significant public health issue, leading to physical, psychological, and social complications. Machine learning (ML) algorithms have shown potential in diagnosing and predicting the functional and neurologic outcomes of subjects with SCI. ML algorithms can predict scores for SCI classification systems and accurately predict outcomes by analyzing large amounts of data. This systematic review aimed to examine the performance of ML algorithms for diagnosing and predicting the outcomes of subjects with SCI.

METHODS

The literature was comprehensively searched for the pertinent studies from inception to May 25, 2023. Therefore, electronic databases of PubMed, Embase, Scopus, and Web of Science were systematically searched with individual search syntax.

RESULTS

A total of 9424 individuals diagnosed with SCI across multiple studies were analyzed. Among the 21 studies included, 5 specifically aimed to evaluate diagnostic accuracy, while the remaining 16 focused on exploring prognostic factors or management strategies.

CONCLUSIONS

ML and deep learning (DL) have shown great potential in various aspects of SCI. ML and DL algorithms have been employed multiple times in predicting and diagnosing patients with SCI. While there are studies on diagnosing acute SCI using DL algorithms, further research is required in this area.

Harnessing Artificial Neural Networks for Spinal Cord Injury Prognosis

Background: Prediction of neurorehabilitation outcomes after a Spinal Cord Injury (SCI) is crucial for healthcare resource management and improving prognosis and rehabilitation strategies. Artificial neural networks (ANNs) have emerged as a promising alternative to conventional statistical approaches for identifying complex prognostic factors in SCI patients. Materials: a database of 1256 SCI patients admitted for rehabilitation was analyzed. Clinical and demographic data and SCI characteristics were used to predict functional outcomes using both ANN and linear regression models. The former was structured with input, hidden, and output layers, while the linear regression identified significant variables affecting outcomes. Both approaches aimed to evaluate and compare their accuracy for rehabilitation outcomes measured by the Spinal Cord Independence Measure (SCIM) score. Results: Both ANN and linear regression models identified key predictors of functional outcomes, such as age, injury level, and initial SCIM scores (correlation with actual outcome: R = 0.75 and 0.73, respectively). When also alimented with parameters recorded during hospitalization, the ANN highlighted the importance of these additional factors, like motor completeness and complications during hospitalization, showing an improvement in its accuracy (R = 0.87). Conclusions: ANN seemed to be not widely superior to classical statistics in general, but, taking into account complex and non-linear relationships among variables, emphasized the impact of complications during the hospitalization on recovery, particularly respiratory issues, deep vein thrombosis, and urological complications. These results suggested that the management of complications is crucial for improving functional recovery in SCI patients.

Lesion characteristics are associated with bowel, bladder, and overall independence following cervical spinal cord injury

CONTEXT/OBJECTIVE

There is a growing global interest in quantifying spinal cord lesions and spared neural tissue using magnetic resonance imaging (MRI) in individuals with spinal cord injury (SCI). The primary objective of this study was to assess the relationships between spinal cord lesion characteristics assessed on MRI and bowel, bladder, and overall independence following SCI.

DESIGN

Retrospective, exploratory study.

PARTICIPANTS

93 individuals with cervical SCI who were enrolled in a local United States Model Systems SCI database from 2010 to 2017.

METHODS

Clinical and MRI data were obtained for potential participants, and MRIs of eligible participants were analyzed. Explanatory variables, captured on MRIs, included intramedullary lesion length (IMLL), midsagittal ventral tissue bridge width (VTBW), midsagittal dorsal tissue bridge width (DTBW), and axial damage ratio (ADR).

OUTCOME MEASURES

Bowel and bladder management scale of the Functional Independence Measure (FIM) and FIM total motor score.

RESULTS

When accounting for all four variables, only ADR was significantly associated with bowel independence (OR = 0.970, 95% CI: 0.942-0.997, P = 0.030), and both ADR and IMLL were strongly associated with bladder independence (OR = 0.967, 95% CI: 0.936-0.999, P = 0.046 and OR = 0.948, 95% CI: 0.919-0.978, P = 0.0007, respectively). 32% of the variation in overall independence scores were explained by all four predictive variables, but only ADR was significantly associated with overall independence after accounting for all other predictive variables (β = -0.469, 95% CI: -0.719, -0.218, P = 0.0004).

CONCLUSIONS

Our results suggest that the MRI-measured extent of spinal cord lesion may be predictive of bowel, bladder, and overall independence following cervical SCI.

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.

Machine Learning Web Application for Predicting Functional Outcomes in Patients With Traumatic Spinal Cord Injury Following Inpatient Rehabilitation

Accurately predicting functional outcomes in patients with spinal cord injury (SCI) helps clinicians set realistic functional recovery goals and improve the home environment after discharge. The present study aimed to develop and validate machine learning (ML) models to predict functional outcomes in patients with SCI and deploy the models within a web application. The study included data from the Japan Association of Rehabilitation Database from January 1, 1991, to December 31, 2015. Patients with SCI who were admitted to an SCI center or transferred to a participating post-acute rehabilitation hospital after receiving acute treatment were enrolled in this database. The primary outcome was functional ambulation at discharge from the rehabilitation hospital. The secondary outcome was the total motor Functional Independence Measure (FIM) score at discharge. We used binary classification models to predict whether functional ambulation was achieved, as well as regression models to predict total motor FIM scores at discharge. In the training dataset (70% random sample) using demographic characteristics and neurological and functional status as predictors, we built prediction performance matrices of multiple ML models and selected the best one for each outcome. We validated each model's predictive performance in the test dataset (the remaining 30%). Among the 4181 patients, 3827 were included in the prediction model for the total motor FIM score. The mean (standard deviation [SD]) age was 50.4 (18.7) years, and 3211 (83.9%) patients were male. There were 3122 patients included in the prediction model for functional ambulation. The CatBoost Classifier and regressor models showed the best performances in the training dataset. On the test dataset, the CatBoost Classifier had an area under the receiver operating characteristic curve of 0.8572 and an accuracy of 0.7769 for predicting functional ambulation. Likewise, the CatBoost Regressor performed well, with an R2 of 0.7859, a mean absolute error of 9.2957, and a root mean square error of 13.4846 for predicting the total motor FIM score. The final models were deployed in a web application to provide functional predictions. The application can be found at http://3.138.174.54:8501. In conclusion, our prediction models developed using ML successfully predicted functional outcomes in patients with SCI and were deployed in an open-access web application.

Guidelines for neuroprognostication in adults with traumatic spinal cord injury

BACKGROUND

Traumatic spinal cord injury (tSCI) impacts patients and their families acutely and often for the long term. The ability of clinicians to share prognostic information about mortality and functional outcomes allows patients and their surrogates to engage in decision-making and plan for the future. These guidelines provide recommendations on the reliability of acute-phase clinical predictors to inform neuroprognostication and guide clinicians in counseling adult patients with tSCI or their surrogates.

METHODS

A narrative systematic review was completed using Grading of Recommendations Assessment, Development, and Evaluation methodology. Candidate predictors, including clinical variables and prediction models, were selected based on clinical relevance and presence of an appropriate body of evidence. The Population/Intervention/Comparator/Outcome/Timing/Setting question was framed as "When counseling patients or surrogates of critically ill patients with traumatic spinal cord injury, should < predictor, with time of assessment if appropriate > be considered a reliable predictor of < outcome, with time frame of assessment >?" Additional full-text screening criteria were used to exclude small and lower quality studies. Following construction of an evidence profile and summary of findings, recommendations were based on four Grading of Recommendations Assessment, Development, and Evaluation criteria: quality of evidence, balance of desirable and undesirable consequences, values and preferences, and resource use. Good practice recommendations addressed essential principles of neuroprognostication that could not be framed in the Population/Intervention/Comparator/Outcome/Timing/Setting format. Throughout the guideline development process, an individual living with tSCI provided perspective on patient-centered priorities.

RESULTS

Six candidate clinical variables and one prediction model were selected. Out of 11,132 articles screened, 369 met inclusion criteria for full-text review and 35 articles met eligibility criteria to guide recommendations. We recommend pathologic findings on magnetic resonance imaging, neurological level of injury, and severity of injury as moderately reliable predictors of American Spinal Cord Injury Impairment Scale improvement and the Dutch Clinical Prediction Rule as a moderately reliable prediction model of independent ambulation at 1 year after injury. No other reliable or moderately reliable predictors of mortality or functional outcome were identified. Good practice recommendations include considering the complete clinical condition as opposed to a single variable and communicating the challenges of likely functional deficits as well as potential for improvement and for long-term quality of life with SCI-related deficits to patients and surrogates.

CONCLUSIONS

These guidelines provide recommendations about the reliability of acute-phase predictors of mortality, functional outcome, American Spinal Injury Association Impairment Scale grade conversion, and recovery of independent ambulation for consideration when counseling patients with tSCI or their surrogates and suggest broad principles of neuroprognostication in this context.

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.

Are Clinical Prediction Rules Used in Spinal Cord Injury Care? A Survey of Practice

Background

Accurate outcome prediction is desirable post spinal cord injury (SCI), reducing uncertainty for patients and supporting personalized treatments. Numerous attempts have been made to create clinical prediction rules that identify patients who are likely to recover function. It is unknown to what extent these rules are routinely used in clinical practice.

Objectives

To better understand knowledge of, and attitudes toward, clinical prediction rules amongst SCI clinicians in the United Kingdom.

Methods

An online survey was distributed via mailing lists of clinical special interest groups and relevant National Health Service Trusts. Respondents answered questions about their knowledge of existing clinical prediction rules and their general attitudes to using them. They also provided information about their level of experience with SCI patients.

Results

One hundred SCI clinicians completed the survey. The majority (71%) were unaware of clinical prediction rules for SCI; only 8% reported using them in clinical practice. Less experienced clinicians were less likely to be aware. Lack of familiarity with prediction rules was reported as being a barrier to their use. The importance of clinical expertise when making prognostic decisions was emphasized. All respondents reported interest in using clinical prediction rules in the future.

Conclusion

The results show widespread lack of awareness of clinical prediction rules amongst SCI clinicians in the United Kingdom. However, clinicians were positive about the potential for clinical prediction rules to support decision-making. More focus should be directed toward refining current rules and improving dissemination within the SCI community.

Studying missingness in spinal cord injury data: challenges and impact of data imputation

BACKGROUND

In the last decades, medical research fields studying rare conditions such as spinal cord injury (SCI) have made extensive efforts to collect large-scale data. However, most analysis methods rely on complete data. This is particularly troublesome when studying clinical data as they are prone to missingness. Often, researchers mitigate this problem by removing patients with missing data from the analyses. Less commonly, imputation methods to infer likely values are applied.

OBJECTIVE

Our objective was to study how handling missing data influences the results reported, taking the example of SCI registries. We aimed to raise awareness on the effects of missing data and provide guidelines to be applied for future research projects, in SCI research and beyond.

METHODS

Using the Sygen clinical trial data (n = 797), we analyzed the impact of the type of variable in which data is missing, the pattern according to which data is missing, and the imputation strategy (e.g. mean imputation, last observation carried forward, multiple imputation).

RESULTS

Our simulations show that mean imputation may lead to results strongly deviating from the underlying expected results. For repeated measures missing at late stages (> = 6 months after injury in this simulation study), carrying the last observation forward seems the preferable option for the imputation. This simulation study could show that a one-size-fit-all imputation strategy falls short in SCI data sets.

CONCLUSIONS

Data-tailored imputation strategies are required (e.g., characterisation of the missingness pattern, last observation carried forward for repeated measures evolving to a plateau over time). Therefore, systematically reporting the extent, kind and decisions made regarding missing data will be essential to improve the interpretation, transparency, and reproducibility of the research presented.

Verification of the Accuracy of Cervical Spinal Cord Injury Prognosis Prediction Using Clinical Data-Based Artificial Neural Networks

BACKGROUND

In patients with cervical spinal cord injury (SCI), we need to make accurate prognostic predictions in the acute phase for more effective rehabilitation. We hypothesized that a multivariate prognosis would be useful for patients with cervical SCI.

METHODS

We made two predictive models using Multiple Linear Regression (MLR) and Artificial Neural Networks (ANNs). We adopted MLR as a conventional predictive model. Both models were created using the same 20 clinical parameters of the acute phase data at the time of admission. The prediction results were classified by the ASIA Impairment Scale. The training data consisted of 60 cases, and prognosis prediction was performed for 20 future cases (test cohort). All patients were treated in the Spinal Injuries Center (SIC) in Fukuoka, Japan.

RESULTS

A total of 16 out of 20 cases were predictable. The correct answer rate of MLR was 31.3%, while the rate of ANNs was 75.0% (number of correct answers: 12).

CONCLUSION

We were able to predict the prognosis of patients with cervical SCI from acute clinical data using ANNs. Performing effective rehabilitation based on this prediction will improve the patient's quality of life after discharge. Although there is room for improvement, ANNs are useful as a prognostic tool for patients with cervical SCI.

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.

Prediction of Lower Urinary Tract, Sexual, and Bowel Function, and Autonomic Dysreflexia after Spinal Cord Injury

Spinal cord injury (SCI) produces damage to the somatic and autonomic pathways that regulate lower urinary tract, sexual, and bowel function, and increases the risk of autonomic dysreflexia. The recovery of these functions has a high impact on health, functioning, and quality of life and is set as the utmost priority by patients. The application of reliable models to predict lower urinary tract, sexual, and bowel function, and autonomic dysreflexia is important for guiding counseling, rehabilitation, and social reintegration. Moreover, a reliable prediction is essential for designing future clinical trials to optimize patients' allocation to different treatment groups. To date, reliable and simple algorithms are available to predict lower urinary tract and bowel outcomes after traumatic and ischemic SCI. Previous studies identified a few risk factors to develop autonomic dysreflexia, albeit a model for prediction still lacks. On the other hand, there is an urgent need for a model to predict the evolution of sexual function. The aim of this review is to examine the available knowledge and models for the prediction of lower urinary tract, sexual, and bowel function, and autonomic dysreflexia after SCI, and critically discuss the research priorities in these fields.

Improvement of predictive accuracies of functional outcomes after subacute stroke inpatient rehabilitation by machine learning models

OBJECTIVES

Stepwise linear regression (SLR) is the most common approach to predicting activities of daily living at discharge with the Functional Independence Measure (FIM) in stroke patients, but noisy nonlinear clinical data decrease the predictive accuracies of SLR. Machine learning is gaining attention in the medical field for such nonlinear data. Previous studies reported that machine learning models, regression tree (RT), ensemble learning (EL), artificial neural networks (ANNs), support vector regression (SVR), and Gaussian process regression (GPR), are robust to such data and increase predictive accuracies. This study aimed to compare the predictive accuracies of SLR and these machine learning models for FIM scores in stroke patients.

METHODS

Subacute stroke patients (N = 1,046) who underwent inpatient rehabilitation participated in this study. Only patients' background characteristics and FIM scores at admission were used to build each predictive model of SLR, RT, EL, ANN, SVR, and GPR with 10-fold cross-validation. The coefficient of determination (R2) and root mean square error (RMSE) values were compared between the actual and predicted discharge FIM scores and FIM gain.

RESULTS

Machine learning models (R2 of RT = 0.75, EL = 0.78, ANN = 0.81, SVR = 0.80, GPR = 0.81) outperformed SLR (0.70) to predict discharge FIM motor scores. The predictive accuracies of machine learning methods for FIM total gain (R2 of RT = 0.48, EL = 0.51, ANN = 0.50, SVR = 0.51, GPR = 0.54) were also better than of SLR (0.22).

CONCLUSIONS

This study suggested that the machine learning models outperformed SLR for predicting FIM prognosis. The machine learning models used only patients' background characteristics and FIM scores at admission and more accurately predicted FIM gain than previous studies. ANN, SVR, and GPR outperformed RT and EL. GPR could have the best predictive accuracy for FIM prognosis.

Use of artificial neural networks in the prognosis of musculoskeletal diseases-a scoping review

To determine the current evidence on artificial neural network (ANN) in prognostic studies of musculoskeletal diseases (MSD) and to assess the accuracy of ANN in predicting the prognosis of patients with MSD. The scoping review was reported under the Preferred Items for Systematic Reviews and the Meta-Analyses extension for Scope Reviews (PRISMA-ScR). Cochrane Library, Embase, Pubmed, and Web of science core collection were searched from inception to January 2023. Studies were eligible if they used ANN to make predictions about MSD prognosis. Variables, model prediction accuracy, and disease type used in the ANN model were extracted and charted, then presented as a table along with narrative synthesis. Eighteen Studies were included in this scoping review, with 16 different types of musculoskeletal diseases. The accuracy of the ANN model predictions ranged from 0.542 to 0.947. ANN models were more accurate compared to traditional logistic regression models. This scoping review suggests that ANN can predict the prognosis of musculoskeletal diseases, which has the potential to be applied to different types of MSD.

Prognosis prediction in traumatic brain injury patients using machine learning algorithms

Predicting treatment outcomes in traumatic brain injury (TBI) patients is challenging worldwide. The present study aimed to achieve the most accurate machine learning (ML) algorithms to predict the outcomes of TBI treatment by evaluating demographic features, laboratory data, imaging indices, and clinical features. We used data from 3347 patients admitted to a tertiary trauma centre in Iran from 2016 to 2021. After the exclusion of incomplete data, 1653 patients remained. We used ML algorithms such as random forest (RF) and decision tree (DT) with ten-fold cross-validation to develop the best prediction model. Our findings reveal that among different variables included in this study, the motor component of the Glasgow coma scale, the condition of pupils, and the condition of cisterns were the most reliable features for predicting in-hospital mortality, while the patients' age takes the place of cisterns condition when considering the long-term survival of TBI patients. Also, we found that the RF algorithm is the best model to predict the short-term mortality of TBI patients. However, the generalized linear model (GLM) algorithm showed the best performance (with an accuracy rate of 82.03 ± 2.34) in predicting the long-term survival of patients. Our results showed that using appropriate markers and with further development, ML has the potential to predict TBI patients' survival in the short- and long-term.

MRI metrics at the epicenter of spinal cord injury are correlated with the stepping process in rhesus monkeys

Clinical evaluations of long-term outcomes in the early-stage spinal cord injury (SCI) focus on macroscopic motor performance and are limited in their prognostic precision. This study was designed to investigate the sensitivity of the magnetic resonance imaging (MRI) indexes to the data-driven gait process after SCI. Ten adult female rhesus monkeys were subjected to thoracic SCI. Kinematics-based gait examinations were performed at 1 (early stage) and 12 (chronic stage) months post-SCI. The proportion of stepping (PS) and gait stability (GS) were calculated as the outcome measures. MRI metrics, which were derived from structural imaging (spinal cord cross-sectional area, SCA) and diffusion tensor imaging (fractional anisotropy, FA; axial diffusivity, λ_//), were acquired in the early stage and compared with functional outcomes by using correlation analysis and stepwise multivariable linear regression. Residual tissue SCA at the injury epicenter and residual tissue FA/remote normal-like tissue FA were correlated with the early-stage PS and GS. The extent of lesion site λ_///residual tissue λ_// in the early stage after SCI was correlated with the chronic-stage GS. The ratios of lesion site λ_// to residual tissue λ_// and early-stage GS were predictive of the improvement in the PS at follow-up. Similarly, the ratios of lesion site λ_// to residual tissue λ_// and early-stage PS best predicted chronic GS recovery. Our findings demonstrate the predictive power of MRI combined with the early data-driven gait indexes for long-term outcomes. Such an approach may help clinicians to predict functional recovery accurately.

Early Trauma Predictors of Mobility in People with Spinal Cord Injury

STUDY DESIGN

A retrospective cohort study.

OBJECTIVE

This study aims to assess the potential value of very early trauma variables such as Abbreviated Injury Scale (AIS) and the Injury Severity Score for predicting independent ambulation following a traumatic spinal cord injury (TSCI).

SUMMARY OF BACKGROUND DATA

Several models for prediction of ambulation early after TSCI have been published and validated. The vast majority rely on the initial examination of American Spinal Injury Association (ASIA) impairment scale and level of injury; however, in many locations and clinical situations this examination is not feasible early after the injury.

METHODS

Patient characteristics, trauma data, and ASIA scores on admission to rehabilitation were collected for each of the 144 individuals in the study. Outcome measure was the indoor mobility item of the Spinal Cord Independence Measure taken upon discharge from rehabilitation. Univariate and multivariable models were created for each predictor, Odds ratios (ORs) were obtained by a multivariable logistic regression analysis, and area under the receiver operator curve was calculated for each model.

RESULTS

We observed a significant correlation between the trauma variables and independent ambulation upon discharge from rehabilitation. Of the early variables, the AIS for the spine region showed the strongest correlation.

CONCLUSION

These findings support using preliminary trauma variables for early prognostication of ambulation following a TSCI, allowing for tailored individual interventions.Level of Evidence: 3.

The potential of prediction models of functioning remains to be fully exploited: A scoping review in the field of spinal cord injury rehabilitation

OBJECTIVE

The study aimed to explore existing prediction models of functioning in spinal cord injury (SCI).

STUDY DESIGN AND SETTING

The databases PubMed, EBSCOhost CINAHL Complete, and IEEE Xplore were searched for relevant literature. The search strategy included published search filters for prediction model and impact studies, index terms and keywords for SCI, and relevant outcome measures able to assess functioning as reflected in the International Classification of Functioning, Disability and Health (ICF). The search was completed in October 2020.

RESULTS

We identified seven prediction model studies reporting twelve prediction models of functioning. The identified prediction models were mainly envisioned to be used for rehabilitation planning, however, also other possible applications were stated. The method predominantly used was regression analysis and the investigated predictors covered mainly the ICF-components of body functions and activities and participation, next to characteristics of the health condition and health interventions.

CONCLUSION

Findings suggest that the development of prediction models of functioning for use in clinical practice remains to be fully exploited. By providing a comprehensive overview of what has been done, this review informs future research on prediction models of functioning in SCI and contributes to an efficient use of research evidence.

Improving Diagnostic Workup Following Traumatic Spinal Cord Injury: Advances in Biomarkers

PURPOSE OF REVIEW

Traumatic spinal cord injury (SCI) is a life-changing event with drastic implications for patients due to sensorimotor impairment and autonomous dysfunction. Current clinical evaluations focus on the assessment of injury level and severity using standardized neurological examinations. However, they fail to predict individual trajectories of recovery, which highlights the need for the development of advanced diagnostics. This narrative review identifies recent advances in the search of clinically relevant biomarkers in the field of SCI.

RECENT FINDINGS

Advanced neuroimaging and molecular biomarkers sensitive to the disease processes initiated by the SCI have been identified. These biomarkers range from advanced neuroimaging techniques, neurophysiological readouts, and molecular biomarkers identifying the concentrations of several proteins in blood and CSF samples. Some of these biomarkers improve current prediction models based on clinical readouts. Validation with larger patient cohorts is warranted. Several biomarkers have been identified-ranging from imaging to molecular markers-that could serve as advanced diagnostic and hence supplement current clinical assessments.

Clinical judgment is a cornerstone for validating and using clinical prediction rules: a head-to-head study on ambulation outcomes for spinal cord injured patients

STUDY DESIGN

Retrospective comparative study.

OBJECTIVE

Clinical prediction rules (CPRs) are an effervescent topic in the medical literature. Recovering ambulation after a traumatic spinal cord injury (tSCI) is a priority for patients and multiple CPRs have been proposed for predicting ambulation outcomes. Our objective is to confront clinical judgment to an established CPR developed for patients with tSCI.

SETTINGS

Level one trauma center specialized in tSCI and its affiliated rehabilitation center.

METHOD

In this retrospective comparative study, six physicians had to predict the ambulation outcome of 68 patients after a tSCI based on information from the acute hospitalization. Ambulation was also predicted according to the CPR of van Middendorp (CPR-vM). The success rate of the CPR-vM and clinicians to predict ambulation was compared using criteria of 5% for defining clinical significance, and a level of statistical significance of 0.05 for bilateral McNemar tests.

RESULTS

There was no statistical difference between the overall performance of physicians (success rate of 79%) and of the CPR-vM (81%) for predicting ambulation. The differences between the CPR-vM and physicians varied clinically and significantly with the level of experience, clinical setting, and field of expertise.

CONCLUSION

Confronting CPRs with the judgment of a group of clinicians should be an integral part of the design and validation of CPRs. Head-to-head comparison of CPRs with clinicians is also a cornerstone for defining the optimal strategy for translation into the clinical practice, and for defining which clinician and specific clinical context would benefit from using the CPR.

An arrow that missed the mark: a pediatric case report of remarkable neurologic improvement following penetrating spinal cord injury

Penetrating spinal cord injuries are rare in children but result in devastating impacts on long-term morbidity and mortality-with little known about the recovery capacity in this age group. We present the case of an eight-year-old child who sustained a penetrating injury through the right anterior thorax. Thoracic CT showed the arrow tip extending through the spinal canal at T6. Neurologic examination revealed no motor or sensory function below T6. The arrow was surgically removed without complications through an anterior-only approach. MRI on post-operative day (POD) 4 showed focal T2 hyperintensity at the T6 spinal cord. Patient was discharged on POD33 with an American Spinal Injury Association (ASIA)-D score and trace voluntary control over bowel and bladder function. Remarkably, four months later, he had near normal bowel and bladder function, with near-intact lower extremity strength and self-sustained ambulation. Follow-up imaging revealed hemicord formation at the level of injury. We review our case of penetrating spinal cord injury in a child and similar reports in the literature. Penetrating thoracic spinal cord trauma portends poor clinical outcomes, particularly when employing available adult prognostic spinal cord injury scoring metrics. Incomplete spinal cord injury, and often-associated spinal shock, can mimic a complete injury-as in our patient, which improved to near-complete motor and sensory restoration of function and resulted in the formation of a split hemicord. This case represents a unique penetrating spinal cord injury with remarkable neurologic recovery, which would advocate against definitive early prognostication in the pediatric population.

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.

The effect of age and injury severity on clinical prediction rules for ambulation among individuals with spinal cord injury

BACKGROUND CONTEXT

While several models for predicting independent ambulation early after traumatic spinal cord injury (SCI) based upon age and specific motor and sensory level findings have been published and validated, their accuracy, especially in individual American Spinal Injury Association [ASIA] Impairment Scale (AIS) classifications, has been questioned. Further, although age is widely used in prediction rules, its role and possible modifications have not been adequately evaluated until now.

PURPOSE

To evaluate the predictive accuracy of existing clinical prediction rules for independent ambulation among individuals at spinal cord injury model systems (SCIMS) Centers as well as the effect of modifying the age parameter from a cutoff of 65 years to 50 years.

STUDY DESIGN

Retrospective analysis of a longitudinal database.

PATIENT SAMPLE

Adult individuals with traumatic SCI.

OUTCOME MEASURES

The FIM locomotor score was used to assess independent walking ability at the 1-year follow-up.

METHODS

In all, 639 patients were enrolled in the SCIMS database between 2011 and 2015, with complete neurological examination data within 15 days following the injury and a follow-up assessment with functional independence measure (FIM) at 1-year post injury. Two previously validated logistic regression models were evaluated for their ability to predict independent walking at 1-year post injury with participants in the SCIMS database. Area under the receiver operating curve (AUC) was calculated for the individual AIS categories and for different age groups. Prediction accuracy was also calculated for a new modified LR model (with cut-off age of 50).

RESULTS

Overall AUC for each of the previous prediction models was found to be consistent with previous reports (0.919 and 0.904). AUCs for grouped AIS levels (A+D, B+C) were consistent with prior reports, moreover, prediction for individual AIS grades continued to reveal lower values. AUCs by different age categories showed a decline in prognostication accuracy with an increase in age, with statistically significant improvement of AUC when age-cut off was reduced to 50.

CONCLUSIONS

We confirmed previous results that former prediction models achieve strong prognostic accuracy by combining AIS subgroups, yet prognostication of the separate AIS groups is less accurate. Further, prognostication of persons with AIS B+C, for whom a clinical prediction model has arguably greater clinical utility, is less accurate than those with AIS A+D. Our findings emphasize that age is an important factor in prognosticating ambulation following SCI. Prediction accuracy declines for older individuals compared with younger ones. To improve prediction of independent ambulation, the age of 50 years may be a better cutoff instead of age of 65.

Predicting clinically significant motor function improvement after contemporary task-oriented interventions using machine learning approaches

BACKGROUND

Accurate prediction of motor recovery after stroke is critical for treatment decisions and planning. Machine learning has been proposed to be a promising technique for outcome prediction because of its high accuracy and ability to process large volumes of data. It has been used to predict acute stroke recovery; however, whether machine learning would be effective for predicting rehabilitation outcomes in chronic stroke patients for common contemporary task-oriented interventions remains largely unexplored. This study aimed to determine the accuracy and performance of machine learning to predict clinically significant motor function improvements after contemporary task-oriented intervention in chronic stroke patients and identify important predictors for building machine learning prediction models.

METHODS

This study was a secondary analysis of data using two common machine learning approaches, which were the k-nearest neighbor (KNN) and artificial neural network (ANN). Chronic stroke patients (N = 239) that received 30 h of task-oriented training including the constraint-induced movement therapy, bilateral arm training, robot-assisted therapy and mirror therapy were included. The Fugl-Meyer assessment scale (FMA) was the main outcome. Potential predictors include age, gender, side of lesion, time since stroke, baseline functional status, motor function and quality of life. We divided the data set into a training set and a test set and used the cross-validation procedure to construct machine learning models based on the training set. After the models were built, we used the test data set to evaluate the accuracy and prediction performance of the models.

RESULTS

Three important predictors were identified, which were time since stroke, baseline functional independence measure (FIM) and baseline FMA scores. Models for predicting motor function improvements were accurate. The prediction accuracy of the KNN model was 85.42% and area under the receiver operating characteristic curve (AUC-ROC) was 0.89. The prediction accuracy of the ANN model was 81.25% and the AUC-ROC was 0.77.

CONCLUSIONS

Incorporating machine learning into clinical outcome prediction using three key predictors including time since stroke, baseline functional and motor ability may help clinicians/therapists to identify patients that are most likely to benefit from contemporary task-oriented interventions. The KNN and ANN models may be potentially useful for predicting clinically significant motor recovery in chronic stroke.

Gap Analysis Regarding Prognostication in Neurocritical Care: A Joint Statement from the German Neurocritical Care Society and the Neurocritical Care Society

Background/Objective

Prognostication is a routine part of the delivery of neurocritical care for most patients with acute neurocritical illnesses. Numerous prognostic models exist for many different conditions. However, there are concerns about significant gaps in knowledge regarding optimal methods of prognostication.

Methods

As part of the Arbeitstagung NeuroIntensivMedizin meeting in February 2018 in Würzburg, Germany, a joint session on prognostication was held between the German NeuroIntensive Care Society and the Neurocritical Care Society. The purpose of this session was to provide presentations and open discussion regarding existing prognostic models for eight common neurocritical care conditions (aneurysmal subarachnoid hemorrhage, intracerebral hemorrhage, acute ischemic stroke, traumatic brain injury, traumatic spinal cord injury, status epilepticus, Guillain–Barré Syndrome, and global cerebral ischemia from cardiac arrest). The goal was to develop a qualitative gap analysis regarding prognostication that could help inform a future framework for clinical studies and guidelines.

Results

Prognostic models exist for all of the conditions presented. However, there are significant gaps in prognostication in each condition. Furthermore, several themes emerged that crossed across several or all diseases presented. Specifically, the self-fulfilling prophecy, lack of accounting for medical comorbidities, and absence of integration of in-hospital care parameters were identified as major gaps in most prognostic models.

Conclusions

Prognostication in neurocritical care is important, and current prognostic models are limited. This gap analysis provides a summary assessment of issues that could be addressed in future studies and evidence-based guidelines in order to improve the process of prognostication.

Nationwide prediction of type 2 diabetes comorbidities

Identification of individuals at risk of developing disease comorbidities represents an important task in tackling the growing personal and societal burdens associated with chronic diseases. We employed machine learning techniques to investigate to what extent data from longitudinal, nationwide Danish health registers can be used to predict individuals at high risk of developing type 2 diabetes (T2D) comorbidities. Leveraging logistic regression-, random forest- and gradient boosting models and register data spanning hospitalizations, drug prescriptions and contacts with primary care contractors from >200,000 individuals newly diagnosed with T2D, we predicted five-year risk of heart failure (HF), myocardial infarction (MI), stroke (ST), cardiovascular disease (CVD) and chronic kidney disease (CKD). For HF, MI, CVD, and CKD, register-based models outperformed a reference model leveraging canonical individual characteristics by achieving area under the receiver operating characteristic curve improvements of 0.06, 0.03, 0.04, and 0.07, respectively. The top 1,000 patients predicted to be at highest risk exhibited observed incidence ratios exceeding 4.99, 3.52, 1.97 and 4.71 respectively. In summary, prediction of T2D comorbidities utilizing Danish registers led to consistent albeit modest performance improvements over reference models, suggesting that register data could be leveraged to systematically identify individuals at risk of developing disease comorbidities.

Development of an unsupervised machine learning algorithm for the prognostication of walking ability in spinal cord injury patients

BACKGROUND CONTEXT

Traumatic spinal cord injury can have a dramatic effect on a patient's life. The degree of neurologic recovery greatly influences a patient's treatment and expected quality of life. This has resulted in the development of machine learning algorithms (MLA) that use acute demographic and neurologic information to prognosticate recovery. The van Middendorp et al. (2011) (vM) logistic regression (LR) model has been established as a reference model for the prediction of walking recovery following spinal cord injury as it has been validated within many different countries. However, an examination of the way in which these prediction models are evaluated is warranted. The area under the receiver operators curve (AUROC) has been consistently used when evaluating model performance, but it has been shown that AUROC overemphasizes the most common event resulting in an inaccurate assessment when the data are imbalanced. Furthermore, there is evidence that the use of more advanced MLA, such as an unsupervised k-means model, may show superior performance compared to LR as they can handle a larger number of features.

PURPOSE

The first objective of the study was to assess the performance of both an unsupervised MLA and LR model with complete admission neurologic information against the vM and Hicks models. Second, a comparison between the accuracy of the AUROC and the F1-score will be made to determine which method is superior for the assessment of diagnostic performance of prediction models on large-scale datasets.

STUDY DESIGN

Retrospective review of a prospective cohort study.

PATIENT SAMPLE

The Rick Hansen Spinal Cord Injury Registry (RHSCIR) was used in this study. All patients enrolled between 2004 and 2017 with complete neurologic examination and Functional Independence Measure outcome data at ≥1 year follow-up or who could walk at discharge were included. The prognostic variables included age (dichotomized at ≥65 years old); American Spinal Injury Association Impairment Scale (AIS) grade; and individual motor, light touch, and pinprick score from L2 to S1.

OUTCOME MEASURES

The Functional Independence Measure locomotor score was used to assess independent walking ability at discharge or 1-year follow-up.

METHODS

An unsupervised MLA with k=2 was chosen in order to identify a "walk" cluster and a "not walk" cluster. Model performance was assessed through the development of a receiver operating characteristic curve with associated AUROC and a precision-recall curve with associated F1-score. The study and the RHSCIR are supported by funding from Health Canada, Western Economic Diversification Canada, and the Governments of Alberta, British Columbia, Manitoba, and Ontario. These funders had no role in the study or study reporting and the authors have no conflicts of interest to report.

RESULTS

No clinically relevant differences were found between with the use of an unsupervised MLA with a greater amount of initial neurologic information compared to the established standards for any AIS classification. Although demonstrated for all separate AIS classifications, most notably, the AUROC for the vM (0.78) and Hicks models (0.76) were found to be superior to that of the new LR model (0.72); however, the vM and Hicks models had more than double the amount of false negative classifications compared to the LR. The F1-scores between these three models were also found to be different but with the vM and Hicks models being lower than the LR (0.85, 0.81, and 0.89, respectively).

CONCLUSIONS

No clinically relevant differences were found between the use of an unsupervised MLA with complete admission neurologic information compared to the previously validated standards; however, when comparing the performance of the AUROC and F1-score, the AUROC showed inaccurate prognostic performance when there was an imbalance toward a greater amount of false negatives. Importantly, the F1-score did not succumb to this imbalance. As AUROC has been used as the standard when evaluating performance of prediction models, consideration as to whether this is the most appropriate method is warranted. Future work should focus on comparing AUROC and F1-scores with other previously validated models.

Bowel Outcome Prediction After Traumatic Spinal Cord Injury: Longitudinal Cohort Study

Background. Predicting functional outcomes after traumatic spinal cord injury (SCI) is essential for counseling, rehabilitation planning, and discharge. Moreover, the outcome prognosis is crucial for patient stratification when designing clinical trials. However, no valid prediction rule is currently available for bowel outcomes after a SCI. Objective. To generate a model for predicting the achievement of independent, reliable bowel management at 1 year after traumatic SCI. Methods. We performed multivariable logistic regression analyses of data for 1250 patients with traumatic SCIs that were included in the European Multicenter Study about Spinal Cord Injury. The resulting model was prospectively validated on data for 186 patients. As potential predictors, we evaluated age, sex, and variables from the International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI) and the Spinal Cord Independence Measure (SCIM), measured within 40 days of the injury. A positive outcome at 1 year post-SCI was assessed with item 7 of the SCIM. Results. The model relied on a single predictor, the ISNCSCI total motor score—that is, the sum of muscle strengths in 5 key muscle groups in each limb. The area under the receiver operating characteristics curve (aROC) was 0.837 (95% CI: 0.815-0.859). The prospective validation confirmed high predictive power: aROC = 0.817 (95% CI: 0.754-0.881). Conclusions. We generated a valid model for predicting independent, reliable bowel management at 1 year after traumatic SCI. Its application could improve counseling, optimize patient-tailored rehabilitation planning, and become crucial for appropriate patient stratification in future clinical trials.

A systematic review shows no performance benefit of machine learning over logistic regression for clinical prediction models

OBJECTIVES

The objective of this study was to compare performance of logistic regression (LR) with machine learning (ML) for clinical prediction modeling in the literature.

STUDY DESIGN AND SETTING

We conducted a Medline literature search (1/2016 to 8/2017) and extracted comparisons between LR and ML models for binary outcomes.

RESULTS

We included 71 of 927 studies. The median sample size was 1,250 (range 72-3,994,872), with 19 predictors considered (range 5-563) and eight events per predictor (range 0.3-6,697). The most common ML methods were classification trees, random forests, artificial neural networks, and support vector machines. In 48 (68%) studies, we observed potential bias in the validation procedures. Sixty-four (90%) studies used the area under the receiver operating characteristic curve (AUC) to assess discrimination. Calibration was not addressed in 56 (79%) studies. We identified 282 comparisons between an LR and ML model (AUC range, 0.52-0.99). For 145 comparisons at low risk of bias, the difference in logit(AUC) between LR and ML was 0.00 (95% confidence interval, -0.18 to 0.18). For 137 comparisons at high risk of bias, logit(AUC) was 0.34 (0.20-0.47) higher for ML.

CONCLUSION

We found no evidence of superior performance of ML over LR. Improvements in methodology and reporting are needed for studies that compare modeling algorithms.

Highlighting discrepancies in walking prediction accuracy for patients with traumatic spinal cord injury: an evaluation of validated prediction models using a Canadian Multicenter Spinal Cord Injury Registry

BACKGROUND CONTEXT

Models for predicting recovery in traumatic spinal cord injury (tSCI) patients have been developed to optimize care. Several models predicting tSCI recovery have been previously validated, yet recent findings question their accuracy, particularly in patients whose prognoses are the least predictable.

PURPOSE

To compare independent ambulatory outcomes in AIS (ASIA [American Spinal Injury Association] Impairment Scale) A, B, C, and D patients, as well as in AIS B+C and AIS A+D patients by applying two existing logistic regression prediction models.

STUDY DESIGN

A prospective cohort study.

PARTICIPANT SAMPLE

Individuals with tSCI enrolled in the pan-Canadian Rick Hansen SCI Registry (RHSCIR) between 2004 and 2016 with complete neurologic examination and Functional Independence Measure (FIM) outcome data.

OUTCOME MEASURES

The FIM locomotor score was used to assess independent walking ability at 1-year follow-up.

METHODS

Two validated prediction models were evaluated for their ability to predict walking 1-year postinjury. Relative prognostic performance was compared with the area under the receiver operating curve (AUC).

RESULTS

In total, 675 tSCI patients were identified for analysis. In model 1, predictive accuracies for 675 AIS A, B, C, and D patients as measured by AUC were 0.730 (95% confidence interval [CI] 0.622-0.838), 0.691 (0.533-0.849), 0.850 (0.771-0.928), and 0.516 (0.320-0.711), respectively. In 160 AIS B+C patients, model 1 generated an AUC of 0.833 (95% CI 0.771-0.895), whereas model 2 generated an AUC of 0.821 (95% CI 0.754-0.887). The AUC for 515 AIS A+D patients was 0.954 (95% CI 0.933-0.975) with model 1 and 0.950 (0.928-0.971) with model 2. The difference in prediction accuracy between the AIS B+C cohort and the AIS A+D cohort was statistically significant using both models (p=.00034; p=.00038). The models were not statistically different in individual or subgroup analyses.

CONCLUSIONS

Previously tested prediction models demonstrated a lower predictive accuracy for AIS B+C than AIS A+D patients. These models were unable to effectively prognosticate AIS A+D patients separately; a failure that was masked when amalgamating the two patient populations. This suggests that former prediction models achieved strong prognostic accuracy by combining AIS classifications coupled with a disproportionately high proportion of AIS A+D patients.

Comparison of artificial neural network and logistic regression models for prediction of outcomes in trauma patients: A systematic review and meta-analysis

BACKGROUND

Currently, two models of artificial neural network (ANN) and logistic regression (LR) are known as models that extensively used in medical sciences. The aim of this study was to compare the ANN and LR models in prediction of Health-related outcomes in traumatic patients using a systematic review.

METHODS

The study was planned and conducted based on the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) checklist. A literature search of published studies was conducted using PubMed, Embase, Web of knowledge, Scopus, and Google Scholar in May 2018. Joanna Briggs Institute (JBI) checklists was used for assessing the quality of the included articles.

RESULTS

The literature searches yielded 326 potentially relevant studies from the primary searches. Overall, the review included 10 unique studies. The results of this study showed that the area under curve (AUC) for the ANN was 0.91, (95% CI 0.89-0.83) and 0.89, (95% CI 0.87-90) for the LR in random effect model. The accuracy rate for ANN and LR in random effect models were 90.5, (95% CI, 87.6-94.2) and 83.2, (95% CI 75.1-91.2), respectively.

CONCLUSION

The results of our study showed that ANN has better performance than LR in predicting the terminal outcomes of traumatic patients in both the AUC and accuracy rate. Using an ANN to predict the final implications of trauma patients can provide more accurate clinical decisions.

A simplified clinical prediction rule for prognosticating independent walking after spinal cord injury: a prospective study from a Canadian multicenter spinal cord injury registry

BACKGROUND CONTEXT

Traumatic spinal cord injury (SCI) is a debilitating condition with limited treatment options for neurologic or functional recovery. The ability to predict the prognosis of walking post injury with emerging prediction models could aid in rehabilitation strategies and reintegration into the community.

PURPOSE

To revalidate an existing clinical prediction model for independent ambulation (van Middendorp et al., 2011) using acute and long-term post-injury follow-up data, and to investigatethe accuracy of a simplified model using prospectively collected data from a Canadian multicenter SCI database, the Rick Hansen Spinal Cord Injury Registry (RHSCIR).

STUDY DESIGN

Prospective cohort study.

PARTICIPANT SAMPLE

The analysis cohort consisted of 278 adult individuals with traumatic SCI enrolled in the RHSCIR for whom complete neurologic examination data and Functional Independence Measure (FIM) outcome data were available.

OUTCOME MEASURES

The FIM locomotor score was used to assess independent walking ability (defined as modified or complete independence in walk or combined walk and wheelchair modality) at 1-year follow-up for each participant.

METHODS

A logistic regression (LR) model based on age and four neurologic variables was applied to our cohort of 278 RHSCIR participants. Additionally, a simplified LR model was created. The Hosmer-Lemeshow goodness of fit test was used to check if the predictive model is applicable to our data set. The performance of the model was verified by calculating the area under the receiver operating characteristic curve (AUC). The accuracy of the model was tested using a cross-validation technique. This study was supported by a grant from The Ottawa Hospital Academic Medical Organization ($50,000 over 2 years). The RHSCIR is sponsored by the Rick Hansen Institute and is supported by funding from Health Canada, Western Economic Diversification Canada, and the provincial governments of Alberta, British Columbia, Manitoba, and Ontario. ET and JP report receiving grants from the Rick Hansen Institute (approximately $60,000 and $30,000 per year, respectively). DMR reports receiving remuneration for consulting services provided to Palladian Health, LLC and Pacira Pharmaceuticals, Inc ($20,000-$30,000 annually), although neither relationship presents a potential conflict of interest with the submitted work. KEH received a grant for involvement in the present study from the Government of Canada as part of the Canada Summer Jobs Program ($3,000). JP reports receiving an educational grant from Medtronic Canada outside of the submitted work ($75,000 annually). TM reports receiving educational fellowship support from AO Spine, AO Trauma, and Medtronic; however, none of these relationships are financial in nature. All remaining authors have no conflicts of interest to disclose.

RESULTS

The fitted prediction model generated 85% overall classification accuracy, 79% sensitivity, and 90% specificity. The prediction model was able to accurately classify independent walking ability (AUC 0.889, 95% confidence interval [CI] 0.846-0.933, p<.001) compared with the existing prediction model, despite the use of a different outcome measure (FIM vs. Spinal Cord Independence Measure) to qualify walking ability. A simplified, three-variable LR model based on age and two neurologic variables had an overall classification accuracy of 84%, with 76% sensitivity and 90% specificity, demonstrating comparable accuracy with its five-variable prediction model counterpart. The AUC was 0.866 (95% CI 0.816-0.916, p<.01), only marginally less than that of the existing prediction model.

CONCLUSIONS

A simplified predictive model with similar accuracy to a more complex model for predicting independent walking was created, which improves utility in a clinical setting. Such models will allow clinicians to better predict the prognosis of ambulation in individuals who have sustained a traumatic SCI.

Neural network prediction of severe lower intestinal bleeding and the need for surgical intervention

BACKGROUND

The prognosis for patients with severe acute lower intestinal bleeding (ALIB) may be assessed by complex artificial neural networks (ANNs) or user-friendly regression-based models. Comparisons between these modalities are limited, and predicting the need for surgical intervention remains elusive. We hypothesized that ANNs would outperform the Strate rule to predict severe bleeding and would also predict the need for surgical intervention.

METHODS

We performed a 4-y retrospective analysis of 147 adult patients who underwent endoscopy, angiography, or surgery for ALIB. Baseline characteristics, Strate risk factors, management parameters, and outcomes were analyzed. The primary outcomes were severe bleeding and surgical intervention. ANNs were created in SPSS. Models were compared by area under the receiver operating characteristic curve (AUROC) with 95% confidence intervals.

RESULTS

The number of Strate risk factors for each patient correlated significantly with the outcome of severe bleeding (r = 0.29, P < 0.001). However, the Strate model was less accurate than an ANN (AUROC 0.66 [0.57-0.75] versus 0.98 [0.95-1.00], respectively) which incorporated six variables present on admission: hemoglobin, systolic blood pressure, outpatient prescription for Aspirin 325 mg daily, Charlson comorbidity index, base deficit ≥5 mEq/L, and international normalized ratio ≥1.5. A similar ANN including hemoglobin nadir and the occurrence of a 20% decrease in hematocrit was effective in predicting the need for surgery (AUROC 0.95 [0.90-1.00]).

CONCLUSIONS

The Strate prediction rule effectively stratified risk for severe ALIB, but was less accurate than an ANN. A separate ANN accurately predicted the need for surgery by combining risk factors for severe bleeding with parameters quantifying blood loss. Optimal prognostication may be achieved by integrating pragmatic regression-based calculators for quick decisions at the bedside and highly accurate ANNs when time and resources permit.