Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 2
Encontrada(s) 1 página(s)

RUC. Repositorio da Universidade da Coruña
Publication Artículo científico (JournalArticle)

Predicting total knee replacement in knee osteoarthritis using a machine learning-guided approach in patients of the Osteoarthritis Initiative (OAI)

  • Blanco García, Francisco J
  • Oreiro Villar, Natividad
  • Vázquez-García, Jorge
  • Morano Torres, Antonio
  • Balboa-Barreiro, Vanesa
  • Rodríguez-Valle, Isabel
  • Relaño, Sara
  • Veronese, Nicola
  • de Andrés, María C.
  • Rego-Pérez, I.
[Abstract] Objective To develop a pragmatic model to predict total knee replacement (TKR) in knee osteoarthritis using non-imaging clinical, genetic and lifestyle data with machine learning (ML)-guided feature selection. Methods We analysed 3790 Osteoarthritis Initiative participants. Nested ML feature selection on the training set identified 15 informative variables. Classifiers were benchmarked, then a multivariable logistic regression was fit on the full cohort. Performance was summarised by discrimination (area under the curve (AUC) with 95% CI) and calibration (Brier score). To assess the incremental value of genetics, we refit an otherwise identical clinical model excluding the Polygenic Risk Score (PRS) and compared specificity at fixed sensitivities using Bonferroni-adjusted McNemar tests. A prespecified analysis examined performance by baseline Kellgren-Lawrence (KL) grade (KL 0–1 vs KL ≥2). Results On the test set, classifier AUCs ranged 0.716–0.748, with Elastic Net and XGBoost performing best. The final logistic model fit on the full cohort achieved AUC 0.765 (95% CI 0.736 to 0.793) with acceptable calibration (Brier 0.097). Performance remained robust by disease stage, with higher discrimination in pre-radiographic knees (KL 0–1: AUC 0.827) and moderate discrimination in KL ≥2 (AUC 0.720); decile plots indicated broadly aligned observed versus predicted risks. PRS added modest, statistically significant gains in specificity at several fixed sensitivities without materially changing AUC. Conclusions We present a pragmatic, non-imaging, ML-informed model that predicts TKR with clinically acceptable discrimination and calibration using routinely collected data. This framework provides a practical basis for individualised risk stratification and decision support without reliance on imaging., This study has been funded by Instituto de Salud Carlos III (ISCIII) through the projects RD21/0002/0009, RD24/0007/0026, PMP22/00101, PMPTA22/00115, PI17/00210, PI22/01165, PI22/01155 and PI23/00913 and co-founded by the European Union. This work was also funded by grants IN607A 2021/07 and IN607D 2021/13 from Axencia Galega de Innovación-Xunta de Galicia. IRP is supported by Contrato Miguel Servet-II Fondo de Investigación Sanitaria (CPII17/00026) SERGAS-stabilized. JVG is supported by grant IN606A 2022/048 from Xunta de Galicia, Spain. This work was also funded by Pfizer and Elli Lilly and Company through the Third Global Awards for Advancing Chronic Pain Research, ADVANCE (grant ID#64122119). This study has been also supported by an Open Access Articles grant from Fundación Pública Galega de Investigación Biomédica INIBIC.
RUC. Repositorio da Universidade da Coruña
Publication Artículo científico (JournalArticle)

mtDNA haplogroup A enhances the effect of obesity on the risk of knee OA in a mexican population

  • Ramos-Louro, Paula
  • Arellano Pérez Vertti, Rubén Daniel
  • López-Reyes, Alberto
  • Martínez-Nava, Gabriela Angélica
  • Espinosa, Rolando
  • Pineda, Carlos
  • González Galarza, Faviel Francisco
  • Argüello Astorga, Rafael
  • Aguilar Muñiz, Lizette Sarai
  • Hernández Terán, Fernando
  • Parra Torres, Nancy Marbella
  • Durán-Sotuela, Alejandro
  • Fernández-Moreno, Mercedes
  • Balboa-Barreiro, Vanesa
  • Blanco García, Francisco J
  • Rego-Pérez, I.
[Abstract] To evaluate the influence of mitochondrial DNA haplogroups on the risk of knee OA in terms of their interaction with obesity, in a population from Mexico. Samples were obtained from (n = 353) knee OA patients (KL grade ≥ I) and (n = 364) healthy controls (KL grade = 0) from Mexico city and Torreon (Mexico). Both Caucasian and Amerindian mtDNA haplogroups were assigned by single base extension assay. A set of clinical and demographic variables, including obesity status, were considered to perform appropriate statistical approaches, including chi-square contingency tables, regression models and interaction analyses. To ensure the robustness of the predictive model, a statistical cross-validation strategy of B = 1000 iterations was used. All the analyses were performed using boot, GmAMisc and epiR package from R software v4.0.2 and SPSS software v24. The frequency distribution of the mtDNA haplogroups between OA patients and healthy controls for obese and non-obese groups showed the haplogroup A as significantly over-represented in knee OA patients within the obese group (OR 2.23; 95% CI 1.22-4.05; p-value = 0.008). The subsequent logistic regression analysis, including as covariate the interaction between obesity and mtDNA haplogroup A, supported the significant association of this interaction (OR 2.57; 95% CI 1.24-5.32; p-value = 0.011). The statistical cross-validation strategy confirmed the robustness of the regression model. The data presented here indicate a link between obesity in knee OA patients and mtDNA haplogroup A., This work is supported by Grants from Fondo de Investigación Sanitaria (PI17/00210, PI16/02124, PI20/00614, RETIC-RIER-RD16/0012/0002 and PRB3-ISCIII-PT17/0019/0014) integrated in the National Plan for Scientific Program, Development and Technological Innovation 2013–2016 and funded by the ISCIII-General Subdirection of Assessment and Promotion of Research-European Regional Development Fund (FEDER) “A way of making Europe” and Grant IN607A2017/11 from Xunta de Galicia. The authors further acknowledge AE CICA-INIBIC (ED431E 2018/03) for financial support. IRP is supported by Contrato Miguel Servet-II Fondo de Investigación Sanitaria (CPII17/00026) and AD-S is supported by Grant IN606A-2018/023 from Xunta de Galicia, Spain. The Biomedical Research Networking Center (CIBER) is an initiative from Instituto de Salud Carlos III (ISCIII), Xunta de Galicia; IN607A2017/11, Xunta de Galicia; ED431E 2018/03, Xunta de Galicia; IN606A-2018/023