Explore the vast range of titles available.

[This corrects the article DOI: 10.1371/journal.pone.0231524.].[This corrects the article DOI: 10.1371/journal.pone.0231524.].

[This corrects the article DOI: 10.1371/journal.pone.0340276.].

Background. Idiopathic inflammatory myopathies (IIM) are a group of rare and heterogeneous diseases. One of the greatest challenges in IIM management is defining and predicting disease flares, which are inconsistently characterized across studies. The aim of the study was to identify predictors of relapse in a retro-prospective, multi-centric IIM cohort during the first two years of disease.   Methods. Patients with IIM subsets—dermatomyositis (DM), polymyositis (PM), anti-synthetase syndrome (ASyS), connective tissue disease-associated IIM (CTD-IIM), or immune-mediated necrotizing myopathy (IMNM)—were included if >=18 yrs and with >= 2 years of follow-up. Relapse was defined as a change in disease activity requiring escalation of immunosuppressive therapy and/or steroids. Patients were classified as monocyclic (no relapse in the first 2 years), polycyclic (>=1 relapse in the first 2 years), or chronic continuous. Clinical characteristics across subgroups were compared using chi-square, Mann–Whitney U, or Student’s t-test. Predictors were identified through multivariable logistic regression.   Results. Of 297 screened patients, 91 were excluded due to missing data. A total of 206 patients were included (155 female; median age at diagnosis 46.9 years). 82 (39.8%) were DM, 56 (27.2%) ASyS, 40 (19.4%) CTD-IIM, 14 (6.8%) PM, and 14 (6.8%) IMNM. During the first 2 years of disease, 84 patients (40.8%) had monocyclic course, 67 (32.5%) polycyclic course, and 55 (26.7%) chronic continuous course. No significant differences in disease course distribution were observed across IIM subsets, although PM and IMNM were more frequently associated with chronic or relapsing patterns (64.2% vs. 35.8%). Myositis autoantibodies were not significantly linked to any disease course. Relapses occurred despite stable treatment for over one year in approximately one-third of cases. Most relapses involved muscle (61.9%), particularly in PM/IMNM (89%), whereas the frequency was notably lower in ASyS (43.2%), where ILD was the leading cause of relapse (61.4%). Refractory disease was the predominant phenotype in PM/IMNM (55.6%) and ASyS (52.3%), while drug tapering was most frequent cause implicated in DM (30%) and CTD-IIM (31%). Muscle and cardiac involvement at baseline were less frequent in monocyclic patients. A more severe disease phenotype—reflected by higher levels of muscle enzymes, and both PtGA and PhGA—was more commonly associated with non-monocyclic disease. Notably, non-monocyclic patients had received baseline low-dose steroids (<0.25 mg/kg; 78.9% vs. 21.1%, p = 0.019) and intravenous immunoglobulins (10.0% vs. 26.0%, p = 0.004). Multivariable analysis confirmed baseline muscle involvement, CK levels, and PtGA as independent predictors of relapse. -   Conclusions. In this large, diverse IIM cohort, non-monocyclic disease (polycyclic and chronic continuous) was associated with baseline myositis, higher serum muscle enzymes, higher global activity scores, and lower steroid induction dosage. These findings highlight the potential value of flare predictors and advocate for standardized definitions of disease activity/relapse in IIM.  

Visceral Leishmaniasis (VL), also known as Kala-Azar, poses a significant global public health challenge and is a neglected disease, with relapses and treatment failures leading to increased morbidity and mortality. This study introduces an explainable machine learning approach to predict VL relapse and identify critical risk factors, thereby aiding patient monitoring and treatment strategies. Leveraging data from a follow-up study of 571 patients, the survival machine learning models are applied, including Random Survival Forest (RSF), Survival Support Vector Machine (SSVM), and eXtreme Gradient Boosting (XGBoost), for relapse prediction. The results demonstrated that RSF, with a C-index of 0.85, outperformed the conventional Cox Proportional Hazard (CPH) model (C-index 0.8), offering improved prediction capabilities by capturing non-linear relationships and variable interactions. To address the lack of transparency (in terms of feature importance) in Machine Learning (ML) models, the SHapley Additive exPlanation (SHAP) method is employed, which enhances model interpretability (feature importance) through visual insights. SHAP dependence plots allowed the healthcare professionals to evaluate which factors encourage the occurrence of the relapse. A statistically significant relationship between HIV co-infection (HR=3.92, 95% CI=2.03–7.58) and VL relapse was identified through -2 log-likelihood ratio and chi-square tests. These results indicate the promise of explainable artificial intelligence (XAI) for making clinical decisions and remedying recurrences in VL.