Prediction Model of Osteonecrosis of the Femoral Head After Femoral Neck Fracture: Machine Learning-Based Development and Validation Study

Using Artificial Intelligence to predict outcomes of operatively managed neck of femur fractures

Aims/Background Patients with neck of femur fractures present a tremendous public health problem that leads to a high incidence of death and dysfunction. An essential factor is the postoperative length of stay, which heavily impacts hospital costs and the quality of care. As an extension of traditional statistical methods, machine learning (ML) provides the possibility of accurately predicting the length of hospital stay. This review assesses how machine learning can effectively use healthcare data to predict the outcomes of patients with operatively managed neck of femurs. Methods A narrative literature review on the use of Artificial Intelligence to predict outcomes in the neck of femurs was undertaken to understand the field and critical considerations of its application. The papers and any relevant references were scrutinised using the specific inclusion and exclusion criteria to produce papers that were used in the analysis. Results Thirteen papers were used in the analysis. The critical themes recognised the different models, the 'backbox' conundrum, predictor identification, validation methodology and the need to improve efficiency and quality of care. Through reviewing the themes in this paper, current issues, and potential avenues of advancing the field are explored. Conclusions This review has demonstrated that the use of machine learning in Orthopaedic pathways is in its infancy. Further work is needed to leverage this technology effectively to improve outcomes.

[Correlation analysis between combined deflection angle and osteonecrosis of femoral head after femoral neck fracture]

Objective

To evaluate the correlation between pelvic incidence (PI) angle, hip deflection angle (HDA), combined deflection angle (CDA) and osteonecrosis of the femoral head (ONFH) after femoral neck fracture, in order to explore early predictive indicators for ONFH occurrence after femoral neck fracture.

Methods

A study was conducted on patients with femoral neck fractures who underwent cannulated screw internal fixation between December 2018 and December 2020. Among them, 208 patients met the selection criteria and were included in the study. According to the occurrence of ONFH, the patients were allocated into ONFH group and non-NOFH group. PI, HDA, and CDA were measured based on the anteroposterior X-ray films of pelvis and axial X-ray films of the affected hip joint before operation, and the differences between the two groups were compared. The receiver operating characteristic curve (ROC) was used to evaluate the value of the above imaging indicators in predicting the occurrence of ONFH.

Results

Among the 208 patients included in the study, 84 patients experienced ONFH during follow-up (ONFH group) and 124 patients did not experience ONFH (non-ONFH group). In the non-ONFH group, there were 59 males and 65 females, the age was 18-86 years (mean, 53.9 years), and the follow-up time was 18-50 months (mean, 33.2 months). In the ONFH group, there were 37 males and 47 females, the age was 18-76 years (mean, 51.6 years), and the follow-up time was 8-45 months (mean, 22.1 months). The PI, HDA, and CDA were significantly larger in the ONFH group than in the non-ONFH group ( P<0.05). ROC curve analysis showed that the critical value of PI was 19.82° (sensitivity of 40.5%, specificity of 86.3%, P<0.05); the critical value of HDA was 20.94° (sensitivity of 77.4%, specificity of 75.8%, P<0.05); and the critical value of CDA was 39.16° (sensitivity of 89.3%, specificity of 83.1%, P<0.05).

Conclusion

There is a correlation between PI, HDA, CDA and the occurrence of ONFH after femoral neck fracture, in which CDA can be used as an important reference indicator. Patients with CDA≥39.16° have a higher risk of ONFH after femoral neck fracture.

Experimental analysis of femoral head intraosseous vascular anastomosis in the treatment of porcine subcapital femoral neck fractures

Introduction

Femoral neck fractures are challenging injuries associated with a compromised blood supply to the femoral head, leading to a high risk of avascular necrosis and poor clinical outcomes. This study aimed to investigate the efficacy of femoral head intraosseous vascular anastomosis in the treatment of porcine sub-capital femoral neck fractures.

Methods

Ten Landrace pigs were used as experimental animal models. The femoral head was completely removed after femoral neck sub-cephalic fracture. It was fixed on the medial side of the knee joint, and the blood supply to the femoral head was reconstructed by anastomosing the femoral head vessels. One week later, blood flow in the femoral head was observed by borehole, digital subtraction angiography examination, and hematoxylin and eosin staining. Further, terminal deoxynucleotidyl transferase-mediated dUTP nick-end labelling tests were performed to detect pathological changes in the femoral head.

Results

After one-week, digital subtraction angiography of the femoral head revealed a blood circulation rate of 70 %, and the blood seepage rate of the borehole was 80 %. Hematoxylin and eosin staining and terminal deoxynucleotidyl transferase-mediated dUTP nick-end labelling test results showed that necrosis of bone marrow cells in the experimental group was significantly improved compared to that in the control group.

Discussion

This study highlights the potential benefits of femoral head intraosseous vascular anastomosis in the treatment of porcine sub-capital femoral neck fractures. Further research and clinical trials are warranted to validate these findings and to explore the translational potential of this technique in human patients.

An machine learning model to predict quality of life subtypes of disabled stroke survivors

OBJECTIVE

Stroke causes serious physical disability with impaired quality of life (QoL) and heavy burden on health. The goal of this study is to explore the impaired QoL typologies and their predicting factors in physically disabled stroke survivors with machine learning approach.

METHODS

Non-negative matrix factorization (NMF) was applied to clustering 308 physically disabled stroke survivors in rural China based on their responses on the short form 36 (SF-36) assessment of quality of life. Principal component analysis (PCA) was conducted to differentiate the subtypes, and the Boruta algorithm was used to identify the variables relevant to the categorization of two subtypes. A gradient boosting machine(GBM) and local interpretable model-agnostic explanation (LIME) algorithms were used to apply to interpret the variables that drove subtype predictions.

RESULTS

Two distinct subtypes emerged, characterized by short form 36 (SF-36) domains. The feature difference between worsen QoL subtype and better QoL subtype was as follows: role-emotion (RE), body pain (BP) and general health (GH), but not physical function (PF); the most relevant predictors of worsen QoL subtypes were help from others, followed by opportunities for community activity and rehabilitation needs, rather than disability severity or duration since stroke.

INTERPRETATION

The results suggest that the rehabilitation programs should be tailored toward their QoL clustering feature; body pain and emotional-behavioral problems are more crucial than motor deficit; stroke survivors with worsen QoL subtype are most in need of social support, return to community, and rehabilitation.

Systematic review of machine-learning models in orthopaedic trauma

Aims

Machine-learning (ML) prediction models in orthopaedic trauma hold great promise in assisting clinicians in various tasks, such as personalized risk stratification. However, an overview of current applications and critical appraisal to peer-reviewed guidelines is lacking. The objectives of this study are to 1) provide an overview of current ML prediction models in orthopaedic trauma; 2) evaluate the completeness of reporting following the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) statement; and 3) assess the risk of bias following the Prediction model Risk Of Bias Assessment Tool (PROBAST) tool.

Methods

A systematic search screening 3,252 studies identified 45 ML-based prediction models in orthopaedic trauma up to January 2023. The TRIPOD statement assessed transparent reporting and the PROBAST tool the risk of bias.

Results

A total of 40 studies reported on training and internal validation; four studies performed both development and external validation, and one study performed only external validation. The most commonly reported outcomes were mortality (33%, 15/45) and length of hospital stay (9%, 4/45), and the majority of prediction models were developed in the hip fracture population (60%, 27/45). The overall median completeness for the TRIPOD statement was 62% (interquartile range 30 to 81%). The overall risk of bias in the PROBAST tool was low in 24% (11/45), high in 69% (31/45), and unclear in 7% (3/45) of the studies. High risk of bias was mainly due to analysis domain concerns including small datasets with low number of outcomes, complete-case analysis in case of missing data, and no reporting of performance measures.

Conclusion

The results of this study showed that despite a myriad of potential clinically useful applications, a substantial part of ML studies in orthopaedic trauma lack transparent reporting, and are at high risk of bias. These problems must be resolved by following established guidelines to instil confidence in ML models among patients and clinicians. Otherwise, there will remain a sizeable gap between the development of ML prediction models and their clinical application in our day-to-day orthopaedic trauma practice.

A prediction model for osteonecrosis of femoral head after internal fixation with multiple cannulated compression screws for adult femoral neck fractures

OBJECTIVES

This study aims to investigate the high-risk factors for osteonecrosis of the femoral head (ONFH) after internal fixation with multiple cannulated compression screws for adult femoral neck fractures and to construct a prediction model.

PATIENTS AND METHODS

Between from January 2012 and December 2020, a total of 268 patients (138 males, 130 females; mean age: 53±10 years; range, 23 to 70 years) with ONFH who had complete follow-up data were included. Closed reduction in combination with open reduction were performed. All patients received internal fixation with multiple cannulated compression screws and were assigned to ONFH and non-ONFH groups. Logistic regression model was utilized to identify independent risk factors for postoperative ONFH, followed by constructing a nomogram prediction model. The predictive ability of the model was evaluated by receiver operating characteristic curve, Hosmer-Lemeshow test, and calibration curve.

RESULTS

Multivariate analysis revealed that older age (odds ratio [OR]: 2.307, 95% confidence interval [CI]: 1.295-4.108], Charlson Comorbidity Index (CCI) ≥2 (OR: 2.214, 95% CI: 1.035-4.739), fracture displacement (OR: 2.426, 95% CI: 1.122-5.247), unsatisfactory reduction (OR: 2.629, 95% CI: 1.275-5.423), postoperative removal of internal fixation implant (OR: 2.200, 95% CI: 1.051-4.604) were independent risk factors for postoperative ONFH (p<0.05). The nomogram prediction model constructed based on these clinical characteristics showed high predictive value (AUC=0.807) and consistency (p>0.05).

CONCLUSION

Age, comorbidity index, fracture type, reduction quality and postoperative removal of internal fixation implant are of utmost importance for postoperative ONFH in patients with femoral neck fractures. The established nomogram prediction model can accurately predict the occurrence of postoperative ONFH.

Predictive modeling of perioperative blood transfusion in lumbar posterior interbody fusion using machine learning

Background: Accurate estimation of perioperative blood transfusion risk in lumbar posterior interbody fusion is essential to reduce the number, cost, and complications associated with blood transfusions. Machine learning algorithms have the potential to outperform traditional prediction methods in predicting perioperative blood transfusion. This study aimed to construct a machine learning-based perioperative transfusion risk prediction model for lumbar posterior interbody fusion in order to improve the efficacy of surgical decision-making. Methods: We retrospectively collected clinical data on 1905 patients who underwent lumbar posterior interbody fusion surgery at the Second Hospital of Shanxi Medical University between January 2021 and March 2023. All the data was randomly divided into a training set and a validation set, and the "feature_importances" method provided by eXtreme Gradient Boosting (XGBoost) algorithm was applied to select statistically significant features on the training set to establish five machine learning prediction models. The optimal model was identified by utilizing the area under the curve (AUC) and the probability calibration curve on the validation set. Shapley additive explanations (SHAP) and local interpretable model-agnostic explanations (LIME) were employed for interpretable analysis of the optimal model. Results: In the postoperative outcomes of patients, the number of hospital days in the transfusion group was longer than that in the non-transfusion group. Additionally, the transfusion group experienced higher total hospital costs, 90-day readmission rates, and complication rates within 90 days after surgery than the non-transfusion group. A total of 9 features were selected for the models. The XGBoost model performed best with an AUC value of 0.958. The SHAP values showed that intraoperative blood loss, intraoperative fluid infusion, and number of fused segments were the top 3 most important features affecting perioperative blood transfusion in lumbar posterior interbody fusion. The LIME algorithm was used to interpret the individualized prediction. Conclusion: Surgery, ASA class, levels fused, total intraoperative blood loss, operative time, and preoperative Hb are viable predictors of perioperative blood transfusion in lumbar posterior interbody fusion. The XGBoost model has demonstrated superior predictive efficacy compared to the traditional logistic regression model, making it a more effective decision-making tool for perioperative blood transfusion.

Predictive value of clinical features and CT radiomics in the efficacy of hip preservation surgery with fibula allograft

BACKGROUND

Despite being an effective treatment for osteonecrosis of the femoral head (ONFH), hip preservation surgery with fibula allograft (HPS&FA) still experiences numerous failures. Developing a prediction model based on clinical and radiomics predictors holds promise for addressing this issue.

METHODS

This study included 112 ONFH patients who underwent HPS&FA and were randomly divided into training and validation cohorts. Clinical data were collected, and clinically significant predictors were identified using univariate and multivariate analyses to develop a clinical prediction model (CPM). Simultaneously, the least absolute shrinkage and selection operator method was employed to select optimal radiomics features from preoperative hip computed tomography images, forming a radiomics prediction model (RPM). Furthermore, to enhance prediction accuracy, a clinical-radiomics prediction model (CRPM) was constructed by integrating all predictors. The predictive performance of the models was evaluated using receiver operating characteristic curve (ROC), area under the curve (AUC), DeLong test, calibration curve, and decision curve analysis.

RESULTS

Age, Japanese Investigation Committee classification, postoperative use of glucocorticoids or alcohol, and non-weightbearing time were identified as clinical predictors. The AUC of the ROC curve for the CPM was 0.847 in the training cohort and 0.762 in the validation cohort. After incorporating radiomics features, the CRPM showed improved AUC values of 0.875 in the training cohort and 0.918 in the validation cohort. Decision curves demonstrated that the CRPM yielded greater medical benefit across most risk thresholds.

CONCLUSION

The CRPM serves as an efficient prediction model for assessing HPS&FA efficacy and holds potential as a personalized perioperative intervention tool to enhance HPS&FA success rates.

An effective prediction model based on XGBoost for the 12-month recurrence of AF patients after RFA

BACKGROUND

Atrial fibrillation (AF) is a common heart rhythm disorder that can lead to complications such as stroke and heart failure. Radiofrequency ablation (RFA) is a procedure used to treat AF, but it is not always successful in maintaining a normal heart rhythm. This study aimed to construct a clinical prediction model based on extreme gradient boosting (XGBoost) for AF recurrence 12 months after ablation.

METHODS

The 27-dimensional data of 359 patients with AF undergoing RFA in the First Affiliated Hospital of Soochow University from October 2018 to November 2021 were retrospectively analysed. We adopted the logistic regression, support vector machine (SVM), random forest (RF) and XGBoost methods to conduct the experiment. To evaluate the performance of the prediction, we used the area under the receiver operating characteristic curve (AUC), the area under the precision-recall curve (AP), and calibration curves of both the training and testing sets. Finally, Shapley additive explanations (SHAP) were utilized to explain the significance of the variables.

RESULTS

Of the 27-dimensional variables, ejection fraction (EF) of the left atrial appendage (LAA), N-terminal probrain natriuretic peptide (NT-proBNP), global peak longitudinal strain of the LAA (LAAGPLS), left atrial diameter (LAD), diabetes mellitus (DM) history, and female sex had a significant role in the predictive model. The experimental results demonstrated that XGBoost exhibited the best performance among these methods, and the accuracy, specificity, sensitivity, precision and F1 score (a measure of test accuracy) of XGBoost were 86.1%, 89.7%, 71.4%, 62.5% and 0.67, respectively. In addition, SHAP analysis also proved that the 6 parameters were decisive for the effect of the XGBoost-based prediction model.

CONCLUSIONS

We proposed an effective model based on XGBoost that can be used to predict the recurrence of AF patients after RFA. This prediction result can guide treatment decisions and help to optimize the management of AF.

Methods to predict osteonecrosis of femoral head after femoral neck fracture: a systematic review of the literature

BACKGROUND

Femoral neck fracture (FNF) is a very common traumatic disorder and a major cause of blood supply disruption to the femoral head, which may lead to a severe long-term complication, osteonecrosis of femoral head (ONFH). Early prediction and evaluation of ONFH after FNF could facilitate early treatment and may prevent or reverse the development of ONFH. In this review paper, we will review all the prediction methods reported in the previous literature.

METHODS

Studies on the prediction of ONFH after FNF were included in PubMed and MEDLINE databases with articles published before October 2022. Further screening criteria were conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses. This study highlights all the advantages and disadvantages of the prediction methods.

RESULTS

There were a total of 36 studies included, involving 11 methods to predict ONFH after FNF. Among radiographic imaging, superselective angiography could directly visualize the blood supply of the femoral head, but it is an invasive examination. As noninvasive detection methods, dynamic enhanced magnetic resonance imaging (MRI) and SPECT/CT are easy to operate, have a high sensitivity, and increase specificity. Though still at the early stage of development in clinical studies, micro-CT is a method of highly accurate quantification that can visualize femoral head intraosseous arteries. The prediction model relates to artificial intelligence and is easy to operate, but there is no consensus on the risk factors of ONFH. For the intraoperative methods, most of them are single studies and lack clinical evidence.

CONCLUSION

After reviewing all the prediction methods, we recommend using dynamic enhanced MRI or single photon emission computed tomography/computed tomography in combination with the intraoperative observation of bleeding from the holes of proximal cannulated screws to predict ONFH after FNF. Moreover, micro-CT is a promising imaging technique in clinical practice.

Patients With Femoral Neck Fractures Are at Risk for Conversion to Arthroplasty After Internal Fixation: A Machine-learning Algorithm

BACKGROUND

Femoral neck fractures are common and are frequently treated with internal fixation. A major disadvantage of internal fixation is the substantially high number of conversions to arthroplasty because of nonunion, malunion, avascular necrosis, or implant failure. A clinical prediction model identifying patients at high risk of conversion to arthroplasty may help clinicians in selecting patients who could have benefited from arthroplasty initially.

QUESTION/PURPOSE

What is the predictive performance of a machine-learning (ML) algorithm to predict conversion to arthroplasty within 24 months after internal fixation in patients with femoral neck fractures?

METHODS

We included 875 patients from the Fixation using Alternative Implants for the Treatment of Hip fractures (FAITH) trial. The FAITH trial consisted of patients with low-energy femoral neck fractures who were randomly assigned to receive a sliding hip screw or cancellous screws for internal fixation. Of these patients, 18% (155 of 875) underwent conversion to THA or hemiarthroplasty within the first 24 months. All patients were randomly divided into a training set (80%) and test set (20%). First, we identified 27 potential patient and fracture characteristics that may have been associated with our primary outcome, based on biomechanical rationale and previous studies. Then, random forest algorithms (an ML learning, decision tree-based algorithm that selects variables) identified 10 predictors of conversion: BMI, cardiac disease, Garden classification, use of cardiac medication, use of pulmonary medication, age, lung disease, osteoarthritis, sex, and the level of the fracture line. Based on these variables, five different ML algorithms were trained to identify patterns related to conversion. The predictive performance of these trained ML algorithms was assessed on the training and test sets based on the following performance measures: (1) discrimination (the model's ability to distinguish patients who had conversion from those who did not; expressed with the area under the receiver operating characteristic curve [AUC]), (2) calibration (the plotted estimated versus the observed probabilities; expressed with the calibration curve intercept and slope), and (3) the overall model performance (Brier score: a composite of discrimination and calibration).

RESULTS

None of the five ML algorithms performed well in predicting conversion to arthroplasty in the training set and the test set; AUCs of the algorithms in the training set ranged from 0.57 to 0.64, slopes of calibration plots ranged from 0.53 to 0.82, calibration intercepts ranged from -0.04 to 0.05, and Brier scores ranged from 0.14 to 0.15. The algorithms were further evaluated in the test set; AUCs ranged from 0.49 to 0.73, calibration slopes ranged from 0.17 to 1.29, calibration intercepts ranged from -1.28 to 0.34, and Brier scores ranged from 0.13 to 0.15.

CONCLUSION

The predictive performance of the trained algorithms was poor, despite the use of one of the best datasets available worldwide on this subject. If the current dataset consisted of different variables or more patients, the performance may have been better. Also, various reasons for conversion to arthroplasty were pooled in this study, but the separate prediction of underlying pathology (such as, avascular necrosis or nonunion) may be more precise. Finally, it may be possible that it is inherently difficult to predict conversion to arthroplasty based on preoperative variables alone. Therefore, future studies should aim to include more variables and to differentiate between the various reasons for arthroplasty.

LEVEL OF EVIDENCE

Level III, prognostic study.

The Application of Machine Learning in Predicting Mortality Risk in Patients With Severe Femoral Neck Fractures: Prediction Model Development Study

BACKGROUND

Femoral neck fracture (FNF) accounts for approximately 3.58% of all fractures in the entire body, exhibiting an increasing trend each year. According to a survey, in 1990, the total number of hip fractures in men and women worldwide was approximately 338,000 and 917,000, respectively. In China, FNFs account for 48.22% of hip fractures. Currently, many studies have been conducted on postdischarge mortality and mortality risk in patients with FNF. However, there have been no definitive studies on in-hospital mortality or its influencing factors in patients with severe FNF admitted to the intensive care unit.

OBJECTIVE

In this paper, 3 machine learning methods were used to construct a nosocomial death prediction model for patients admitted to intensive care units to assist clinicians in early clinical decision-making.

METHODS

A retrospective analysis was conducted using information of a patient with FNF from the Medical Information Mart for Intensive Care III. After balancing the data set using the Synthetic Minority Oversampling Technique algorithm, patients were randomly separated into a 70% training set and a 30% testing set for the development and validation, respectively, of the prediction model. Random forest, extreme gradient boosting, and backpropagation neural network prediction models were constructed with nosocomial death as the outcome. Model performance was assessed using the area under the receiver operating characteristic curve, accuracy, precision, sensitivity, and specificity. The predictive value of the models was verified in comparison to the traditional logistic model.

RESULTS

A total of 366 patients with FNFs were selected, including 48 cases (13.1%) of in-hospital death. Data from 636 patients were obtained by balancing the data set with the in-hospital death group to survival group as 1:1. The 3 machine learning models exhibited high predictive accuracy, and the area under the receiver operating characteristic curve of the random forest, extreme gradient boosting, and backpropagation neural network were 0.98, 0.97, and 0.95, respectively, all with higher predictive performance than the traditional logistic regression model. Ranking the importance of the feature variables, the top 10 feature variables that were meaningful for predicting the risk of in-hospital death of patients were the Simplified Acute Physiology Score II, lactate, creatinine, gender, vitamin D, calcium, creatine kinase, creatine kinase isoenzyme, white blood cell, and age.

CONCLUSIONS

Death risk assessment models constructed using machine learning have positive significance for predicting the in-hospital mortality of patients with severe disease and provide a valid basis for reducing in-hospital mortality and improving patient prognosis.