Abstract:Renming Chen, Qin Huang, Lihua Chen Department of Nephropathy and Rheumatology,The Central Hospital of Enshi Tujia and Miao Autonomous Prefecture, Enshi, 445000, People's Republic of China Correspondence: Lihua Chen, Department of Nephropathy and Rheumatology,The Central Hospital of Enshi Tujia and Miao Autonomous Prefecture, 158 Wuyang County Street, Enshi, Hubei, People's Republic of China, Tel +86 0718-8263471, Email Objective: Fracture is a critical unfavorable prognostic factor in patients with rheumatoid arthritis(RA) and osteoporosis. At present, models involving clinical indices that accurately predict fracture are still uncommon. We addressed this gap by developing machine learning (ML)-based predictive models to individualize the risk of fracture in elderly patients with RA and osteoporosis and to identify a high-risk group for fracture. Methods: 487 patients diagnosed with RA and osteoporosis at the Central Hospital of Enshi Tujia and Miao Autonomous Prefecture were randomly divided into a training cohort (used for building the model) and a validation cohort (used for validating the model). Five ML-assisted models were developed from candidate clinical features using two-step estimation methods. The receiver operating characteristic curve (ROC), decision curve analysis (DCA), and clinical impact curve (CIC) were performed to evaluate the robustness and clinical practicability of each model. Results: A total of twenty-two candidate variables were included, and the prediction model was established by an ML-based algorithm. The areas under the ROC curve (AUCs) of the random forest classifier (RFC) model, artificial neural network (ANN), support vector machine (SVM), eXtreme gradient boosting (XGBoost), decision tree (DT), probability of major osteoporotic fractures (PMOF), and probability of hip fracture (PHF) ranged from 0.695 to 0.878. Among them, RFC obtained the optimal prediction efficiency via adding serum selenium and clinical indices, that is, glucocorticoid, and erythrocyte sedimentation rate (ESR). Conclusion: Based on the classic clinical parameters, the fracture risk of RA patients with osteoporosis can be accurately predicted. In particular, RFC prediction model shows good discrimination ability in identifying high-risk patients with fracture. Keywords: rheumatoid arthritis, osteoporosis, fracture, machine learning algorithm, risk factor, predictive model Rheumatoid arthritis (RA) is a chronic inflammatory joint disease, which can lead to cartilage and bone damage and disability. 1,2 In addition, local and systemic bone loss is one of the major extraarticular complications of RA and leads to an increased risk of brittle fracture, which further impairs functional ability, quality of life, and life expectancy. 3,4 Alarmingly, the incidence rate of osteoporosis in elderly-onset rheumatoid arthritis (EORA) patients has an upward trend in clinical practice. 5,6 The secondary fracture and osteoporosis-related pain in RA patients with osteoporosis can directly lead to the reduction of the quality of life of patients. Therefore, timely fracture risk management is urgently needed for RA patients with osteoporosis. In the elderly, the presence of complications and the increase of drug-related adverse reactions pose specific treatment challenges. 6 Although glucocorticoids can reduce clinical activity in the short term and may also reduce structural damage in the medium term, these benefits are offset by many adverse reactions (such as osteoporosis, fracture, etc.). 7 Therefore, it is of positive significance for clinical treatment to clarify the fracture risk and related influencing factors of RA patients with osteoporosis. At least, to support the balance between over treatment and under treatment, it is necessary to study and propose many strategies to select patients who may omit fracture. Currently, most risk stratification systems and prediction models are based on classical clinical parameters. For example, a genome-wide polygenic risk score (PRS) was more strongly associated with the risk of fracture, which could define patients with high-risk fracture. 8 Susan et al established FRAX( ® ) calculations to predict the risk of fracture in patients with RA. 9 Samaneh et al developed the trabecular bone score (TBS) that predicts fracture risk better than bone mineral density (BMD). However, these risk stratification systems and prediction models do not seem to be suitable for accurately predicting fractures, especially in elderly RA patients with osteoporosis. It is worth mentioning that the factors related to bone health play an important role in preventing osteoporosis and its adverse consequences. For instance, a meta-analysis -Abstract Truncated-

A Novel Fracture Prediction Model Using Machine Learning in a Community‐Based Cohort

Clinical Value of Machine Learning Models in Early Osteoporosis Prediction Based on Demographic and Routine Biochemical Indicators

Machine Learning to Predict Long and Short Term Fracture Risk in Postmenopausal Women

Fracture risk prediction in postmenopausal women with traditional and machine learning models in a nationwide, prospective cohort study in Switzerland with validation in the UK Biobank

Osteoporosis, fracture and survival: Application of machine learning in breast cancer prediction models

Predicting fracture risk for elderly osteoporosis patients by hybrid machine learning model

Enhanced osteoporotic fracture prediction in postmenopausal women using Bayesian optimization of machine learning models with genetic risk score

Predicting Fracture Risk for Chinese Elderly Osteoporosis Patients by Hybrid Model

Predictive value of machine learning on fracture risk in osteoporosis: a systematic review and meta-analysis

Fragility Fractures in Postmenopausal Women: Development of 5-Year Prediction Models Using the FRISBEE Study

Development of a Spine X-Ray-Based Fracture Prediction Model Using a Deep Learning Algorithm

Osteoporotic Fracture Risk Assessment Using Bone Mineral Density in Korean: A Community-based Cohort Study

Predicting fragility fractures based on frailty and bone mineral density among rural community-dwelling older adults

Predicting fractures in an international cohort using risk factor algorithms without BMD

Developing Novel Prognostic Biomarkers for Multivariate Fracture Risk Prediction Algorithms

Development and Validation of Machine Learning Models for Prediction of Fracture Risk in Patients with Elderly-Onset Rheumatoid Arthritis

Machine Learning Algorithms for Predicting the Risk of Fracture in Patients with Diabetes in China

A novel clinical prediction model for hip fractures: a development and validation study in the total population of Sweden

A Computed Tomography–Based Fracture Prediction Model With Images of Vertebral Bones and Muscles by Employing Deep Learning: Development and Validation Study

Establish and validate the reliability of predictive models in bone mineral density by deep learning as examination tool for women