Prediction of the risk of C5 palsy after posterior laminectomy and fusion with cervical myelopathy using a support vector machine: an analysis of 184 consecutive patients

Limitations in Evaluating Machine Learning Models for Imbalanced Binary Outcome Classification in Spine Surgery: A Systematic Review

Clinical prediction models for spine surgery applications are on the rise, with an increasing reliance on machine learning (ML) and deep learning (DL). Many of the predicted outcomes are uncommon; therefore, to ensure the models' effectiveness in clinical practice it is crucial to properly evaluate them. This systematic review aims to identify and evaluate current research-based ML and DL models applied for spine surgery, specifically those predicting binary outcomes with a focus on their evaluation metrics. Overall, 60 papers were included, and the findings were reported according to the PRISMA guidelines. A total of 13 papers focused on lengths of stay (LOS), 12 on readmissions, 12 on non-home discharge, 6 on mortality, and 5 on reoperations. The target outcomes exhibited data imbalances ranging from 0.44% to 42.4%. A total of 59 papers reported the model's area under the receiver operating characteristic (AUROC), 28 mentioned accuracies, 33 provided sensitivity, 29 discussed specificity, 28 addressed positive predictive value (PPV), 24 included the negative predictive value (NPV), 25 indicated the Brier score with 10 providing a null model Brier, and 8 detailed the F1 score. Additionally, data visualization varied among the included papers. This review discusses the use of appropriate evaluation schemes in ML and identifies several common errors and potential bias sources in the literature. Embracing these recommendations as the field advances may facilitate the integration of reliable and effective ML models in clinical settings.

The use of machine learning for predicting candidates for outpatient spine surgery: a review

While spine surgery has historically been performed in the inpatient setting, in recent years there has been growing interest in performing certain cervical and lumbar spine procedures on an outpatient basis. While conducting these procedures in the outpatient setting may be preferable for both the surgeon and the patient, appropriate patient selection is crucial. The employment of machine learning techniques for data analysis and outcome prediction has grown in recent years within spine surgery literature. Machine learning is a form of statistics often applied to large datasets that creates predictive models, with minimal to no human intervention, that can be applied to previously unseen data. Machine learning techniques may outperform traditional logistic regression with regards to predictive accuracy when analyzing complex datasets. Researchers have applied machine learning to develop algorithms to aid in patient selection for spinal surgery and to predict postoperative outcomes. Furthermore, there has been increasing interest in using machine learning to assist in the selection of patients who may be appropriate candidates for outpatient cervical and lumbar spine surgery. The goal of this review is to discuss the current literature utilizing machine learning to predict appropriate patients for cervical and lumbar spine surgery, candidates for outpatient spine surgery, and outcomes following these procedures.

Comparison of Surgical Outcomes of Anterior and Posterior Fusion Surgeries for K-line (-) Cervical Ossification of the Posterior Longitudinal Ligament: A Prospective Multicenter Study

STUDY DESIGN

A prospective multicenter study.

OBJECTIVE

The objective of this study is to compare the surgical outcomes of anterior and posterior fusion surgeries in patients with K-line (-) cervical ossification of the posterior longitudinal ligament (OPLL).

SUMMARY OF BACKGROUND DATA

Although laminoplasty is effective for patients with K-line (+) OPLL, fusion surgery is recommended for those with K-line (-) OPLL. However, whether the anterior or posterior approach is preferable for this pathology has not been effectively determined.

MATERIALS AND METHODS

A total of 478 patients with myelopathy due to cervical OPLL from 28 institutions were prospectively registered from 2014 to 2017 and followed up for two years. Of the 478 patients, 45 and 46 with K-line (-) underwent anterior and posterior fusion surgeries, respectively. After adjusting for confounders in baseline characteristics using a propensity score-matched analysis, 54 patients in both the anterior and posterior groups (27 patients each) were evaluated. Clinical outcomes were assessed using the cervical Japanese Orthopaedic Association and the Japanese Orthopaedic Association Cervical Myelopathy Evaluation Questionnaire.

RESULTS

Both approaches showed comparable neurological and functional recovery. The cervical range of motion was significantly restricted in the posterior group because of the large number of fused vertebrae compared with the anterior group. The incidence of surgical complications was comparable between the cohorts, but the posterior group demonstrated a higher frequency of segmental motor paralysis, whereas the anterior group more frequently reported postoperative dysphagia.

CONCLUSIONS

Clinical improvement was comparable between anterior and posterior fusion surgeries for patients with K-line (-) OPLL. The ideal surgical approach should be informed based on the balance between the surgeon's technical preference and the risk of complications.

A degradome-based prognostic signature that correlates with immune infiltration and tumor mutation burden in breast cancer

Introduction

Female breast cancer is the most common malignancy worldwide, with a high disease burden. The degradome is the most abundant class of cellular enzymes that play an essential role in regulating cellular activity. Dysregulation of the degradome may disrupt cellular homeostasis and trigger carcinogenesis. Thus we attempted to understand the prognostic role of degradome in breast cancer by means of establishing a prognostic signature based on degradome-related genes (DRGs) and assessed its clinical utility in multiple dimensions.

Methods

A total of 625 DRGs were obtained for analysis. Transcriptome data and clinical information of patients with breast cancer from TCGA-BRCA, METABRIC and GSE96058 were collected. NetworkAnalyst and cBioPortal were also utilized for analysis. LASSO regression analysis was employed to construct the degradome signature. Investigations of the degradome signature concerning clinical association, functional characterization, mutation landscape, immune infiltration, immune checkpoint expression and drug priority were orchestrated. Cell phenotype assays including colony formation, CCK8, transwell and wound healing were conducted in MCF-7 and MDA-MB-435S breast cancer cell lines, respectively.

Results

A 10-gene signature was developed and verified as an independent prognostic predictor combined with other clinicopathological parameters in breast cancer. The prognostic nomogram based on risk score (calculated based on the degradome signature) showed favourable capability in survival prediction and advantage in clinical benefit. High risk scores were associated with a higher degree of clinicopathological events (T4 stage and HER2-positive) and mutation frequency. Regulation of toll-like receptors and several cell cycle promoting activities were upregulated in the high-risk group. PIK3CA and TP53 mutations were dominant in the low- and high-risk groups, respectively. A significantly positive correlation was observed between the risk score and tumor mutation burden. The infiltration levels of immune cells and the expressions of immune checkpoints were significantly influenced by the risk score. Additionally, the degradome signature adequately predicted the survival of patients undergoing endocrinotherapy or radiotherapy. Patients in the low-risk group may achieve complete response after the first round of chemotherapy with cyclophosphamide and docetaxel, whereas patients in the high-risk group may benefit from 5-flfluorouracil. Several regulators of the PI3K/AKT/mTOR signaling pathway and the CDK family/PARP family were identified as potential molecular targets in the low- and high-risk groups, respectively. In vitro experiments further revealed that the knockdown of ABHD12 and USP41 significantly inhibit the proliferation, invasion and migration of breast cancer cells.

Conclusion

Multidimensional evaluation verified the clinical utility of the degradome signature in predicting prognosis, risk stratification and guiding treatment for patients with breast cancer.

A nomogram model for predicting ocular GVHD following allo-HSCT based on risk factors

OBJECTIVE

To develop and validate a nomogram model for predicting chronic ocular graft-versus-host disease (coGVHD) in patients after allogenic haematopoietic stem cell transplantation (allo-HSCT).

METHODS

This study included 61 patients who survived at least 100 days after allo-HSCT. Risk factors for coGVHD were screened using LASSO regression, then the variables selected were subjected to logistic regression. Nomogram was established to further confirm the risk factors for coGVHD. Receiver operating characteristic (ROC) curves were constructed to assess the performance of the predictive model with the training and test sets. Odds ratios and 95% confidence intervals (95% CIs) were calculated by using logistic regression analysis.

RESULTS

Among the 61 patients, 38 were diagnosed with coGVHD. We selected five texture features: lymphocytes (LYM) (OR = 2.26), plasma thromboplastin antecedent (PTA) (OR = 1.19), CD3 + CD25 + cells (OR = 1.38), CD3 + HLA-DR + cells (OR = 0.95), and the ocular surface disease index (OSDI) (OR = 1.44). The areas under the ROC curve (AUCs) of the nomogram with the training and test sets were 0.979 (95% CI, 0.895-1.000) and 0.969 (95% CI, 0.846-1.000), respectively.And the Hosmer-Lemeshow test was nonsignificant with the training (p = 0.9949) and test sets (p = 0.9691).

CONCLUSION

We constructed a nomogram that can assess the risk of coGVHD in patients after allo-HSCT and help minimize the irreversible loss of vision caused by the disease in high-risk populations.

Comparison of the effectiveness of different machine learning algorithms in predicting new fractures after PKP for osteoporotic vertebral compression fractures

BACKGROUND

The use of machine learning has the potential to estimate the probability of a second classification event more accurately than traditional statistical methods, and few previous studies on predicting new fractures after osteoporotic vertebral compression fractures (OVCFs) have focussed on this point. The aim of this study was to explore whether several different machine learning models could produce better predictions than logistic regression models and to select an optimal model.

METHODS

A retrospective analysis of 529 patients who underwent percutaneous kyphoplasty (PKP) for OVCFs at our institution between June 2017 and June 2020 was performed. The patient data were used to create machine learning (including decision trees (DT), random forests (RF), support vector machines (SVM), gradient boosting machines (GBM), neural networks (NNET), and regularized discriminant analysis (RDA)) and logistic regression models (LR) to estimate the probability of new fractures occurring after surgery. The dataset was divided into a training set (75%) and a test set (25%), and machine learning models were built in the training set after ten cross-validations, after which each model was evaluated in the test set, and model performance was assessed by comparing the area under the curve (AUC) of each model.

RESULTS

Among the six machine learning algorithms, except that the AUC of DT [0.775 (95% CI 0.728-0.822)] was lower than that of LR [0.831 (95% CI 0.783-0.878)], RA [0.953 (95% CI 0.927-0.980)], GBM [0.941 (95% CI 0.911-0.971)], SVM [0.869 (95% CI 0.827-0.910), NNET [0.869 (95% CI 0.826-0.912)], and RDA [0.890 (95% CI 0.851-0.929)] were all better than LR.

CONCLUSIONS

For prediction of the probability of new fracture after PKP, machine learning algorithms outperformed logistic regression, with random forest having the strongest predictive power.

A tool for predicting overall survival in patients with Ewing sarcoma: a multicenter retrospective study

OBJECTIVE

The aim of this study was to establish and validate a clinical prediction model for assessing the risk of metastasis and patient survival in Ewing's sarcoma (ES).

METHODS

Patients diagnosed with ES from the Surveillance, Epidemiology and End Results (SEER) database for the period 2010-2016 were extracted, and the data after exclusion of vacant terms was used as the training set (n=767). Prediction models predicting patients' overall survival (OS) at 1 and 3 years were created by cox regression analysis and visualized using Nomogram and web calculator. Multicenter data from four medical institutions were used as the validation set (n=51), and the model consistency was verified using calibration plots, and receiver operating characteristic (ROC) verified the predictive ability of the model. Finally, a clinical decision curve was used to demonstrate the clinical utility of the model.

RESULTS

The results of multivariate cox regression showed that age, , bone metastasis, tumor size, and chemotherapy were independent prognostic factors of ES patients. Internal and external validation results: calibration plots showed that the model had a good agreement for patient survival at 1 and 3 years; ROC showed that it possessed a good predictive ability and clinical decision curve proved that it possessed good clinical utility.

CONCLUSIONS

The tool built in this paper to predict 1- and 3-year survival in ES patients ( https://drwenleli0910.shinyapps.io/EwingApp/ ) has a good identification and predictive power.

An External-Validated Prediction Model to Predict Lung Metastasis among Osteosarcoma: A Multicenter Analysis Based on Machine Learning

BACKGROUND

Lung metastasis greatly affects medical therapeutic strategies in osteosarcoma. This study aimed to develop and validate a clinical prediction model to predict the risk of lung metastasis among osteosarcoma patients based on machine learning (ML) algorithms.

METHODS

We retrospectively collected osteosarcoma patients from the Surveillance Epidemiology and End Results (SEER) database and from four hospitals in China. Six ML algorithms, including logistic regression (LR), gradient boosting machine (GBM), extreme gradient boosting (XGBoost), random forest (RF), decision tree (DT), and multilayer perceptron (MLP), were applied to build predictive models for predicting lung metastasis using patient's demographics, clinical characteristics, and therapeutic variables from the SEER database. The model was internally validated using 10-fold cross-validation to calculate the mean area under the curve (AUC) and the model was externally validated using the Chinese multicenter osteosarcoma data. Relative importance ranking of predictors was plotted to understand the importance of each predictor in different ML algorithms. The correlation heat map of predictors was plotted to understand the correlation of each predictor, selecting the 10-fold cross-validation with the highest AUC value in the external validation ROC curve to build a web calculator.

RESULTS

Of all enrolled patients from the SEER database, 17.73% (194/1094) developed lung metastasis. The multiple logistic regression analysis showed that sex, N stage, T stage, surgery, and bone metastasis were all independent risk factors for lung metastasis. In predicting lung metastasis, the mean AUCs of the six ML algorithms ranged from 0.711 to 0.738 in internal validation and 0.697 to 0.729 in external validation. Among the six ML algorithms, the extreme gradient boosting (XGBoost) model had the highest AUC value with an average internal AUC of 0.738 and an external AUC of 0.729. The best performing ML algorithm model was used to build a web calculator to facilitate clinicians to calculate the risk of lung metastasis for each patient.

CONCLUSIONS

The XGBoost model may have the best prediction effect and the online calculator based on this model can help doctors to determine the lung metastasis risk of osteosarcoma patients and help to make individualized medical strategies.

An Evolution Gaining Momentum—The Growing Role of Artificial Intelligence in the Diagnosis and Treatment of Spinal Diseases

In recent years, applications using artificial intelligence have been gaining importance in the diagnosis and treatment of spinal diseases. In our review, we describe the basic features of artificial intelligence which are currently applied in the field of spine diagnosis and treatment, and we provide an orientation of the recent technical developments and their applications. Furthermore, we point out the possible limitations and challenges in dealing with such technological advances. Despite the momentary limitations in practical application, artificial intelligence is gaining ground in the field of spine treatment. As an applying physician, it is therefore necessary to engage with it in order to benefit from those advances in the interest of the patient and to prevent these applications being misused by non-medical partners.

Artificial Intelligence-Driven Prediction Modeling and Decision Making in Spine Surgery Using Hybrid Machine Learning Models

Healthcare systems worldwide generate vast amounts of data from many different sources. Although of high complexity for a human being, it is essential to determine the patterns and minor variations in the genomic, radiological, laboratory, or clinical data that reliably differentiate phenotypes or allow high predictive accuracy in health-related tasks. Convolutional neural networks (CNN) are increasingly applied to image data for various tasks. Its use for non-imaging data becomes feasible through different modern machine learning techniques, converting non-imaging data into images before inputting them into the CNN model. Considering also that healthcare providers do not solely use one data modality for their decisions, this approach opens the door for multi-input/mixed data models which use a combination of patient information, such as genomic, radiological, and clinical data, to train a hybrid deep learning model. Thus, this reflects the main characteristic of artificial intelligence: simulating natural human behavior. The present review focuses on key advances in machine and deep learning, allowing for multi-perspective pattern recognition across the entire information set of patients in spine surgery. This is the first review of artificial intelligence focusing on hybrid models for deep learning applications in spine surgery, to the best of our knowledge. This is especially interesting as future tools are unlikely to use solely one data modality. The techniques discussed could become important in establishing a new approach to decision-making in spine surgery based on three fundamental pillars: (1) patient-specific, (2) artificial intelligence-driven, (3) integrating multimodal data. The findings reveal promising research that already took place to develop multi-input mixed-data hybrid decision-supporting models. Their implementation in spine surgery may hence be only a matter of time.