Machine learning identifies right index finger tenderness as key signal of DAS28-CRP based psoriatic arthritis activity.
Journal
Scientific reports
ISSN: 2045-2322
Titre abrégé: Sci Rep
Pays: England
ID NLM: 101563288
Informations de publication
Date de publication:
19 Dec 2023
19 Dec 2023
Historique:
received:
18
06
2023
accepted:
09
12
2023
medline:
21
12
2023
pubmed:
21
12
2023
entrez:
20
12
2023
Statut:
epublish
Résumé
Psoriatic arthritis (PsA) is a chronic inflammatory systemic disease whose activity is often assessed using the Disease Activity Score 28 (DAS28-CRP). The present study was designed to investigate the significance of individual components within the score for PsA activity. A cohort of 80 PsA patients (44 women and 36 men, aged 56.3 ± 12 years) with a range of disease activity from remission to moderate was analyzed using unsupervised and supervised methods applied to the DAS28-CRP components. Machine learning-based permutation importance identified tenderness in the metacarpophalangeal joint of the right index finger as the most informative item of the DAS28-CRP for PsA activity staging. This symptom alone allowed a machine learned (random forests) classifier to identify PsA remission with 67% balanced accuracy in new cases. Projection of the DAS28-CRP data onto an emergent self-organizing map of artificial neurons identified outliers, which following augmentation of group sizes by emergent self-organizing maps based generative artificial intelligence (AI) could be defined as subgroups particularly characterized by either tenderness or swelling of specific joints. AI-assisted re-evaluation of the DAS28-CRP for PsA has narrowed the score items to a most relevant symptom, and generative AI has been useful for identifying and characterizing small subgroups of patients whose symptom patterns differ from the majority. These findings represent an important step toward precision medicine that can address outliers.
Identifiants
pubmed: 38123604
doi: 10.1038/s41598-023-49574-4
pii: 10.1038/s41598-023-49574-4
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
22710Subventions
Organisme : Deutsche Forschungsgemeinschaft
ID : SFB 1039/Z01
Organisme : Deutsche Forschungsgemeinschaft
ID : DFG LO 612/16-1
Organisme : Fraunhofer-Gesellschaft
ID : Fraunhofer Cluster of Excellence for Immune Mediated diseases CIMD
Organisme : Fraunhofer-Gesellschaft
ID : Fraunhofer Cluster of Excellence for Immune Mediated diseases CIMD
Organisme : Innovative Medicines Initiative 2 Joint Undertaking (JU)
ID : 101007757
Organisme : Innovative Medicines Initiative 2 Joint Undertaking (JU)
ID : 101007757
Informations de copyright
© 2023. The Author(s).
Références
Zabotti, A. et al. Predictors, risk factors, and incidence rates of psoriatic arthritis development in psoriasis patients: A systematic literature review and meta-analysis. Rheumatol. Ther. 8, 1519–1534. https://doi.org/10.1007/s40744-021-00378-w (2021).
doi: 10.1007/s40744-021-00378-w
pubmed: 34596875
pmcid: 8572278
Pennington, S. R. & FitzGerald, O. Early origins of psoriatic arthritis: Clinical, genetic and molecular biomarkers of progression from psoriasis to psoriatic arthritis. Front. Med. 8, 72394. https://doi.org/10.3389/fmed.2021.723944 (2021).
doi: 10.3389/fmed.2021.723944
Singh, J. A. et al. 2015 American college of rheumatology guideline for the treatment of rheumatoid arthritis. Arthritis Care Res. (Hoboken) 68, 1–25. https://doi.org/10.1002/acr.22783 (2016).
doi: 10.1002/acr.22783
pubmed: 26545825
Mease, P. J. Measures of psoriatic arthritis: Tender and Swollen Joint Assessment, Psoriasis Area and Severity Index (PASI), Nail Psoriasis Severity Index (NAPSI), Modified Nail Psoriasis Severity Index (mNAPSI), Mander/Newcastle Enthesitis Index (MEI), Leeds Enthesitis Index (LEI), Spondyloarthritis Research Consortium of Canada (SPARCC), Maastricht Ankylosing Spondylitis Enthesis Score (MASES), Leeds Dactylitis Index (LDI), Patient Global for Psoriatic Arthritis, Dermatology Life Quality Index (DLQI), Psoriatic Arthritis Quality of Life (PsAQOL), Functional Assessment of Chronic Illness Therapy-Fatigue (FACIT-F), Psoriatic Arthritis Response Criteria (PsARC), Psoriatic Arthritis Joint Activity Index (PsAJAI), Disease Activity in Psoriatic Arthritis (DAPSA), and Composite Psoriatic Disease Activity Index (CPDAI). Arthritis Care Res. (Hoboken) 63(Suppl 11), S64-85. https://doi.org/10.1002/acr.20577 (2011).
doi: 10.1002/acr.20577
pubmed: 22588772
Salaffi, F., Ciapetti, A., Carotti, M., Gasparini, S. & Gutierrez, M. Disease activity in psoriatic arthritis: Comparison of the discriminative capacity and construct validity of six composite indices in a real world. Biomed. Res. Int. 2014, 528105. https://doi.org/10.1155/2014/528105 (2014).
doi: 10.1155/2014/528105
pubmed: 24967375
pmcid: 4055291
Lötsch, J. & Ultsch, A. Enhancing explainable machine learning by reconsidering initially unselected items in feature selection for classification. BioMedInformatics 2, 701–714 (2022).
doi: 10.3390/biomedinformatics2040047
Ihaka, R. & Gentleman, R. R: A language for data analysis and graphics. J. Comput. Graph. Stat. 5, 299–314. https://doi.org/10.1080/10618600.1996.10474713 (1996).
doi: 10.1080/10618600.1996.10474713
R Development Core Team. R: A Language and Environment for Statistical Computing. (2008).
Van Rossum, G. & Drake Jr, F. L. Python tutorial. Vol. 620 (Centrum voor Wiskunde en Informatica Amsterdam, 1995).
Hotelling, H. Analysis of a complex of statistical variables into principal components. J. Educ. Psychol. 24, 498–520. https://doi.org/10.1037/h0070888 (1933).
doi: 10.1037/h0070888
Pearson, K. L. I. I. I. On lines and planes of closest fit to systems of points in space. Lond. Edinb. Dublin Philos. Mag. J. Sci. 2, 559–572. https://doi.org/10.1080/14786440109462720 (1901).
doi: 10.1080/14786440109462720
Le, S., Josse, J. & Husson, F. C. FactoMineR: A package for multivariate analysis. J. Stat. Softw. 25, 1–18 (2008).
doi: 10.18637/jss.v025.i01
Kohonen, T. Self-organized formation of topologically correct feature maps. Biol. Cybernet. 43, 59–69 (1982).
doi: 10.1007/BF00337288
Ultsch, A. Maps for Visualization of High-Dimensional Data Spaces. WSOM, 225–230 (2003).
Lötsch, J., Lerch, F., Djaldetti, R., Tegeder, I. & Ultsch, A. Identification of disease-distinct complex biomarker patterns by means of unsupervised machine-learning using an interactive R toolbox (Umatrix). BMC Big Data Anal. https://doi.org/10.1186/s41044-41018-40032-41041 (2018).
doi: 10.1186/s41044-41018-40032-41041
Ultsch, A. & Lötsch, J. Machine-learned cluster identification in high-dimensional data. J. Biomed. Inform. 66, 95–104. https://doi.org/10.1016/j.jbi.2016.12.011 (2017).
doi: 10.1016/j.jbi.2016.12.011
pubmed: 28040499
pmcid: 5313598
Ultsch, A. & Sieman, H. P. Kohonen's self organizing feature maps for exploratory data analysis. in INNC'90, Int. Neural Network Conference. 305–308 (Kluwer, Dordrecht, Netherlands, 1990).
Lötsch, J. & Ultsch, A. in Advances in Intelligent Systems and Computing Vol. 295 (eds T. Villmann, F-M. Schleif, M. Kaden, & M Lange) 248–257 (Springer, 2014).
Pearson, K. On the criterion that a given system of deviations from the probable in the case of a correlated system of variables is such that it can be reasonably supposed to have arisen from random sampling. Philos. Mag. Ser. 5(50), 157–175 (1900).
doi: 10.1080/14786440009463897
Meyer, D., Zeileis, A. & Hornik, K. vcd: Visualizing Categorical Data. R package version 1.4-11. (2023).
Ho, T. K. in Proceedings of the Third International Conference on Document Analysis and Recognition (Volume 1)—Volume 1 278 (IEEE Computer Society, 1995).
Breiman, L. Random forests. Mach. Learn. 45, 5–32. https://doi.org/10.1023/a:1010933404324 (2001).
doi: 10.1023/a:1010933404324
Chen, R.-C., Dewi, C., Huang, S.-W. & Caraka, R. E. Selecting critical features for data classification based on machine learning methods. J. Big Data 7, 52. https://doi.org/10.1186/s40537-020-00327-4 (2020).
doi: 10.1186/s40537-020-00327-4
Couronné, R., Probst, P. & Boulesteix, A.-L. Random forest versus logistic regression: A large-scale benchmark experiment. BMC Bioinform. 19, 270. https://doi.org/10.1186/s12859-018-2264-5 (2018).
doi: 10.1186/s12859-018-2264-5
Svetnik, V. et al. Boosting: An ensemble learning tool for compound classification and QSAR modeling. J. Chem. Inf. Model. 45, 786–799. https://doi.org/10.1021/ci0500379 (2005).
doi: 10.1021/ci0500379
pubmed: 15921468
Xu, H. et al. When are Deep Networks really better than Decision Forests at small sample sizes, and how?, https://doi.org/10.48550/ARXIV.2108.13637 (2021).
Harris, C. R. et al. Array programming with NumPy. Nature 585, 357–362. https://doi.org/10.1038/s41586-020-2649-2 (2020).
doi: 10.1038/s41586-020-2649-2
pubmed: 32939066
pmcid: 7759461
The pandas development team. pandas-dev/pandas: Pandas. (Zenodo, 2010). https://doi.org/10.5281/zenodo.3509134
Virtanen, P. et al. SciPy 10: Fundamental algorithms for scientific computing in Python. Nat. Methods 17, 261–272. https://doi.org/10.1038/s41592-019-0686-2 (2020).
doi: 10.1038/s41592-019-0686-2
pubmed: 32015543
pmcid: 7056644
Pedregosa, F. et al. Scikit-learn: Machine learning in Python. J. Mach. Learn. Res. 12, 2825–2830 (2011).
Ultsch, A. & Lötsch, J. Computed ABC analysis for rational selection of most informative variables in multivariate data. PLoS ONE 10, e0129767. https://doi.org/10.1371/journal.pone.0129767 (2015).
doi: 10.1371/journal.pone.0129767
pubmed: 26061064
pmcid: 4465645
Juran, J. M. The non-Pareto principle; Mea culpa. Qual. Prog. 8, 8–9 (1975).
Lötsch, J. & Ultsch, A. Recursive computed ABC (cABC) analysis as a precise method for reducing machine learning based feature sets to their minimum informative size. Sci. Rep. 13, 5470. https://doi.org/10.1038/s41598-023-32396-9 (2023).
doi: 10.1038/s41598-023-32396-9
pubmed: 37016033
pmcid: 10073099
Varma, S. & Simon, R. Bias in error estimation when using cross-validation for model selection. BMC Bioinform. 7, 91. https://doi.org/10.1186/1471-2105-7-91 (2006).
doi: 10.1186/1471-2105-7-91
Good, P. I. Resampling Methods: A Practical Guide to Data Analysis (Birkhäuser, 2006).
Brodersen, K. H., Ong, C. S., Stephan, K. E. & Buhmann, J. M. in Pattern Recognition (ICPR), 2010 20th International Conference on. 3121–3124.
Peterson, W., Birdsall, T. & Fox, W. The theory of signal detectability. Trans. IRE Prof. Group Inf. Theory 4, 171–212. https://doi.org/10.1109/TIT.1954.1057460 (1954).
doi: 10.1109/TIT.1954.1057460
Ultsch, A. & Lötsch, J. Generative learning with emergent self-organizing neuronal networks. In Conference of the International Federation of Classification Societies. (2017).
Wishart, D. S. et al. DrugBank 5.0: a major update to the DrugBank database for 2018. Nucleic Acids Res. 46, D1074–D1082. https://doi.org/10.1093/nar/gkx1037 (2018).
Wishart, D. S. et al. DrugBank: A comprehensive resource for in silico drug discovery and exploration. Nucleic Acids Res. 34, D668-672. https://doi.org/10.1093/nar/gkj067 (2006).
doi: 10.1093/nar/gkj067
pubmed: 16381955
Ali, M. & Ezzat, A. dbparser: DrugBank Database XML Parser. R package version 2.0.1. (2023).
Anderson, J. et al. Rheumatoid arthritis disease activity measures: American College of Rheumatology recommendations for use in clinical practice. Arthritis Care Res. (Hoboken) 64, 640–647. https://doi.org/10.1002/acr.21649 (2012).
doi: 10.1002/acr.21649
pubmed: 22473918
Kruskal, W. H. & Wallis, W. A. Use of ranks in one-criterion variance anaylsis. J. Am. Stat. Assoc. 47, 583–621 (1952).
doi: 10.1080/01621459.1952.10483441
Perez-Chada, L. M. & Merola, J. F. Comorbidities associated with psoriatic arthritis: Review and update. Clin. Immunol. 214, 108397. https://doi.org/10.1016/j.clim.2020.108397 (2020).
doi: 10.1016/j.clim.2020.108397
pubmed: 32229290
Felten, R., Duret, P. M., Gottenberg, J. E., Spielmann, L. & Messer, L. At the crossroads of gout and psoriatic arthritis: “psout”. Clin. Rheumatol. 39, 1405–1413. https://doi.org/10.1007/s10067-020-04981-0 (2020).
doi: 10.1007/s10067-020-04981-0
pubmed: 32062768
Moll, J. M. & Wright, V. Psoriatic arthritis. Semin. Arthritis Rheum. 2 (1973).
Acosta Felquer, M. L. & FitzGerald, O. Peripheral joint involvement in psoriatic arthritis patients. Clin. Exp. Rheumatol. 33, S26-30 (2015).
pubmed: 26471860
Kessler, J. et al. Psoriatic arthritis and physical activity: A systematic review. Clin. Rheumatol. 40, 4379–4389. https://doi.org/10.1007/s10067-021-05739-y (2021).
doi: 10.1007/s10067-021-05739-y
pubmed: 33913069
McGonagle, D., Tan, A. L., Watad, A. & Helliwell, P. Pathophysiology, assessment and treatment of psoriatic dactylitis. Nat. Rev. Rheumatol. 15, 113–122. https://doi.org/10.1038/s41584-018-0147-9 (2019).
doi: 10.1038/s41584-018-0147-9
pubmed: 30610219
Prevoo, M. L. et al. Modified disease activity scores that include twenty-eight-joint counts. Development and validation in a prospective longitudinal study of patients with rheumatoid arthritis. Arthritis Rheum. 38, 44–48. https://doi.org/10.1002/art.1780380107 (1995).
doi: 10.1002/art.1780380107
pubmed: 7818570
Schoels, M. Psoriatic arthritis indices. Clin. Exp. Rheumatol. 32, S-109-S−112 (2014).
Ogdie, A., Coates, L. C. & Gladman, D. D. Treatment guidelines in psoriatic arthritis. Rheumatology (Oxford) 59, i37–i46. https://doi.org/10.1093/rheumatology/kez383 (2020).
doi: 10.1093/rheumatology/kez383
pubmed: 32159790
Gladman, D. et al. Tofacitinib for psoriatic arthritis in patients with an inadequate response to TNF inhibitors. N. Engl. J. Med. 377, 1525–1536. https://doi.org/10.1056/NEJMoa1615977 (2017).
doi: 10.1056/NEJMoa1615977
pubmed: 29045207
Creswell, A. & Bharath, A. A. Adversarial training for sketch retrieval (Springer International Publishing, Amsterdam, The Netherlands, 2016).
Cheng, Y. et al. Diagnosis of metacarpophalangeal synovitis with musculoskeletal ultrasound images. Ultrasound. Med. Biol. 48, 488–496. https://doi.org/10.1016/j.ultrasmedbio.2021.11.003 (2022).
doi: 10.1016/j.ultrasmedbio.2021.11.003
pubmed: 34930637
Mumtaz, A. et al. Development of a preliminary composite disease activity index in psoriatic arthritis. Ann. Rheum. Dis. 70, 272–277. https://doi.org/10.1136/ard.2010.129379 (2011).
doi: 10.1136/ard.2010.129379
pubmed: 21115550
Wickham, H. ggplot2: Elegant Graphics for Data Analysis (Springer, 2009).
doi: 10.1007/978-0-387-98141-3
Gu, Z., Eils, R. & Schlesner, M. Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics 32, 2847–2849. https://doi.org/10.1093/bioinformatics/btw313 (2016).
doi: 10.1093/bioinformatics/btw313
pubmed: 27207943
Lötsch, J. & Ultsch, A. Comparative assessment of projection and clustering method combinations in the analysis of biomedical data. (2023).
Cohen, A. On the graphical display of the significant components in a two-way contingency table. Commun. Stat. Theory Methods A9, 1025–1041 (1980).
doi: 10.1080/03610928008827940
Meyer, D., Zeileis, A. & Hornik, K. The Strucplot framework: Visualizing multi-way contingency tables with vcd. J. Stat. Softw. 17, 1–48 (2006).
doi: 10.18637/jss.v017.i03
Waskom, M. L. seaborn: Statistical data visualization. J. Open Source Softw. 6, 3021 (2021).
doi: 10.21105/joss.03021
Pedersen, T. ggforce: Accelerating ‘ggplot2'. R package version 0.4.1 (2022).
Attali, D. & Baker, C. ggExtra: Add Marginal Histograms to ‘ggplot2', and More ‘ggplot2' Enhancements. R package version 0.10.1. (2023).