No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Machine learning classification of vitamin D levels in spondyloarthritis patients
... This integration enabled us to conduct a comprehensive feature selection using various methods, including mutual information. While the authors previously applied the mutual information technique in studies, 4,5 we have now enhanced its methodological rigour by incorporating bootstrapping and cross-validation. Additionally, we used other feature selection models, such as random forest and logistic regression, and created a ranking to assess the importance of each variable. ...
Background
The effectiveness of anti-tumour necrosis factor (TNF) therapy in spondyloarthritis is traditionally associated with factors such as age, obesity and disease subtypes. However, less-explored aspects, such as mental health, socioeconomic status and work type may also play a crucial role in determining inflammatory activity and therapeutic response.
Objectives
To identify the most significant factors explaining inflammatory activity levels in patients treated with anti-TNF therapy and to develop an interpretable machine-learning model with good performance and minimal overfitting.
Design
This is an observational, cross-sectional and multicentre study with socio-demographical and clinical data extracted from the Registry of Spondyloarthritis of Spanish Rheumatology (REGISPONSER) and Ibero-American Registry of Spondyloarthropathies (RESPONDIA) registries.
Methods
We selected patients receiving anti-TNF therapy and applied five feature selection methods to identify key factors. We evaluated these factors using 182 machine learning models, and, finally, we selected a decision tree model that offered comparable performance with reduced overfitting.
Results
Activity levels appear strongly influenced by quality-of-life indicators, particularly the SF-12 physical and mental components and Ankylosing Spondylitis Quality of Life scores. While factors such as age, weight, years of treatment and age at diagnosis have relevance, they are not necessary to obtain a pruned tree with similar cross-validated mean accuracy.
Conclusion
Recognizing the central role of physical and mental well-being in managing disease activity can lead to better therapeutic strategies for chronic disease management.
Axial spondyloarthritis (axSpA) is a chronic inflammatory disease which predominantly involves the axial skeleton and sacroiliac joints. The aetiology remains unknown but is thought to be immune driven. Vitamin D is a focus for research in numerous autoimmune conditions, especially because it is now thought to have an important role in immunoregulation. It has been hypothesised that low levels of vitamin D increase the risk of autoimmune disease. Considering that vitamin D is implicated in immune regulation and autoimmunity, a question that arises is whether vitamin D deficiency can lead to increased disease activity in axSpA. Through this narrative review of the literature the authors explore potential links between vitamin D and axSpA. This review highlights that larger and more methodologically robust prospective longitudinal studies are required to answer this key question. There was considerable heterogeneity between studies, including in the definition of vitamin D deficiency, latitude where the study took place, and seasonal variation. Another clinically relevant aspect to address is whether correcting vitamin D deficiency leads to improved makers of disease activity in patients with ankylosing spondylitis. This may justify mandatory food fortification and specific supplementation programmes in countries at risk. For example, in Finland there is a low prevalence of vitamin D deficiency in the general population because of food fortification.
Novel machine learning methods open the door to advances in rheumatology through application to complex, high-dimensional data, otherwise difficult to analyse. Results from such efforts could provide better classification of disease, decision support for therapy selection, and automated interpretation of clinical images. Nevertheless, such data-driven approaches could potentially model noise, or miss true clinical phenomena. One proposed solution to ensure clinically meaningful machine learning models is to involve primary stakeholders in their development and interpretation. Including patient and health care professionals’ input and priorities, in combination with statistical fit measures, allows for any resulting models to be well fit, meaningful, and fit for practice in the wider rheumatological community. Here we describe outputs from workshops that involved healthcare professionals, and young people from the Your Rheum Young Person’s Advisory Group, in the development of complex machine learning models. These were developed to better describe trajectory of early juvenile idiopathic arthritis disease, as part of the CLUSTER consortium. We further provide key instructions for reproducibility of this process.Involving people living with, and managing, a disease investigated using machine learning techniques, is feasible, impactful and empowering for all those involved.
Over the past decade, there has been a paradigm shift in how clinical data are collected, processed and utilized. Machine learning and artificial intelligence, fueled by breakthroughs in high-performance computing, data availability and algorithmic innovations, are paving the way to effective analyses of large, multi-dimensional collections of patient histories, laboratory results, treatments, and outcomes. In the new era of machine learning and predictive analytics, the impact on clinical decision-making in all clinical areas, including rheumatology, will be unprecedented. Here we provide a critical review of the machine-learning methods currently used in the analysis of clinical data, the advantages and limitations of these methods, and how they can be leveraged within the field of rheumatology.
Objective:
Vitamin D plays an important role in mineral turnover and bone remodeling and there are increasing data about its immunomodulatory potential in different rheumatologic disorders. Deficiency of vitamin D is frequent in patients with spondyloarthritis (SpA) and some data suggest its association with increased disease activity and structural damage. However, its exact role in the pathogenesis of SpA and its association with disease activity are still a matter of debate.
Material and methods:
A cross-sectional study of patients diagnosed with axial spondyloarthritis (axSpA) and peripheral spondyloarthritis (perSpA) according to Assessment of Spondyloarthritis International Society classification criteria was performed. The correlation between concentration of 25-hydroxyvitamin D - 25(OH)D - and disease activity scores (Bath Ankylosing Spondylitis Disease Activity Index - BASDAI, Ankylosing Spondylitis Disease Activity Score - ASDAS), inflammatory markers (C-reactive protein - CRP, erythrocyte sedimentation rate - ESR) and clinical symptoms (arthritis, enthesitis, dactylitis) was performed.
Results:
We included 40 patients with axSpA and 23 patients with perSpA. The mean concentration of 25(OH)D was 24.9 ng/ml (SD 12.49). Forty-seven (74.6%) patients had 25(OH)D below the recommended threshold (< 30 ng/ml). We found no statistically significant negative correlation between the level of 25(OH)D and disease activity of axSpA and perSpA in terms of clinical symptoms (arthritis, enthesitis, dactylitis), inflammatory markers (ESR, CRP) and disease activity scores (BASDAI, ASDAS). These results did not change after adjustment for supplementation of vitamin D and seasonal variation.
Conclusions:
Our data show no correlation between the concentration of 25(OH)D in the serum and disease activity in two subgroups of SpA. However, this does not exclude the potential role of vitamin D in pathogenesis of SpA. Further studies are required to evaluate the optimal range of 25(OH)D serum concentration in axSpA and perSpA patients with its possible immunomodulatory potential and influence on disease activity.
Background
Spondyloarthritides (SpAs) encompass heterogeneous diseases sharing similar genetic background, pathogenic mechanisms, and phenotypic features. Vitamin D is essential for calcium metabolism and skeletal homeostasis. Some recent evidences reported supplemental functions of vitamin D, such as modulation of inflammatory reactions.
Objective
To analyze published data about a possible association between vitamin D and SpAs.
Results
Vitamin D could play a role in immune reactions, influencing both immune and adaptive response. Vitamin D deficiency is more frequent in SpAs than in general population: an active and more severe disease infers patients’ mobility and reduces sunlight exposure. Quiescent inflammatory bowel disease, frequently associated with SpAs, could worsen vitamin D deficiency. All the parameters related to UVB exposure are the most important determinants for vitamin D status and need to be considered evaluating the vitamin D levels in SpAs.
Apart from musculoskeletal problems, patients affected by SpAs frequently suffer from other comorbidities, especially cardiovascular diseases and osteoporosis, and vitamin D status could have a relevance in this field. Bone is involved in SpAs with a dualistic role, coexisting trabecular bone resorption and new bone formation.
It seems rational to monitor vitamin D levels in SpA subjects and to target it to global health threshold.
Conclusion
Literature data were not completely in agreement about a possible relation between poor vitamin D status and onset or worse disease course of SpAs. In fact, these results come from cross-sectional studies, which affect our ability to infer causality. Therefore, large, randomized controlled trials are needed.
Objective
Ankylosing Spondylitis (AS) is an inflammatory rheumatic disease that mainly affects the axial spine. Osteopenia and osteoporosis are the main complications of AS. Vitamin D has functions on the immune system. In this study, we aimed to compare vitamin D levels and Bone Mineral Density (BMD) values between AS patients and controls.
Methods
A total of 68 patients with axial AS and 34 healthy controls were enrolled in this study conducted between March 2018 and May 2018. Vitamin D concentrations, BMD values, disease activity, back mobility, functionality and radiologic damage were evaluated.
Results
Vitamin D concentrations, the total BMD-femur and BMD-femur neck values were significantly lower in AS patients (p = 0.001, p = 0.011 and p = 0.003). No significant correlations were detected between vitamin D levels and BMD-femur total, BMD-femur neck values, disease activity, back mobility, functionality and radiologic damage scores (p > 0.05). Disease activity parameters were significantly and negatively correlated with total BMD-femur and BMD-femur neck values (p < 0.05).
Conclusion
Our study demonstrates that AS patients have lower vitamin D levels, total BMD-femur and BMD-femur neck values. Higher disease activity increases bone loss in AS. Regular measurement of BMD and vitamin D should be kept in mind when planning a treatment in AS.
The Chi-square statistic is a non-parametric (distribution free) tool designed to analyze group differences when the dependent variable is measured at a nominal level. Like all non-parametric statistics, the Chi-square is robust with respect to the distribution of the data. Specifically, it does not require equality of variances among the study groups or homoscedasticity in the data. It permits evaluation of both dichotomous independent variables, and of multiple group studies. Unlike many other non-parametric and some parametric statistics, the calculations needed to compute the Chi-square provide considerable information about how each of the groups performed in the study. This richness of detail allows the researcher to understand the results and thus to derive more detailed information from this statistic than from many others.
The Chi-square is a significance statistic, and should be followed with a strength statistic. The Cramer’s V is the most common strength test used to test the data when a significant Chi-square result has been obtained. Advantages of the Chi-square include its robustness with respect to distribution of the data, its ease of computation, the detailed information that can be derived from the test, its use in studies for which parametric assumptions cannot be met, and its flexibility in handling data from both two group and multiple group studies. Limitations include its sample size requirements, difficulty of interpretation when there are large numbers of categories (20 or more) in the independent or dependent variables, and tendency of the Cramer’s V to produce relative low correlation measures, even for highly significant results.
Objective:
The aim of the present study was to evaluate the plasma vitamin D (vit D) levels and their association with the disease activity in patients with ankylosing spondylitis (AS) and undifferentiated spondyloarthritis (uSpA) compared with healthy populations.
Methods:
This study included 161 spondyloarthritis patients (113 uSpA patients and 48 AS patients) attending our rheumatology out-patient clinic, along with 92 controls.
Results:
The plasma vit D levels were 18 μg/L (8-38) in the AS group, 20 μg/L (4-92.3) in the uSpA group and 24.3 μg/L (7.2-76.8) in the control group. The plasma vit D levels of the AS patients were significantly lower than those of the patients in the control group (p=0.004). The men in the AS group had significanly lower vit D levels than those in the control group (p=0.005). On the other hand, the women in the uSpA group had significanly lower vit D levels than those in the control group (p=0.011). The vit D levels were inversely related to both erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) in the AS patients (p=0.002, R=-0.428; p<0.001, R=-0.592, respectively). This correlation was not demonstrated in the uSpA patients. The vit D levels were not found to correlate with the Bath ankylosing spondylitis disease activity index (BASDAI) levels in either the AS or uSpA patients.
Conclusion:
25-hydroxy-vit D deficiency is frequently observed in patients with SpAs. In this study, vit D deficiency was much more prominent in the male AS patients. On the other hand, among women, the uSpA patients exhibited much more prominent vit D deficiency than the control group subjects. The acute phase response may inversely affect the vit D levels in AS patients.
Scikit-learn is a Python module integrating a wide range of state-of-the-art
machine learning algorithms for medium-scale supervised and unsupervised
problems. This package focuses on bringing machine learning to non-specialists
using a general-purpose high-level language. Emphasis is put on ease of use,
performance, documentation, and API consistency. It has minimal dependencies
and is distributed under the simplified BSD license, encouraging its use in
both academic and commercial settings. Source code, binaries, and documentation
can be downloaded from http://scikit-learn.sourceforge.net.
The field of spondyloarthritis (SpA) has experienced major progress in the last decade, especially with regard to new treatments, earlier diagnosis, imaging technology and a better definition of outcome parameters for clinical trials. In the present work, the Assessment in SpondyloArthritis international Society (ASAS) provides a comprehensive handbook on the most relevant aspects for the assessments of spondyloarthritis, covering classification criteria, MRI and x rays for sacroiliac joints and the spine, a complete set of all measurements relevant for clinical trials and international recommendations for the management of SpA. The handbook focuses at this time on axial SpA, with ankylosing spondylitis (AS) being the prototype disease, for which recent progress has been faster than in peripheral SpA. The target audience includes rheumatologists, trial methodologists and any doctor and/or medical student interested in SpA. The focus of this handbook is on practicality, with many examples of MRI and x ray images, which will help to standardise not only patient care but also the design of clinical studies.
We present two classes of improved estimators for mutual information M(X,Y), from samples of random points distributed according to some joint probability density mu(x,y). In contrast to conventional estimators based on binnings, they are based on entropy estimates from k -nearest neighbor distances. This means that they are data efficient (with k=1 we resolve structures down to the smallest possible scales), adaptive (the resolution is higher where data are more numerous), and have minimal bias. Indeed, the bias of the underlying entropy estimates is mainly due to nonuniformity of the density at the smallest resolved scale, giving typically systematic errors which scale as functions of k/N for N points. Numerically, we find that both families become exact for independent distributions, i.e. the estimator M(X,Y) vanishes (up to statistical fluctuations) if mu(x,y)=mu(x)mu(y). This holds for all tested marginal distributions and for all dimensions of x and y. In addition, we give estimators for redundancies between more than two random variables. We compare our algorithms in detail with existing algorithms. Finally, we demonstrate the usefulness of our estimators for assessing the actual independence of components obtained from independent component analysis (ICA), for improving ICA, and for estimating the reliability of blind source separation.
Resumen
Introducción
Las espondiloartritis son un grupo de enfermedades inflamatorias crónicas con afectación principalmente del esqueleto axial y también de articulaciones periféricas. En cuanto al metabolismo óseo de estos pacientes, se ha observado en algunos estudios que existen niveles más bajos de vitamina D en pacientes con espondiloartritis.
Objetivo
Estimar la prevalencia de déficit/insuficiencia de vitamina D, el metabolismo fosfocálcico y sus implicaciones en una cohorte de pacientes con espondiloartritis.
Metodología
Estudio observacional, descriptivo y transversal. Se llevó a cabo una revisión retrospectiva de la base de datos de pacientes con espondiloartritis que fueron atendidos en las consultas externas del Servicio de Reumatología del Hospital General Universitario de Ciudad Real entre junio del 2018 y junio del 2019. Las variables se describieron usando medidas de frecuencia o medidas de tendencia central/dispersión según correspondiera.
Resultados
Se analizaron 115 pacientes, de los cuales 64 fueron hombres y 51 mujeres, con una edad media de 45,97 años (± 13,41 DE). Del total de los pacientes, 59 presentaron espondilitis anquilosante, 24 artropatía psoriásica, 9 artritis asociada a enfermedad inflamatoria intestinal, 12 espondiloartritis axial no radiográfica y 11 artritis reactiva. Los niveles de vitamina D fueron de 23,81 ng/ml (± 10,5 DE), con un 77,4% de los pacientes con cifras de déficit/insuficiencia de vitamina D. Agrupados por el subtipo de espondiloartritis y según las cifras de déficit/insuficiencia de vitamina D, 45 pacientes tenían espondilitis anquilosante, 19 artropatía psoriásica, 9 artritis asociada a enfermedad inflamatoria intestinal, 7 espondiloartritis axiales no radiográficas y 9 artritis reactivas. Además, el déficit de vitamina D (< 20 ng/ml) se presentaba la mayoría de las veces en las estaciones de primavera e invierno, con 31 y 26 pacientes respectivamente.
Conclusiones
Una optimización de los niveles de vitamina D puede implicar una mejoría en la situación clínica del paciente, medida tanto por BASDAI y DAPSA como por PCR y VSG. En consecuencia, se recomienda la monitorización y suplementación de vitamina D en pacientes con hipovitaminosis D.
In this study, a relationship between low levels of vitamin D and other markers of inflammation is observed in patients with spondyloarthritis. Therefore, insufficient levels of vitamin D could be used as a marker of inflammation, so they would help to have a better follow-up of the patient.PurposeThe aim of this study is to determine the association between 25-hydroxyvitamin D deficiency and level of disease activity in patients suffering from spondyloarthritis.MethodologyThis study is an observational, descriptive and transversal study. A retrospective review of patients with spondyloarthritis under treatment at the Rheumatology Department of Hospital General Universitario de Ciudad Real between September 2016 and September 2018 is under consideration. The association between 25-hydroxyvitamin D deficiency and levels of disease activity was calculated using odds ratio (OR) with a 95% confidence level.ResultsInitial study developments are reported. A group of 147 patients was analyzed. The average disease activity was 4.3 (according to BASDAI) and 11.4 (according to DAPSA). Twenty-eight percent of patients showed increased levels of acute-phase reactants, and 82% exhibited 25-hydroxyvitamin D deficiency/insufficiency. Analysis of the association between vitamin D deficiency/insufficiency and BASDAI/DAPSA activity degree revealed an OR of 7.9 (CI95%: 2.92–21.23, p = < 0.0001).Conclusion
In this study, an association was found between those patients with spondyloarthritis who had a 25-hydroxyvitamin D deficiency and disease activity. Despite these results, a 1-year follow-up of these patients using two or more measurements of activity and 25-hydroxyvitamin D levels would be needed to corroborate these data.
Introduction:
This study aimed to compare serum vitamin D levels in Spondyloarthritis (SpA) patients and control group and to evaluate the associations between vitamin D and disease activity in SpA patients.
Methodology:
In this study, 86 SpA patients according to the International Criteria and 117 age and sex-matched healthy controls were included. In patients, clinical examination was performed and Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) and Bath Ankylosing Spondylitis Functional Index (BASFI) were recorded. Serum 25(OH)D3 concentration was measured using ELISA kit.
Results:
Serum 25(OH)D3 levels in SpA patients were significantly lower than healthy controls (p < 0.001). Vitamin D deficiency and insufficiency frequency in the SpA group was significantly more than control group (p < 0.001). No significant difference was observed in the activity of SpA in different serum levels of 25(OH)D. Also, no significant correlations were observed between serum 25(OH)D3 with clinical findings as well as with BASDAI and BASFI (P > 0.05).
Discussion and conclusion:
Although our study revealed lower serum 25(OH)D3 levels in SpA patients compared to healthy controls, there were not any significant correlations between its serum levels with severity of disease. However, correction of vitamin D status may be beneficial in controlling inflammation and disease activity.
This study aimed to evaluate the relationship between the 25-hydroxyvitamin D [25(OH)D] level and rheumatoid arthritis (RA) and the correlation between serum vitamin D level and RA activity. Methods We searched the PUBMED, EMBASE, and Cochrane databases and performed a meta-analysis examining the vitamin D level and prevalence of vitamin D deficiency in patients with RA compared to healthy controls and the correlation coefficients between the vitamin D level and disease activity score 28 (DAS28) in RA patients. Results Fifteen studies that included a total of 1,143 RA patients and 963 controls were available for this meta-analysis. The meta-analysis showed that the serum vitamin D level in the RA group was significantly lower than that in the control group (SMD=-0.608, 95% CI=-1.105-[-0.017], p=0.017). In addition, the prevalence of vitamin D deficiency was significantly higher in the RA group than in the control group (55.2% vs. 33.2%; OR = 2.460, 95% CI = 1.135-5.332, p=0.023). Thirteen studies evaluated the correlation between the vitamin D level and its activity in 924 RA patients. Meta-analysis showed a significant inverse correlation between the vitamin D level and DAS28 (Correlation coefficient =-0.278, 95% CI =-0.393-[-0.153], p=1.8 x 10⁻⁵). Conclusion Our meta-analysis demonstrates that serum vitamin D level is significantly low in patients with RA, vitamin D deficiency is prevalent in RA patients compared to controls, and the vitamin D level correlates inversely with RA activity. Our meta-analysis suggests that the vitamin D level is associated with susceptibility to RA and RA activity.
Bayesian optimization techniques have been successfully applied to robotics,
planning, sensor placement, recommendation, advertising, intelligent user
interfaces and automatic algorithm configuration. Despite these successes, the
approach is restricted to problems of moderate dimension, and several workshops
on Bayesian optimization have identified its scaling to high-dimensions as one
of the holy grails of the field. In this paper, we introduce a novel random
embedding idea to attack this problem. The resulting Random EMbedding Bayesian
Optimization (REMBO) algorithm is very simple, has important invariance
properties, and applies to domains with both categorical and continuous
variables. We present a thorough theoretical analysis of REMBO, including
regret bounds that only depend on the problem's intrinsic dimensionality.
Empirical results confirm that REMBO can effectively solve problems with
billions of dimensions, provided the intrinsic dimensionality is low. They also
show that REMBO achieves state-of-the-art performance in optimizing the 47
discrete parameters of a popular mixed integer linear programming solver.
The concepts for classification, diagnosis and referral of patients with axial spondyloarthitis differ, although they of course basically relate to the same disease. While classification criteria and referral strategies concentrate largely on patients with chronic back pain with an age at onset before 45 years, the rheumatologist can make a diagnosis of axial SpA in patients with late onset or in patients with back pain for only some weeks if other items are fulfilled. Early recognition of patients with axial SpA is important to establish the diagnosis, potentially start therapeutic interventions and avoid unnecessary health care procedures.
We study the numerical performance of a limited memory quasi-Newton method for large scale optimization, which we call the L-BFGS method. We compare its performance with that of the method developed by Buckley and LeNir (1985), which combines cycles of BFGS steps and conjugate direction steps. Our numerical tests indicate that the L-BFGS method is faster than the method of Buckley and LeNir, and is better able to use additional storage to accelerate convergence. We show that the L-BFGS method can be greatly accelerated by means of a simple scaling. We then compare the L-BFGS method with the partitioned quasi-Newton method of Griewank and Toint (1982a). The results show that, for some problems, the partitioned quasi-Newton method is clearly superior to the L-BFGS method. However we find that for other problems the L-BFGS method is very competitive due to its low iteration cost. We also study the convergence properties of the L-BFGS method, and prove global convergence on uniformly convex problems.
In this study, the relation between osteoporosis and vitamin D and the disease activity in patients with ankylosing spondylitis (AS) was investigated. A hundred patients with AS and 58 healthy individuals were included in the study. In addition to the routine blood and urine tests, serum 25-(OH)D3, parathormone (PTH), C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), total calcium, ionized calcium, and phosphorous levels of all participants were also measured. Bone mineral density (BMD) measurements were performed at the anterior-posterior and lateral lumbar and femur regions. Anterior-posterior and lateral thoracic and lumbosacral radiography was performed on all participants. The disease activity was evaluated by Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), functional status by Bath Ankylosing Spondylitis Functional Index (BASFI), and mobility by Bath Ankylosing Spondylitis Metrology Index (BASMI). In the patient group, BMD values obtained from the lateral lumbar and femur regions and serum vitamin D levels were lower than the control group. A negative relation was determined between the lateral lumbar BMD values and ESR, CRP, and BASDAI scores of patients with AS. The ESR, CRP levels, and BASMI scores of the AS patients with osteoporosis were significantly higher, when compared to patients without osteoporosis. The negative correlation between serum 25-(OH)D3 level and ESR, CRP levels did not reach a statistically significant level in patients with AS; the positive correlation between PTH levels and ESR, and the negative correlation between CRP and BASDAI also did not reach a statistically significant level. Vitamin D deficiency in AS may indirectly lead to osteoporosis by causing an increase in the inflammatory activity. The present authors believe that it would be beneficial to monitorize vitamin D levels together with BMD measurements in order to determine the patients under osteoporosis risk.
Abrudescu A Vitamin D levels in ankylosing spondylitis: does deficiency correspond to disease activity?
- Pokhai
Vitamin D status in spondyloarthritis: results of the ASAS-COMOSPA international study
- Fernandez
Old and new treatment targets in axial spondyloarthritis
- Rios