Artificial intelligence for the detection of sacroiliitis on magnetic resonance imaging in patients with axial spondyloarthritis

The role of deep learning in diagnostic imaging of spondyloarthropathies: a systematic review

AIM

Diagnostic imaging is an integral part of identifying spondyloarthropathies (SpA), yet the interpretation of these images can be challenging. This review evaluated the use of deep learning models to enhance the diagnostic accuracy of SpA imaging.

METHODS

Following PRISMA guidelines, we systematically searched major databases up to February 2024, focusing on studies that applied deep learning to SpA imaging. Performance metrics, model types, and diagnostic tasks were extracted and analyzed. Study quality was assessed using QUADAS-2.

RESULTS

We analyzed 21 studies employing deep learning in SpA imaging diagnosis across MRI, CT, and X-ray modalities. These models, particularly advanced CNNs and U-Nets, demonstrated high accuracy in diagnosing SpA, differentiating arthritis forms, and assessing disease progression. Performance metrics frequently surpassed traditional methods, with some models achieving AUCs up to 0.98 and matching expert radiologist performance.

CONCLUSION

This systematic review underscores the effectiveness of deep learning in SpA imaging diagnostics across MRI, CT, and X-ray modalities. The studies reviewed demonstrated high diagnostic accuracy. However, the presence of small sample sizes in some studies highlights the need for more extensive datasets and further prospective and external validation to enhance the generalizability of these AI models.

KEY POINTS

Question How can deep learning models improve diagnostic accuracy in imaging for spondyloarthropathies (SpA), addressing challenges in early detection and differentiation from other forms of arthritis? Findings Deep learning models, especially CNNs and U-Nets, showed high accuracy in SpA imaging across MRI, CT, and X-ray, often matching or surpassing expert radiologists. Clinical relevance Deep learning models can enhance diagnostic precision in SpA imaging, potentially reducing diagnostic delays and improving treatment decisions, but further validation on larger datasets is required for clinical integration.

Development and validation of machine learning algorithms for early detection of ankylosing spondylitis using magnetic resonance images

BackgroundAnkylosing spondylitis (AS) is a chronic inflammatory disease affecting the sacroiliac joints and spine, often leading to disability if not diagnosed and treated early.ObjectiveIn this study, we present the development and validation of machine learning (ML) algorithms for AS detection only using Short Tau Inversion Recovery (STIR) sequenced magnetic resonance (MR) images.MethodsThe detection process is based on creating Gray Level Co-occurrence Matrices (GLCM) from MR images, followed by the computation of Haralick features and the training of ML-based models. A total of 696 MR images (AS+: 348, AS-: 348) were utilized for AS detection. Models were trained and tested on 70% of the dataset using a 10-fold cross-validation method to prevent overfitting, while the remaining 30% of the data was used for validation. In addition, care was taken to ensure that different images from the same patient were not split between the training and validation sets during this separation process to prevent potential data leakage.ResultsThe proposed ML-based model demonstrated superior performance during the validation phase (accuracy: 0.885, AUC: 0.941). The results of our study show promising outcomes when compared to previous works employing GLCM-based ML detection models.Conclusions: This study introduces a new perspective on AS detection, focusing on the assignment of ML techniques to STIR-sequenced MR images with a notable absence of literature on interpreting ML models for AS detection. This typology also addresses a lack of knowledge, as most models do not provide practical interpretability or knowledge alongside accurate prediction. The system also offers an effective strategy for early and correct diagnosis of AS, which is important for timely intervention and treatment planning.

Deep learning analysis for rheumatologic imaging: current trends, future directions, and the role of human

Rheumatic diseases, such as rheumatoid arthritis (RA), osteoarthritis (OA), and spondyloarthritis (SpA), present diagnostic and management challenges due to their impact on connective tissues and the musculoskeletal system. Traditional imaging techniques, including plain radiography, ultrasounds, computed tomography, and magnetic resonance imaging (MRI), play a critical role in diagnosing and monitoring these conditions, but face limitations like inter-observer variability and time-consuming assessments. Recently, deep learning (DL), a subset of artificial intelligence, has emerged as a promising tool for enhancing medical imaging analysis. Convolutional neural networks, a DL model type, have shown great potential in medical image classification, segmentation, and anomaly detection, often surpassing human performance in tasks like tumor identification and disease severity grading. In rheumatology, DL models have been applied to plain radiography, ultrasounds, and MRI for assessing joint damage, synovial inflammation, and disease progression in RA, OA, and SpA patients. Despite the promise of DL, challenges such as data bias, limited explainability, and the need for large annotated datasets remain significant barriers to its widespread adoption. Furthermore, human oversight and value judgment are essential for ensuring the ethical use and effective implementation of DL in clinical settings. This review provides a comprehensive overview of DL's applications in rheumatologic imaging and explores its future potential in enhancing diagnosis, treatment decisions, and personalized medicine.

Clinical implications of teleradiology in rheumatic and musculoskeletal diseases: improving rheumatic care

Teleradiology, the transmission of radiologic images for remote assessment and consultation, has transformed modern medical care by mitigating geographical inequities and improving diagnostic accuracy. This technology employs telecommunications, digital imaging, and data-sharing systems developments to deliver swift and precise image analysis across various healthcare environments. Teleradiology has been essential in identifying and controlling diseases, including osteoarthritis, osteoporosis, rheumatoid arthritis, and spondyloarthritis, especially in musculoskeletal radiology and rheumatology. The combination of teleradiology and telemedicine has transformed multidisciplinary cooperation, enhancing communication among radiologists, rheumatologists, and other healthcare practitioners to provide patient-centered treatment. It has markedly enhanced access to highly specialized knowledge, especially in rural and disadvantaged areas, facilitating prompt consultations and alleviating patient travel constraints. However, despite its benefits, teleradiology encounters several challenges, including standardization issues, ethical dilemmas, and infrastructure constraints. The absence of uniform standards and inequalities in access to high-speed Internet and digital health records impede extensive implementation. Addressing these constraints is crucial to fully utilizing teleradiology's potential in musculoskeletal and rheumatic care. This article highlights the transformational potential of teleradiology and its incorporation into telemedicine for musculoskeletal and rheumatological treatment. Teleradiology is set to enhance global healthcare delivery by addressing disparities in healthcare access, fostering multidisciplinary cooperation, and utilizing advanced technologies. It underscores the necessity for ongoing innovation and investment in infrastructure, education, and standards to optimize the advantages of this crucial technology and guarantee equitable, efficient, and high-quality care for all patients.

Advancements in Imaging Techniques for Early Diagnosis and Management of Axial Spondyloarthritis

PURPOSE OF REVIEW

We aimed to introduce recent finding of imaging studies used in axial spondyloarthritis (axSpA).

RECENT FINDINGS

Using low-dose whole spine CT (CT syndesmophyte score [CTSS]) improved diagnostic accuracy for evaluating spinal structural progression than previous method (modified Stoke Ankylosing Spondylitis Spinal Score [mSASSS]) in axSpA. The novel definition of positive finding of sacroiliac joint (SIJ) and spine magnetic resonance imaging (MRI) enabled to diagnose axSpA earlier than plain radiography. In addition, novel MRI protocol such as volumetric interpolated breath-hold examination improved detection rate of structural change of axial joints in axSpA, Nuclear medicine imaging showed potential for diagnosis and predicting progression of axSpA. Ultrasonography guided injection is useful for controlling local joint pain of axSpA. AxSpA is characterised by pain and inflammation of axial joints such as the SIJ and spine. Detection of active inflammatory lesions using MRI has expanded the subtypes of axSpA to include non-radiographic axSpA (nr-axSpA). In addition, many other imaging techniques have improved, and can now detect structural and early inflammatory lesions of the axial joints. In addition, a method for quantitative measurement of syndesmophytes by CTSS has been developed; this method is more accurate and sensitive than the mSASSS for detecting spinal structural damage. Here, we discuss the current knowledge and clinical advances in computed tomography, MRI, nuclear medicine imaging, and ultrasonography as imaging methods for axSpA.

Unmet needs in axial spondyloarthritis. Proceedings of the French spondyloarthritis taskforce workshop

The progress observed over the last 30 years in the field of axial spondyloarthritis (axSpA) has not made it possible to answer all the current questions. This manuscript represents the proceedings of the meeting of the French spondyloArthitiS Task force (FAST) in Besançon on September 28 and 29, 2023. Different points of discussion were thus individualized as unmet needs: biomarkers for early diagnosis and disease activity, a common electronic file dedicated to SpA nationwide, a better comprehension of dysbiosis in the disease, a check-list for addressing to the rheumatologist, adapt patient reported outcomes thresholds for female gender, implementation of comorbidities screening programs, new imaging tools, in research cellular and multi omics approaches, grouping, at a nationwide level, different cohorts and registries, therapeutic strategy studies, consensual definition of difficult to treat disease and management, preclinical stage of the disease, mastering AI as a tool in the various aspects of research. These elements may represent a framework for the research agenda in axSpA for the years to come.

Regional Differences in Diagnosis Journey and Healthcare Utilization: Results from the International Map of Axial Spondyloarthritis (IMAS)

INTRODUCTION

To assess differences in the diagnosis journey and access to care in a large sample of patients with axial spondyloarthritis (axSpA) from around the world, included in the International Map of Axial Spondyloarthritis (IMAS).

METHODS

IMAS was a cross-sectional online survey (2017-2022) of 5557 unselected patients with axSpA from 27 countries. Across five worldwide geographic regions, the patient journey until diagnosis and healthcare utilization in the last 12 months prior to survey were evaluated. Univariable and multivariable linear regression was used to analyze factors associated with higher healthcare utilization.

RESULTS

Of 5557 participants in IMAS, the diagnosis took an average of 7.4 years, requiring more than two visits to HCPs (77.7% general practitioner and 51.3% rheumatologist), and more than two diagnostic tests [67.5% performed human leukocyte antigen B27 (HLA-B27), 64.2% x-ray, and 59.1% magnetic resonance imaging (MRI) scans]. North America and Europe were the regions with the highest number of healthcare professional (HCP) visits for diagnosis, while the lowest number of visits was in the Asian region. In the previous 12 months, 94.9% (n = 5272) used at least one healthcare resource, with an average of 29 uses per year. The regions with the highest healthcare utilization were Latin America, Europe, and North America. In the multiple linear regression, factors associated with higher number of healthcare utilization were younger age (b =  - 0.311), female gender (b = 7.736), higher disease activity (b = 1.461), poorer mental health (b = 0.624), greater functional limitation (b = 0.300), greater spinal stiffness (b = 1.527), and longer diagnostic delay (b = 0.104).

CONCLUSION

The diagnosis of axSpA usually takes more than two visits to HCPs and at least 7 years. After diagnosis, axSpA is associated with frequent healthcare resource use. Younger age, female gender, higher disease activity, poorer mental health, greater functional limitation, greater spinal stiffness, and longer diagnostic delay are associated with higher healthcare utilization. Europe and North America use more HCP visits and diagnostic tests before and after diagnosis than the other regions.

Artificial intelligence and machine learning in axial spondyloarthritis

PURPOSE OF REVIEW

To evaluate the current applications and prospects of artificial intelligence and machine learning in diagnosing and managing axial spondyloarthritis (axSpA), focusing on their role in medical imaging, predictive modelling, and patient monitoring.

RECENT FINDINGS

Artificial intelligence, particularly deep learning, is showing promise in diagnosing axSpA assisting with X-ray, computed tomography (CT) and MRI analyses, with some models matching or outperforming radiologists in detecting sacroiliitis and markers. Moreover, it is increasingly being used in predictive modelling of disease progression and personalized treatment, and could aid risk assessment, treatment response and clinical subtype identification. Variable study designs, sample sizes and the predominance of retrospective, single-centre studies still limit the generalizability of results.

SUMMARY

Artificial intelligence technologies have significant potential to advance the diagnosis and treatment of axSpA, providing more accurate, efficient and personalized healthcare solutions. However, their integration into clinical practice requires rigorous validation, ethical and legal considerations, and comprehensive training for healthcare professionals. Future advances in artificial intelligence could complement clinical expertise and improve patient care through improved diagnostic accuracy and tailored therapeutic strategies, but the challenge remains to ensure that these technologies are validated in prospective multicentre trials and ethically integrated into patient care.

An update in the pharmacological management of axial spondyloarthritis

INTRODUCTION

Significant progress has been made in the diagnosis and management of axial spondyloarthritis (AxSpA) over recent decades. A greater understanding of the immunopathogenesis of the disease has paved the way for the development of targeted treatments. Their efficacy has been demonstrated in randomized controlled trials, meta-analyses and one head-to-head study of biologic DMARDs. Treatment decisions in AxSpA are currently influenced by patient choice, co-morbidity, clinician familiarity and cost.

AREAS COVERED

We review the clinical trials that underpin the evidence base for treatments in AxSpA. We also cover the meta-analyses and head-to-head data that seek to support clinicians in personalizing treatment decisions. Further, we discuss the recent international guidelines that provide clinicians with treatment pathways and guidance.

EXPERT OPINION

We conclude that treatment decisions in managing both radiographic and non-radiographic AxSpA should be based on shared decision-making with patients, the clinical effectiveness of drug class, co-morbidity and cost. At present, we have limited head-to-head data to prioritize one drug class over another for first-line treatment but can recommend tumor necrosis factor (TNF), interleukin 17 (IL17) and JAK inhibition as being comparable in terms of clinical, structural and patient-reported outcome measures. Further real-world data may guide treatment decision-making in individual patients.

Advancing Rheumatology Care Through Machine Learning

Rheumatologic diseases are marked by their complexity, involving immune-, metabolic- and mechanically mediated processes which can affect different organ systems. Despite a growing arsenal of targeted medications, many rheumatology patients fail to achieve full remission. Assessing disease activity remains challenging, as patients prioritize different symptoms and disease phenotypes vary. This is also reflected in clinical trials where the efficacy of drugs is not necessarily measured in an optimal way with the traditional outcome assessment. The recent COVID-19 pandemic has catalyzed a digital transformation in healthcare, embracing telemonitoring and patient-reported data via apps and wearables. As a further driver of digital medicine, electronic medical record (EMR) providers are actively engaged in developing algorithms for clinical decision support, heralding a shift towards patient-centered, decentralized care. Machine learning algorithms have emerged as valuable tools for handling the increasing volume of patient data, promising to enhance treatment quality and patient well-being. Convolutional neural networks (CNN) are particularly promising for radiological image analysis, aiding in the detection of specific lesions such as erosions, sacroiliitis, or osteoarthritis, with several FDA-approved applications. Clinical predictions, including numerical disease activity forecasts and medication choices, offer the potential to optimize treatment strategies. Numeric predictions can be integrated into clinical workflows, allowing for shared decision making with patients. Clustering patients based on disease characteristics provides a personalized care approach. Digital biomarkers, such as patient-reported outcomes and wearables data, offer insights into disease progression and therapy response more flexibly and outside patient consultations. In association with patient-reported outcomes, disease-specific digital biomarkers via image recognition or single-camera motion capture enables more efficient remote patient monitoring. Digital biomarkers may also play a major role in clinical trials in the future as continuous, disease-specific outcome measurement facilitating decentralized studies. Prediction models can help with patient selection in clinical trials, such as by predicting high disease activity. Efforts are underway to integrate these advancements into clinical workflows using digital pathways and remote patient monitoring platforms. In summary, machine learning, digital biomarkers, and advanced imaging technologies hold immense promise for enhancing clinical decision support and clinical trials in rheumatology. Effective integration will require a multidisciplinary approach and continued validation through prospective studies.

Radiological Insights into Sacroiliitis: A Narrative Review

Sacroiliitis is the inflammation of the sacroiliac joint, the largest axial joint in the human body, contributing to 25% of lower back pain cases. It can be detected using various imaging techniques like radiography, MRI, and CT scans. Treatments range from conservative methods to invasive procedures. Recent advancements in artificial intelligence offer precise detection of this condition through imaging. Treatment options range from physical therapy and medications to invasive methods like joint injections and surgery. Future management looks promising with advanced imaging, regenerative medicine, and biologic therapies, especially for conditions like ankylosing spondylitis. We conducted a review on sacroiliitis using imaging data from sources like PubMed and Scopus. Only English studies focusing on sacroiliitis's radiological aspects were included. The findings were organized and presented narratively.