1
|
Håkansson S, Tuci M, Bolliger M, Curt A, Jutzeler CR, Brüningk SC. Data-driven prediction of spinal cord injury recovery: An exploration of current status and future perspectives. Exp Neurol 2024; 380:114913. [PMID: 39097073 DOI: 10.1016/j.expneurol.2024.114913] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/30/2024] [Revised: 07/24/2024] [Accepted: 07/30/2024] [Indexed: 08/05/2024]
Abstract
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).
Collapse
Affiliation(s)
- Samuel Håkansson
- ETH Zürich, Department of Health Sciences and Technology (D-HEST), Zürich, Switzerland; Swiss Institute of Bioinformatics (SIB), Lausanne, Switzerland.
| | - Miklovana Tuci
- ETH Zürich, Department of Health Sciences and Technology (D-HEST), Zürich, Switzerland; Spinal Cord Injury Center, University Hospital Balgrist, University of Zürich, Switzerland
| | - Marc Bolliger
- Spinal Cord Injury Center, University Hospital Balgrist, University of Zürich, Switzerland
| | - Armin Curt
- Spinal Cord Injury Center, University Hospital Balgrist, University of Zürich, Switzerland
| | - Catherine R Jutzeler
- ETH Zürich, Department of Health Sciences and Technology (D-HEST), Zürich, Switzerland; Swiss Institute of Bioinformatics (SIB), Lausanne, Switzerland
| | - Sarah C Brüningk
- ETH Zürich, Department of Health Sciences and Technology (D-HEST), Zürich, Switzerland; Swiss Institute of Bioinformatics (SIB), Lausanne, Switzerland
| |
Collapse
|
2
|
Onate D, Hogan C, Fitzgerald K, White KT, Tansey K. Recommendations for clinical decision-making when offering exoskeletons for community use in individuals with spinal cord injury. FRONTIERS IN REHABILITATION SCIENCES 2024; 5:1428708. [PMID: 39206134 PMCID: PMC11349703 DOI: 10.3389/fresc.2024.1428708] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 05/06/2024] [Accepted: 07/29/2024] [Indexed: 09/04/2024]
Abstract
Approved in 2014 by the Food and Drug Administration (FDA) for use with a trained companion, personal powered exoskeletons (PPE) for individuals with spinal cord injury (SCI) provide an opportunity for the appropriate candidate to ambulate in their home and community. As an adjunct to wheeled mobility, PPE use allows those individuals who desire to ambulate the opportunity to experience the potential physiological and psychosocial benefits of assisted walking outside of a rehabilitation setting. There exists, however, a knowledge gap for clinicians regarding appropriate candidate selection for use, as well as who might benefit from ambulating with a PPE. The purpose of this paper is to provide guidance for clinicians working with individuals living with SCI by outlining an expert consensus for a PPE decision-making algorithm, as well as a discussion of potential physiological and psychosocial benefits from PPE use based on early evidence in publication.
Collapse
Affiliation(s)
| | - Cassandra Hogan
- United States Department of Veterans Affairs, James A. Haley Veterans' Hospital, Tampa, FL, United States
| | - Kathryn Fitzgerald
- United States Department of Veterans Affairs, James A. Haley Veterans' Hospital, Tampa, FL, United States
| | - Kevin T White
- United States Department of Veterans Affairs, James A. Haley Veterans' Hospital, Tampa, FL, United States
- Department of Neurology, University of South Florida, Tampa, FL, United States
| | - Keith Tansey
- Center for Neuroscience and Neurological Recovery, Methodist Rehabilitation Center, Jackson, MS, United States
- United States Department of Veterans Affairs, G.V. (Sonny) Montgomery VA Medical Center, Jackson, MS, United States
- Department of Neurosurgery, University of Mississippi Medical Center, Jackson, MS, United States
| |
Collapse
|
3
|
Draganich C, Anderson D, Dornan GJ, Sevigny M, Berliner J, Charlifue S, Welch A, Smith A. Predictive modeling of ambulatory outcomes after spinal cord injury using machine learning. Spinal Cord 2024; 62:446-453. [PMID: 38890506 DOI: 10.1038/s41393-024-01008-2] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/03/2023] [Revised: 05/12/2024] [Accepted: 06/06/2024] [Indexed: 06/20/2024]
Abstract
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.
Collapse
Affiliation(s)
- Christina Draganich
- University of Colorado Department of Physical Medicine and Rehabilitation, Aurora, CO, USA.
| | | | | | | | - Jeffrey Berliner
- University of Colorado Department of Physical Medicine and Rehabilitation, Aurora, CO, USA
- Craig Hospital, Englewood, CO, USA
| | | | | | - Andrew Smith
- University of Colorado Department of Physical Medicine and Rehabilitation, Aurora, CO, USA
| |
Collapse
|
4
|
Boyles RH, Alexander CM, Belsi A, Strutton PH. Are Clinical Prediction Rules Used in Spinal Cord Injury Care? A Survey of Practice. Top Spinal Cord Inj Rehabil 2024; 30:45-58. [PMID: 38433737 PMCID: PMC10906376 DOI: 10.46292/sci23-00069] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 03/05/2024]
Abstract
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.
Collapse
Affiliation(s)
- Rowan H. Boyles
- Department of Surgery and Cancer, Imperial College London, London, United Kingdom
- Department of Therapies, Imperial College Healthcare NHS Trust, London, United Kingdom
| | - Caroline M. Alexander
- Department of Surgery and Cancer, Imperial College London, London, United Kingdom
- Department of Therapies, Imperial College Healthcare NHS Trust, London, United Kingdom
| | - Athina Belsi
- Department of Surgery and Cancer, Imperial College London, London, United Kingdom
| | - Paul H. Strutton
- Department of Surgery and Cancer, Imperial College London, London, United Kingdom
| |
Collapse
|
5
|
Smith AC, Draganich C, Thornton WA, Berliner JC, Lennarson PJ, Rejc E, Sevigny M, Charlifue S, Tefertiller C, Weber KA. A Single Dermatome Clinical Prediction Rule for Independent Walking 1 Year After Spinal Cord Injury. Arch Phys Med Rehabil 2024; 105:10-19. [PMID: 37414239 PMCID: PMC10766862 DOI: 10.1016/j.apmr.2023.06.015] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/24/2023] [Revised: 04/24/2023] [Accepted: 06/22/2023] [Indexed: 07/08/2023]
Abstract
OBJECTIVE To derive and validate a simple, accurate CPR to predict future independent walking ability after SCI at the bedside that does not rely on motor scores and is predictive for those initially classified in the middle of the SCI severity spectrum. DESIGN Retrospective cohort study. Binary variables were derived, indicating degrees of sensation to evaluate predictive value of pinprick and light touch variables across dermatomes. The optimal single sensory modality and dermatome was used to derive our CPR, which was validated on an independent dataset. SETTING Analysis of SCI Model Systems dataset. PARTICIPANTS Individuals with traumatic SCI. The data of 3679 participants (N=3679) were included with 623 participants comprising the derivation dataset and 3056 comprising the validation dataset. INTERVENTIONS Not applicable. MAIN OUTCOME MEASURES Self-reported ability to walk both indoors and outdoors. RESULTS Pinprick testing at S1 over lateral heels, within 31 days of SCI, accurately identified future independent walkers 1 year after SCI. Normal pinprick in both lateral heels provided good prognosis, any pinprick sensation in either lateral heel provided fair prognosis, and no sensation provided poor prognosis. This CPR performed satisfactorily in the middle SCI severity subgroup. CONCLUSIONS In this large multi-site study, we derived and validated a simple, accurate CPR using only pinprick sensory testing at lateral heels that predicts future independent walking after SCI.
Collapse
Affiliation(s)
- Andrew C Smith
- Department of Physical Medicine and Rehabilitation, University of Colorado School of Medicine, Aurora, CO.
| | - Christina Draganich
- Department of Physical Medicine and Rehabilitation, University of Colorado School of Medicine, Aurora, CO; Craig Hospital, Englewood, CO
| | - Wesley A Thornton
- Department of Physical Medicine and Rehabilitation, University of Colorado School of Medicine, Aurora, CO
| | - Jeffrey C Berliner
- Department of Physical Medicine and Rehabilitation, University of Colorado School of Medicine, Aurora, CO
| | - Peter J Lennarson
- Department of Neurosurgery, University of Colorado School of Medicine, Aurora, CO
| | - Enrico Rejc
- Department of Neurosurgery, University of Louisville School of Medicine, Louisville, KY; Department of Medicine, University of Udine, Udine, Italy
| | | | | | | | - Kenneth A Weber
- Department of Anesthesiology, Perioperative and Pain Medicine, Stanford University School of Medicine, Palo Alto, CA
| |
Collapse
|
6
|
Draganich C, Weber KA, Thornton WA, Berliner JC, Sevigny M, Charlifue S, Tefertiller C, Smith AC. Predicting Outdoor Walking 1 Year After Spinal Cord Injury: A Retrospective, Multisite External Validation Study. J Neurol Phys Ther 2023; 47:155-161. [PMID: 36630206 PMCID: PMC10329972 DOI: 10.1097/npt.0000000000000428] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/12/2023]
Abstract
BACKGROUND AND PURPOSE Predicting future outdoor walking ability after spinal cord injury (SCI) is important, as this is associated with community engagement and social participation. A clinical prediction rule (CPR) was derived for predicting outdoor walking 1 year after SCI. While promising, this CPR has not been validated, which is necessary to establish its clinical value. The objective of this study was to externally validate the CPR using a multisite dataset. METHODS This was a retrospective analysis of US SCI Model Systems data from 12 centers. L3 motor score, L5 motor score, and S1 sensory score were used as predictor variables. The dataset was split into testing and training datasets. The testing dataset was used as a holdout dataset to provide an unbiased estimate of prediction performance. The training dataset was used to determine the optimal CPR threshold through a "leave-one-site-out" cross-validation framework. The primary outcome was self-reported outdoor walking ability 1 year after SCI. RESULTS A total of 3721 participants' data were included. Using the optimal CPR threshold (CPR ≥ 33 threshold value), we were able to predict outdoor walking 1 year with high cross-validated accuracy and prediction performance. For the entire dataset, area under receiver operator characteristic curve was 0.900 (95% confidence interval: 0.890-0.910; P < 0.0001). DISCUSSION AND CONCLUSIONS The outdoor walking CPR has been externally validated. Future research should conduct a clinical outcomes and cost-benefit impact analysis for implementing this CPR. Our results support that clinicians may use this 3-variable CPR for prediction of future outdoor walking ability.Video Abstract available for more insights from the authors (see the Video, Supplemental Digital Content 1, available at: http://links.lww.com/JNPT/A411 ).
Collapse
Affiliation(s)
- Christina Draganich
- University of Colorado School of Medicine, Department of Physical Medicine and Rehabilitation, Aurora, CO USA
- Craig Hospital, Englewood, CO USA
| | - Kenneth A. Weber
- Stanford University School of Medicine, Department of Anesthesiology, Perioperative and Pain Medicine, Palo Alto, CA USA
| | - Wesley A. Thornton
- University of Colorado School of Medicine, Department of Physical Medicine and Rehabilitation, Aurora, CO USA
| | - Jeffrey C. Berliner
- University of Colorado School of Medicine, Department of Physical Medicine and Rehabilitation, Aurora, CO USA
| | | | | | | | - Andrew C. Smith
- University of Colorado School of Medicine, Department of Physical Medicine and Rehabilitation, Aurora, CO USA
| |
Collapse
|