Application of machine learning techniques to explore the occurrence of macrophage activation syndrome in Still’s disease: results from the GIRRCS AOSD Study Group and the AIDA Network Still’s Disease Registry

Objectives: This study aims to explore the application of machine learning techniques in assessing macrophage activation syndrome (MAS) in Still's disease. Methods: A multicenter, observational, prospective study was conducted, including patients with Still's disease enrolled in the Gruppo Italiano di Ricerca in Reumatologia Clinica e Sperimentale (GIRRCS) AOSD Study Group and the AutoInflammatory Disease Alliance (AIDA) Network Still's Disease Registry. Results: A total of 737 patients (age: 35.5 ± 17.8, male sex: 44.7%) with Still's disease were assessed; 11.4% were affected by MAS, and 3% had a poor prognosis. First, random forest imputation was applied to the original dataset. Subsequently, a machine-learning-driven assessment was developed to explore MAS occurrence. Collectively, regression models, an exploration decision tree, and a random forest were applied, suggesting the importance of ferritin, age, C-reactive protein (CRP), and systemic score. A logistic regression model accounting for data leakage concerns was then generated using these variables, and missing values were imputed using random forest imputation. This analysis supported the role of the selected variables, which were further combined across different clinical scenarios to estimate the probability of MAS. The highest risk of MAS was estimated for patients simultaneously characterized by age ≥ 45 years, ferritin ≥ 4,178.10 ng/mL, CRP ≥ 27.15 mg/L, and a systemic score ≥ 7, corresponding to a 34.7% probability of MAS, as well as for those characterized by ferritin ≥ 4,178.10 ng/mL, CRP ≥ 27.15 mg/L, and systemic score ≥ 7, corresponding to a 33.5% probability of MAS. Conclusions: A machine-learning-driven prediction of MAS was explored in Still's disease, highlighting the importance of age of onset, hyperferritinaemia, increased CRP, and multiorgan involvement. A combination of these features may suggest a clinician-friendly algorithm for stratifying the probability of MAS during Still's disease.

Authors

Institutions

Publication Details

Journal
Frontiers in Immunology
Published
2026-04-14
DOI
https://doi.org/10.3389/fimmu.2026.1811317
Primary Topic
Autoimmune and Inflammatory Disorders Research
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Application of machine learning techniques to explore the occurrence of macrophage activation syndrome in Still’s disease: results from the GIRRCS AOSD Study Group and the AIDA Network Still’s Disease Registry

Bruno Frediani, Andrea Hinojosa‐Azaola, Moustafa Ali Saad, Ibrahim A. AlMaghlouth et al.
Frontiers in Immunology
Autoimmune and Inflammatory Disorders Research
article

Application of machine learning techniques to explore the occurrence of macrophage activation syndrome in Still’s disease: results from the GIRRCS AOSD Study Group and the AIDA Network Still’s Disease Registry

Bruno Frediani, Andrea Hinojosa‐Azaola, Moustafa Ali Saad, Ibrahim A. AlMaghlouth, F. Iannone, Αναστάσιος Καραμανάκος, Benson Ogunjimi, G. H. Ragab, Gian Domenico Sebastiani, Amato de Paulis, Carlomaurizio Montecucco, Valentina Pucino, Maria Sole Chimenti, Emanuela Del Giudice, Valeria Caggiano, Paola Cipriani, Lampros Fotis, Ombretta Viapiana, Lorenzo Dagna, Paola Parronchi, Ludovico De Stefano, Piero Ruscitti, Samar Tharwat, Paolo Sfriso, Alberto Balistreri, Luca Navarini, Francesco Ursini, Sara Bindoli, Francesco Carubbi, Marco Valenti, Edoardo Biancalana, Mahmoud Ghanema, H Direskeneli, Katerina Kourtesi, Francesco Caso, B. Suzon, P P Sfikakis, Lidia La Barbera, Ewa Więsik–Szewczyk, Maria Morrone, E. Bartoloni, Francesco Masedu, İbrahim Yahya Çakır, Angela Mauro, Annamaria Iagnocco, Federico Perosa, Donato Rigante, Maria Cristina Maggio, Eduardo Martín‐Nares, Luisa Costa, J Pelizza, Şükran Erten, Serena Bugatti, Antonio Gidaro, Flavia Riccio, Giuliana Guggino, Alberto Lo Gullo, Rabia Deniz, Alessandro Conforti, Ilenia Di Cola, Micol Frassi, Carla Gaggiano, Marcella Prete, Francesco La Torre, Daniela Iacono, Maissa Thabet, Ezgi D. Batu, Patrizia Barone, Giacomo Emmi, Oksana Boyarchuk, Maria Alessio, Jessica Sbalchiero, Henrique A. Mayrink Giardini, Jiram Torres-Ruiz, Abdelhfeez Moshrif, Antonio Vitale, Tetiana Kovalchuk, Özgül Soysal Gündüz, Piercarlo Sarzi-Puttini, Mariachiara Visconti, Corrado Campochiaro, Francesco Ciccia, Giovanni Conti, Fabiola Atzeni, Claudia Fabiani, Marcello Govoni, Luca Cantarini, Seza Ozen, Alma Nunzia Olivieri, Roberto Giacomelli, Luciana Breda, Sulaiman M. Al-Mayouf, Cemal Bes, Joanna Makowska, José Hernández-Rodríguez, Abdurrahman Tufan, Giuseppe Lopalco
article en

Abstract

Objectives: This study aims to explore the application of machine learning techniques in assessing macrophage activation syndrome (MAS) in Still's disease. Methods: A multicenter, observational, prospective study was conducted, including patients with Still's disease enrolled in the Gruppo Italiano di Ricerca in Reumatologia Clinica e Sperimentale (GIRRCS) AOSD Study Group and the AutoInflammatory Disease Alliance (AIDA) Network Still's Disease Registry. Results: A total of 737 patients (age: 35.5 ± 17.8, male sex: 44.7%) with Still's disease were assessed; 11.4% were affected by MAS, and 3% had a poor prognosis. First, random forest imputation was applied to the original dataset. Subsequently, a machine-learning-driven assessment was developed to explore MAS occurrence. Collectively, regression models, an exploration decision tree, and a random forest were applied, suggesting the importance of ferritin, age, C-reactive protein (CRP), and systemic score. A logistic regression model accounting for data leakage concerns was then generated using these variables, and missing values were imputed using random forest imputation. This analysis supported the role of the selected variables, which were further combined across different clinical scenarios to estimate the probability of MAS. The highest risk of MAS was estimated for patients simultaneously characterized by age ≥ 45 years, ferritin ≥ 4,178.10 ng/mL, CRP ≥ 27.15 mg/L, and a systemic score ≥ 7, corresponding to a 34.7% probability of MAS, as well as for those characterized by ferritin ≥ 4,178.10 ng/mL, CRP ≥ 27.15 mg/L, and systemic score ≥ 7, corresponding to a 33.5% probability of MAS. Conclusions: A machine-learning-driven prediction of MAS was explored in Still's disease, highlighting the importance of age of onset, hyperferritinaemia, increased CRP, and multiorgan involvement. A combination of these features may suggest a clinician-friendly algorithm for stratifying the probability of MAS during Still's disease.

Frontiers in ImmunologyVol. 17
University of Siena (IT), Università Cattolica del Sacro Cuore (IT), University of Pisa (IT), University of Messina (IT), University of Rome Tor Vergata (IT), National Academy of Medicine (US), University of Trieste (IT), University of Maryland Medical System (US), Cairo University (EG), Ankara University (TR), Vita-Salute San Raffaele University (IT), Università Campus Bio-Medico (IT), Mansoura University (EG), Nara University (JP), Manisa Celal Bayar University (TR), National and Kapodistrian University of Athens (GR), University of Ferrara (IT), University of Pavia (IT), University of L'Aquila (IT), University of Perugia (IT), King Faisal Specialist Hospital & Research Centre (SA), Monash Medical Centre (AU), Martin University (US), King Saud University (SA), Agostino Gemelli University Polyclinic (IT), University of Chieti-Pescara (IT), Luigi Sacco Hospital (IT), Hôpital Farhat Hached (TN), Azienda Ospedaliera Universitaria Integrata Verona (IT), Azienda Ospedaliera San Camillo-Forlanini (IT), Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán (MX), San Salvatore Hospital (IT), Memorial Ankara Hospital (TR), Antwerp University Hospital (BE), Department of Health (ES), IRCCS Ospedale Galeazzi - Sant'Ambrogio (IT), Ospedale San Paolo (IT), Laiko General Hospital of Athens (GR), Infectious Diseases Institute (UG), Federico II University Hospital (IT), Uman National University of Horticulture (UA), Istituto Ortopedico Rizzoli (IT), Medical University of Lodz (PL), Gazi Hastanesi (TR), Al-Azhar University (ID), Arcispedale Sant'Anna (IT), University of Campania "Luigi Vanvitelli" (IT), Mother Hospital (IN), Universitair Ziekenhuis Brussel (BE), Hospital Materno-Infantil (ES), IRCCS Ospedale San Raffaele (IT), Centre Hospitalier Universitaire de Martinique (MQ), Azienda Ospedaliera Universitaria Policlinico "G. Martino" (IT), University General Hospital Attikon (GR), Ospedale Pediatrico Giovanni XXIII (IT), ERN GUARD-Heart (NL), Campus Bio Medico University Hospital (IT), Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo (BR), Policlinico San Matteo Fondazione (IT), Istituti di Ricovero e Cura a Carattere Scientifico (IT), ASST Fatebenefratelli Sacco (IT), I.Horbachevsky Ternopil National Medical University (UA), Center for Rheumatology (US), Consorci Institut D'Investigacions Biomediques August Pi I Sunyer (ES), Al Azhar University (QA), Azienda Ospedaliera Ospedali Riuniti Papardo Piemonte (IT), Martin University Hospital (SK), Evangelismos Hospital (GR), University of Florence (IT), University of Turin (IT), University of Bari Aldo Moro (IT), Hacettepe University (TR), University of Naples Federico II (IT), Marmara University (TR), Sapienza University of Rome (IT), University of Health Sciences Antigua (AG), University of Bologna (IT), Gazi University (TR)
Openalex Percentile: Top 23%
Autoimmune and Inflammatory Disorders Research
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.