Background/Objectives: Silicosis, a fibrotic lung disease, is re-emerging globally, driven by an aggressive form linked to engineered stone processing that rapidly progresses to progressive massive fibrosis (PMF). The standard diagnostic approach, chest X-ray (CXR), is subject to considerable inter-observer variability, making the distinction between simple silicosis (SS) and PMF particularly challenging. The purpose of this study was to develop and validate an automated multimodal framework for silicosis staging by integrating artificial intelligence (AI), CXR images, and routine blood biomarkers. Methods: We developed three fusion architectures, early, late, and hybrid, connecting blood biomarker analysis with CXR analysis. Deep learning and conventional (shallow) machine learning models were combined. The models were trained and validated on a cohort of 94 patients with engineered stone silicosis, providing 341 paired CXR and biomarker samples. A patient-aware 5-fold cross-validation was used to ensure the model’s generalizability and prevent patient data leakage between folds. Results: The hybrid and late fusion models achieved the best performance for disease staging, yielding an area under the receiver operating characteristic (ROC) curve (AUC) of 0.85. This multimodal approach outperformed both the unimodal CXR-based model (AUC = 0.83) and the biomarker-based model (AUC = 0.70). Conclusions: This study reveals that AI-based techniques that utilize a multimodal fusion approach have the potential to outperform single-modality methods have the potential to serve as an objective decision support tool for clinicians, leading to more consistent staging and improved patient management.

Background/Objectives: Silicosis caused by dust from engineered stone (ES) exposure is an emerging occupational lung disease that severely impacts respiratory health. This study aimed to analyze the association between cytokine profiles and disease severity and progression in patients with engineered stone silicosis (ESS) to assess their potential as biomarkers of progression and their usefulness to stratify risk. Methods: A longitudinal study was conducted with a seven-year follow-up (2017-2024) on 72 workers with simple silicosis (SS) or progressive massive fibrosis (PMF), all with a history of cutting, polishing, and finishing ES countertops. Data on lung function and levels of 27 cytokines were collected at four control points. Machine learning (ML) models were built to classify the disease stage and predict its progression. Results: 39% of patients with SS progressed to PMF. Significant differences in the expression of some cytokines were observed between ESS stages, suggesting a role in the evolution of the inflammatory process. Specifically, higher levels of IL-1RA, IL-8, IL-9, and IFN-$\gamma$ were found at checkpoint 1 in patients with PMF compared to SS. The longitudinal analysis revealed a significant relationship between IL-1RA and MCP-1$\alpha$ and disease duration with MCP-1$\alpha$ also being associated with time and disease grade. Machine learning (ML) models were built using the cytokines selected through a sequential backward feature selection. The Support Vector Machine model achieved an accuracy of 83% in classifying disease stage (SS, PMF), and of 77% in predicting the disease progression. Conclusions: The findings suggest that cytokines can be used as dynamic biomarkers to reflect underlying inflammatory processes and monitor disease evolution. The performance of ML algorithms to predict diagnostic status based on cytokine profiles highlights their clinical value in supporting early diagnosis, monitoring disease progression, and guiding clinical decisions.