Development of a machine-learning based model for predicting multidimensional outcome after surgery for degenerative disorders of the spine

Scoping Review of Machine Learning and Patient-Reported Outcomes in Spine Surgery

Machine learning is an evolving branch of artificial intelligence that is being applied in neurosurgical research. In spine surgery, machine learning has been used for radiographic characterization of cranial and spinal pathology and in predicting postoperative outcomes such as complications, functional recovery, and pain relief. A relevant application is the investigation of patient-reported outcome measures (PROMs) after spine surgery. Although a multitude of PROMs have been described and validated, there is currently no consensus regarding which questionnaires should be utilized. Additionally, studies have reported varying degrees of accuracy in predicting patient outcomes based on questionnaire responses. PROMs currently lack standardization, which renders them difficult to compare across studies. The purpose of this manuscript is to identify applications of machine learning to predict PROMs after spine surgery.

Revolutionizing spinal interventions: a systematic review of artificial intelligence technology applications in contemporary surgery

Leveraging its ability to handle large and complex datasets, artificial intelligence can uncover subtle patterns and correlations that human observation may overlook. This is particularly valuable for understanding the intricate dynamics of spinal surgery and its multifaceted impacts on patient prognosis. This review aims to delineate the role of artificial intelligence in spinal surgery. A search of the PubMed database from 1992 to 2023 was conducted using relevant English publications related to the application of artificial intelligence in spinal surgery. The search strategy involved a combination of the following keywords: "Artificial neural network," "deep learning," "artificial intelligence," "spinal," "musculoskeletal," "lumbar," "vertebra," "disc," "cervical," "cord," "stenosis," "procedure," "operation," "surgery," "preoperative," "postoperative," and "operative." A total of 1,182 articles were retrieved. After a careful evaluation of abstracts, 90 articles were found to meet the inclusion criteria for this review. Our review highlights various applications of artificial neural networks in spinal disease management, including (1) assessing surgical indications, (2) assisting in surgical procedures, (3) preoperatively predicting surgical outcomes, and (4) estimating the occurrence of various surgical complications and adverse events. By utilizing these technologies, surgical outcomes can be improved, ultimately enhancing the quality of life for patients.

Patient-reported outcome of lumbar decompression with instrumented fusion for low-grade spondylolisthesis: influence of pathology and baseline symptoms

INTRODUCTION

Low-grade isthmic and degenerative spondylolisthesis (DS) of the lumbar spine are distinct pathologies but both can be treated with lumbar decompression with fusion. In a very large cohort, we compared patient-reported outcome in relation to the pathology and chief complaint at baseline.

METHODS

This was a retrospective analysis using the EUROSPINE Spine Tango Registry. We included 582 patients (age 60 ± 15 years; 65% female), divided into four groups based on two variables: type of spondylolisthesis and chief pain complaint (leg pain (LP) versus back pain). Patients completed the COMI preoperatively and up to 5 years follow-up (FU), and rated global treatment outcome (GTO). Regression models were used to predict COMI-scores at FU. Pain scores and satisfaction ratings were analysed.

RESULTS

All patients experienced pronounced reductions in COMI scores. Relative to the other groups, the DS-LP group showed between 5% and 11% greater COMI score reduction (p < 0.01 up to 2 years' FU). This group also performed best with respect to pain outcomes and satisfaction. Long-term GTO was 93% at the 5 year FU, compared with between 82% and 86% in the other groups.

CONCLUSION

Regardless of the type of spondylolisthesis, all groups experienced an improvement in COMI score after surgery. Patients with DS and LP as their chief complaint appear to benefit more than other patients. These results are the first to show that the type of the spondylolisthesis and its chief complaint have an impact on surgical outcome. They will be informative for the consent process prior to surgery and can be used to build predictive models for individual outcome.

The use of the Core Yellow Flags Index for the assessment of psychosocial distress in patients undergoing surgery of the cervical spine

BACKGROUND

Psychosocial distress (the presence of yellow flags) has been linked to poor outcomes in spine surgery. The Core Yellow Flags Index (CYFI), a short instrument assessing the 4 main yellow flags, was developed for use in patients undergoing lumbar spine surgery. This study evaluated its ability to predict outcome in patients undergoing cervical spine surgery.

METHODS

Patients with degenerative spinal disorders (excluding myelopathy) operated in one centre, from 2015 to 2019, were asked to complete the CYFI at baseline and the Core Outcome Measures Index (COMI) at baseline and 3 and 12 months after surgery. The relationship between CYFI and COMI scores at baseline as well as the predictive ability of the CYFI on the COMI follow-up scores were tested using structural equation modelling.

RESULTS

From 731 eligible patients, 547 (61.0 ± 12.5 years; 57.2% female) completed forms at all three timepoints. On a cross-sectional basis, preoperative CYFI and COMI scores were highly correlated (β = 0.54, in men and 0.51 in women; each p < 0.001). CYFI added significantly and independently to the prediction of COMI at 3 months' FU in men (β = 0.36) and 12 months' FU in men and women (both β = 0.20) (all p < 0.001).

CONCLUSION

The CYFI had a low to moderate but significant and independent association with cervical spine surgery outcomes. Implementing the CYFI in the preoperative workup of these patients could help refine outcome predictions and better manage patient expectations.

Deep Learning Prediction of Cervical Spine Surgery Revision Outcomes Using Standard Laboratory and Operative Variables

BACKGROUND

Cervical spine procedures represent a major proportion of all spine surgery. Mitigating the revision rate following cervical procedures requires careful patient selection. While complication risk has successfully been predicted, revision risk has proven more challenging. This is likely due to the absence of granular variables in claims databases. The objective of this study was to develop a state-of-the-art model of revision prediction of cervical spine surgery using laboratory and operative variables.

METHODS

Using the Stanford Research Repository, patients undergoing a cervical spine procedure between 2016 and 2022 were identified (N = 3151), and recent laboratory values were collected. Patients were classified into separate cohorts by revision outcome and time frame. Machine and deep learning models were trained to predict each revision outcome from laboratory and operative variables.

RESULTS

Red blood cell count, hemoglobin, hematocrit, mean corpuscular hemoglobin concentration, red blood cell distribution width, platelet count, carbon dioxide, anion gap, and calcium all were significantly associated with ≥1 revision cohorts. For the prediction of 3-month revision, the deep neural network achieved an area under the receiver operating characteristic curve of 0.833. The model demonstrated increased performance for anterior versus posterior and arthrodesis versus decompression procedures.

CONCLUSIONS

Our deep learning approach successfully predicted 3-month revision outcomes from demographic variables, standard laboratory values, and operative variables in a cervical spine surgery cohort. This work used standard laboratory values and operative codes as meaningful predictive variables for revision outcome prediction. The increased performance on certain procedures evidences the need for careful development and validation of one-size-fits-all risk scores for spine procedures.

Clinical Decision Support System Used in Spinal Disorders: Scoping Review

BACKGROUND

Spinal disorders are highly prevalent worldwide with high socioeconomic costs. This cost is associated with the demand for treatment and productivity loss, prompting the exploration of technologies to improve patient outcomes. Clinical decision support systems (CDSSs) are computerized systems that are increasingly used to facilitate safe and efficient health care. Their applications range in depth and can be found across health care specialties.

OBJECTIVE

This scoping review aims to explore the use of CDSSs in patients with spinal disorders.

METHODS

We used the Joanna Briggs Institute methodological guidance for this scoping review and reported according to the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) statement. Databases, including PubMed, Embase, Cochrane, CINAHL, Web of Science, Scopus, ProQuest, and PsycINFO, were searched from inception until October 11, 2022. The included studies examined the use of digitalized CDSSs in patients with spinal disorders.

RESULTS

A total of 4 major CDSS functions were identified from 31 studies: preventing unnecessary imaging (n=8, 26%), aiding diagnosis (n=6, 19%), aiding prognosis (n=11, 35%), and recommending treatment options (n=6, 20%). Most studies used the knowledge-based system. Logistic regression was the most commonly used method, followed by decision tree algorithms. The use of CDSSs to aid in the management of spinal disorders was generally accepted over the threat to physicians' clinical decision-making autonomy.

CONCLUSIONS

Although the effectiveness was frequently evaluated by examining the agreement between the decisions made by the CDSSs and the health care providers, comparing the CDSS recommendations with actual clinical outcomes would be preferable. In addition, future studies on CDSS development should focus on system integration, considering end user's needs and preferences, and external validation and impact studies to assess effectiveness and generalizability.

TRIAL REGISTRATION

OSF Registries osf.io/dyz3f; https://osf.io/dyz3f.

How well can the clinician appraise the patient's perception of the severity and impact of their back problem?

PURPOSE

A main concern of patients with back problems is pain and its impact on function and quality of life. These are subjective phenomena, and should be probed during the clinical consultation so that the physician can ascertain the extent of the problem. This study evaluated the agreement between clinicians' and patients' independent ratings of patient status on the Core Outcome Measures Index (COMI).

METHODS

This was an analysis of the data from 5 spine specialists and 108 patients, in two centres. Prior to the consultation, the patient completed the COMI. After the consultation, the clinician (blind to the patient's version) also completed a COMI. Concordance was assessed by % agreement, Kappa values, Bland-Altman plots, Spearman rank, Intraclass Correlation Coefficients and comparisons of mean values, as appropriate.

RESULTS

Agreement regarding the "main problem" (back pain, leg/buttock pain, sensory disturbances, other) was 83%, Kappa = 0.70 (95%CI 0.58-0.81). Moderate/strong correlations were found between the doctors' and patients' COMI-item ratings (0.48-0.74; p < 0.0001), although compared with the patients' ratings the doctors systematically underestimated absolute values for leg pain (p = 0.002) and dissatisfaction with symptom state (p = 0.002), and overestimated how much the patient's function was impaired (p = 0.029).

CONCLUSION

The doctors were able to ascertain the location of the main problem and the multidimensional outcome score with good accuracy, but some individual domains were systematically underestimated (pain, symptom-specific well-being) or overestimated (impairment of function). More detailed/direct questioning on these domains during the consultation might deliver a better appreciation of the impact of the back problem on the patient's daily life.

Classification and prediction of spinal disease based on the SMOTE-RFE-XGBoost model

Spinal diseases are killers that cause long-term disturbance to people with complex and diverse symptoms and may cause other conditions. At present, the diagnosis and treatment of the main diseases mainly depend on the professional level and clinical experience of doctors, which is a breakthrough problem in the field of medicine. This article proposes the SMOTE-RFE-XGBoost model, which takes the physical angle of human bone as the research index for feature selection and classification model construction to predict spinal diseases. The research process is as follows: two groups of people with normal and abnormal spine conditions are taken as the research objects of this article, and the synthetic minority oversampling technique (SMOTE) algorithm is used to address category imbalance. Three methods, least absolute shrinkage and selection operator (LASSO), tree-based feature selection, and recursive feature elimination (RFE), are used for feature selection. Logistic regression (LR), support vector machine (SVM), parsimonious Bayes, decision tree (DT), random forest (RF), gradient boosting tree (GBT), extreme gradient boosting (XGBoost), and ridge regression models are used to classify the samples, construct single classification models and combine classification models and rank the feature importance. According to the accuracy and mean square error (MSE) values, the SMOTE-RFE-XGBoost combined model has the best classification, with accuracy, MSE and F1 values of 97.56%, 0.1111 and 0.8696, respectively. The importance of four indicators, lumbar slippage, cervical tilt, pelvic radius and pelvic tilt, was higher.

Does loss to follow-up lead to an overestimation of treatment success? Findings from a spine surgery registry of over 15,000 patients

PURPOSE

Patient-reported outcome measures (PROMs) are integral to the assessment of treatment success, but loss to follow-up (attrition) may lead to bias in the results reported. We sought to evaluate the extent, nature and implications of attrition in a long-established, single-centre spine registry.

METHODS

The registry contained the data of 15,264 consecutive spine surgery patients. PROMs included the Core Outcome Measures Index (COMI) and a rating of the Global Treatment Outcome (GTO) and Satisfaction with Care. Baseline characteristics associated with returning a 12-month PROM (= "responder") were analysed (logistic regression). The 3-month outcomes of 12-month responders versus 12-month non-responders were compared (ANOVA and Chi-square).

RESULTS

In total, 14,758/15,264 (97%) patients (60 ± 17y; 46% men) had consented to the use of their registry data for research. Preoperative, 3-month post-operative and 12-month post-operative PROMs were returned by 91, 90 and 86%, respectively. Factors associated with being a 12-month responder included: greater age, born in the country of the study, no private/semi-private insurance, better baseline status (lower COMI score), fewer previous surgeries, less comorbidity and no perioperative medical complications. 12-month non-responders had shown significantly worse outcomes in their 3-month PROMs than had 12-month responders (respectively, 66% vs 80% good GTO ("treatment helped/helped a lot"); 77% vs 88% satisfied/very satisfied; and 49% vs 63% achieved MCIC on COMI).

CONCLUSION

Although attrition in this cohort was relatively low, 12-month non-responders displayed distinctive characteristics and their early outcomes were significantly worse than those of 12-month responders. If loss to follow-up is not addressed, treatment success will likely be overestimated, with erroneously optimistic results being reported.

Developing a triage predictive model for access to a spinal surgeon using clinical variables and natural language processing of radiology reports

PURPOSE

To utilize natural language processing (NLP) of MRI reports and various clinical variables to develop a preliminary model predictive of the need for surgery in patients with low back and neck pain. Such a model would be beneficial for informing clinical practice decisions and help reduce the number of unnecessary surgical referrals, streamlining the surgical process.

METHODS

A historical cohort study was conducted using de-identified data from patients referred to a spine assessment clinic. Various demographic, clinical, and radiological variables were included as potential predictors. Full-text radiology reports of patients' MRI findings were vectorized using NLP before applying machine learning algorithms to develop models predicting who underwent surgery. Outputs from these models were then entered into a logistic regression model with clinical variables to develop a preliminary model predictive of surgical recommendations.

RESULTS

Of the 398 patients assessed, 71 underwent spine surgery. NLP variables were significant predictors in univariate analysis but did not remain in the final logistic regression model. An outcome of receiving surgery was predicted by a primary symptom of low back and leg pain (adjusted odds ratio 2.81), distal pain indicated by a pain diagram (adjusted odds ratio 2.49) and self-reported difficulties walking (adjusted odds ratio 2.73).

CONCLUSION

A logistic regression model was created to predict which patients may require spine surgery. Simple clinical variables appeared more predictive than variables created using NLP. However, additional research with more data samples is needed to validate this model and fully evaluate the usefulness of NLP for this task.