Comprehensive AI-assisted tool for ankylosing spondylitis based on multicenter research outperforms human experts

Introduction The diagnosis and treatment of ankylosing spondylitis (AS) is a difficult task, especially in less developed countries without access to experts. To address this issue, a comprehensive artificial intelligence (AI) tool was created to help diagnose and predict the course of AS. Methods In this retrospective study, a dataset of 5389 pelvic radiographs (PXRs) from patients treated at a single medical center between March 2014 and April 2022 was used to create an ensemble deep learning (DL) model for diagnosing AS. The model was then tested on an additional 583 images from three other medical centers, and its performance was evaluated using the area under the receiver operating characteristic curve analysis, accuracy, precision, recall, and F1 scores. Furthermore, clinical prediction models for identifying high-risk patients and triaging patients were developed and validated using clinical data from 356 patients. Results The ensemble DL model demonstrated impressive performance in a multicenter external test set, with precision, recall, and area under the receiver operating characteristic curve values of 0.90, 0.89, and 0.96, respectively. This performance surpassed that of human experts, and the model also significantly improved the experts' diagnostic accuracy. Furthermore, the model's diagnosis results based on smartphone-captured images were comparable to those of human experts. Additionally, a clinical prediction model was established that accurately categorizes patients with AS into high-and low-risk groups with distinct clinical trajectories. This provides a strong foundation for individualized care. Discussion In this study, an exceptionally comprehensive AI tool was developed for the diagnosis and management of AS in complex clinical scenarios, especially in underdeveloped or rural areas that lack access to experts. This tool is highly beneficial in providing an efficient and effective system of diagnosis and management.

[1]  J.-S. Pan,et al.  Deep Learning-Based Universal Expert-Level Recognizing Pathological Images of Hepatocellular Carcinoma and beyond , 2020, medRxiv.

[2]  OUP accepted manuscript , 2022, Rheumatology.

[3]  G. Collins,et al.  Risk of bias in studies on prediction models developed using supervised machine learning techniques: systematic review , 2021, BMJ.

[4]  B. Qian,et al.  Failure patterns and related risk factors of sagittal reconstruction following pedicle subtraction osteotomy in patients with ankylosing spondylitis and thoracolumbar kyphosis. , 2021, Neurosurgical focus.

[5]  Zenglin Xu,et al.  Machine learning based early mortality prediction in the emergency department , 2021, Int. J. Medical Informatics.

[6]  Z. Siddique,et al.  Detection of COVID-19 Patients from CT Scan and Chest X-ray Data Using Modified MobileNetV2 and LIME , 2021, Healthcare.

[7]  Yixin Zhou,et al.  Assessment of hip involvement in patients with ankylosing spondylitis: reliability and validity of the Hip Inflammation MRI Scoring System , 2021, BMC Musculoskeletal Disorders.

[8]  G. Lee,et al.  Method for Diagnosing the Bone Marrow Edema of Sacroiliac Joint in Patients with Axial Spondyloarthritis Using Magnetic Resonance Image Analysis Based on Deep Learning , 2021, Diagnostics.

[9]  Wei Liu,et al.  Analysis of Bone Strength and Bone Turnover Markers in Ankylosing Spondylitis with Radiological Hip Involvement , 2021, Medical science monitor : international medical journal of experimental and clinical research.

[10]  M. Brown,et al.  Ankylosing spondylitis: an autoimmune or autoinflammatory disease? , 2021, Nature Reviews Rheumatology.

[11]  Rittika Shamsuddin,et al.  CEFEs: A CNN Explainable Framework for ECG Signals , 2021, Artif. Intell. Medicine.

[12]  D. M. van der Heijde,et al.  Tofacitinib for the treatment of ankylosing spondylitis: a phase III, randomised, double-blind, placebo-controlled study , 2021, Annals of the Rheumatic Diseases.

[13]  E. Trucco,et al.  Using machine learning approaches for multi-omics data analysis: A review. , 2021, Biotechnology advances.

[14]  J. Gu,et al.  Serum Metabolomics Signatures Associated With Ankylosing Spondylitis and TNF Inhibitor Therapy , 2021, Frontiers in Immunology.

[15]  Le Lu,et al.  A scalable physician-level deep learning algorithm detects universal trauma on pelvic radiographs , 2021, Nature Communications.

[16]  D. Berry,et al.  Primary Total Hip Arthroplasty in Patients With Ankylosing Spondylitis. , 2021, The Journal of arthroplasty.

[17]  W. Zhou,et al.  An Ensembled Deep Learning Model Outperforms Human Experts in Diagnosing Biliary Atresia from Sonographic Gallbladder Images , 2020, medRxiv.

[18]  Sijia Huang,et al.  DeepProg: an ensemble of deep-learning and machine-learning models for prognosis prediction using multi-omics data , 2021, Genome Medicine.

[19]  Zewei Chen,et al.  Self-Optimizing Support Vector Elastic Net. , 2020, Analytical chemistry.

[20]  Zhifa Zhang,et al.  Improvement of Sleep Quality in Patients with Ankylosing Spondylitis Kyphosis after Corrective Surgery. , 2020, Spine.

[21]  M. Garrido-Cumbrera,et al.  Social economic costs of ankylosing spondylitis in Spain. , 2020, Clinical and experimental rheumatology.

[22]  L. See.  The Patient's Perspective on the Burden of Disease in Ankylosing Spondylitis. , 2020, Rheumatic diseases clinics of North America.

[23]  M. Dougados,et al.  Dual neutralisation of interleukin-17A and interleukin-17F with bimekizumab in patients with active ankylosing spondylitis: results from a 48-week phase IIb, randomised, double-blind, placebo-controlled, dose-ranging study , 2020, Annals of the Rheumatic Diseases.

[24]  T. Webster,et al.  Advances in nanomedicine for the treatment of ankylosing spondylitis , 2019, International journal of nanomedicine.

[25]  Qiang Huang,et al.  Two-Stage Treatment for Ankylosing Spondylitis With Severe Hip Contracture. , 2019, Orthopedics.

[26]  D. Lin,et al.  Bilateral total hip arthroplasty in ankylosing spondylitis: a systematic review , 2019, EFORT open reviews.

[27]  William Graf,et al.  Deep learning for cellular image analysis , 2019, Nature Methods.

[28]  R. Barzilay,et al.  A Deep Learning Mammography-based Model for Improved Breast Cancer Risk Prediction. , 2019, Radiology.

[29]  T. Helbich,et al.  Impact of Machine Learning With Multiparametric Magnetic Resonance Imaging of the Breast for Early Prediction of Response to Neoadjuvant Chemotherapy and Survival Outcomes in Breast Cancer Patients , 2019, Investigative radiology.

[30]  Srini Tridandapani,et al.  Deep Learning in Radiology. , 2018, Academic radiology.

[31]  A. Chhabra,et al.  Ankylosing Spondylitis and Axial Spondyloarthritis. , 2016, The New England journal of medicine.

[32]  E. Mohammadi,et al.  Barriers and facilitators related to the implementation of a physiological track and trigger system: A systematic review of the qualitative evidence , 2017, International journal for quality in health care : journal of the International Society for Quality in Health Care.

[33]  Robert C. Wolpert,et al.  A Review of the , 1985 .