MOSAIC: An Artificial Intelligence–Based Framework for Multimodal Analysis, Classification, and Personalized Prognostic Assessment in Rare Cancers
JCO Clinical Cancer Informatics · 2024-06 · (8)
DOI: 10.1200/cci.24.00008 ↗Kurz gefasst
Fachartikel in der Zeitschrift JCO Clinical Cancer Informatics, 2024-06. Verfasst von Saverio D'Amico und 30 weiteren Autor:innen. Beteiligte Einrichtung: Humanitas Clinical and Research Center-IRCCS, Milan, Italy. Hauptthemen laut MEDLINE: Artificial Intelligence, Precision Medicine. Die Arbeit wurde bisher 19-mal zitiert (Crossref) und nennt 29 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.
Abstract (Originalsprache)
PURPOSE Rare cancers constitute over 20% of human neoplasms, often affecting patients with unmet medical needs. The development of effective classification and prognostication systems is crucial to improve the decision-making process and drive innovative treatment strategies. We have created and implemented MOSAIC, an artificial intelligence (AI)–based framework designed for multimodal analysis, classification, and personalized prognostic assessment in rare cancers. Clinical validation was performed on myelodysplastic syndrome (MDS), a rare hematologic cancer with clinical and genomic heterogeneities. METHODS We analyzed 4,427 patients with MDS divided into training and validation cohorts. Deep learning methods were applied to integrate and impute clinical/genomic features. Clustering was performed by combining Uniform Manifold Approximation and Projection for Dimension Reduction + Hierarchical Density-Based Spatial Clustering of Applications with Noise (UMAP + HDBSCAN) methods, compared with the conventional Hierarchical Dirichlet Process (HDP). Linear and AI-based nonlinear approaches were compared for survival prediction. Explainable AI (Shapley Additive Explanations approach [SHAP]) and federated learning were used to improve the interpretation and the performance of the clinical models, integrating them into distributed infrastructure. RESULTS UMAP + HDBSCAN clustering obtained a more granular patient stratification, achieving a higher average silhouette coefficient (0.16) with respect to HDP (0.01) and higher balanced accuracy in cluster classification by Random Forest (92.7% ± 1.3% and 85.8% ± 0.8%). AI methods for survival prediction outperform conventional statistical techniques and the reference prognostic tool for MDS. Nonlinear Gradient Boosting Survival stands in the internal (Concordance-Index [C-Index], 0.77; SD, 0.01) and external validation (C-Index, 0.74; SD, 0.02). SHAP analysis revealed that similar features drove patients' subgroups and outcomes in both training and validation cohorts. Federated implementation improved the accuracy of developed models. CONCLUSION MOSAIC provides an explainable and robust framework to optimize classification and prognostic assessment of rare cancers. AI-based approaches demonstrated superior accuracy in capturing genomic similarities and providing individual prognostic information compared with conventional statistical methods. Its federated implementation ensures broad clinical application, guaranteeing high performance and data protection.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Humanitas Clinical and Research Center-IRCCS, Milan, Italy.
- Train s.r.l., Milan, Italy.
- Department of Physics and Astronomy (DIFA), Bologna, Italy.
- Computational Biomedicine Unit, Department of Medical Sciences, University of Turin, Turin, Italy.
- Department of Signals, Systems and Radiocommunications, Polytechnic University of Madrid, Madrid, Spain.
- University of Copenhagen, Copenhagen, Denmark.
ORCID-Profile: Saverio D'Amico, Lorenzo Dall'Olio, Patricia Alonso, Daniele Dall'Olio, Claudia Sala, Elisabetta Sauta, Gianluca Asti, Luca Lanino, Elena Zazzetti, Mattia Delleani, Maria Elena Bicchieri, Pierandrea Morandini, Juan Parras, Lin Pierre Zhao, Uwe Platzbecker, Maria Diez-Campelo, Valeria Santini, Piero Fariselli, Tiziana Sanavia, Matteo Giovanni Della Porta, Gastone Castellani
Zitieren
APA
D'Amico, S., Dall'Olio, L., Rollo, C., Alonso, P., Prada-Luengo, I., Dall'Olio, D., Sala, C., Sauta, E., Asti, G., Lanino, L., Maggioni, G., Campagna, A., Zazzetti, E., Delleani, M., Bicchieri, M. E., Morandini, P., Savevski, V., Arroyo, B., Parras, J., …, & Castellani, G. (2024). MOSAIC: An Artificial Intelligence–Based Framework for Multimodal Analysis, Classification, and Personalized Prognostic Assessment in Rare Cancers. JCO Clinical Cancer Informatics. https://doi.org/10.1200/cci.24.00008
Vancouver
D'Amico S, Dall'Olio L, Rollo C, Alonso P, Prada-Luengo I, Dall'Olio D, et al. MOSAIC: An Artificial Intelligence–Based Framework for Multimodal Analysis, Classification, and Personalized Prognostic Assessment in Rare Cancers. JCO Clinical Cancer Informatics. 2024. doi:10.1200/cci.24.00008
BibTeX
@article{DAmico2024,
author = {D'Amico, Saverio and Dall'Olio, Lorenzo and Rollo, Cesare and Alonso, Patricia and Prada-Luengo, Iñigo and Dall'Olio, Daniele and Sala, Claudia and Sauta, Elisabetta and Asti, Gianluca and Lanino, Luca and Maggioni, Giulia and Campagna, Alessia and Zazzetti, Elena and Delleani, Mattia and Bicchieri, Maria Elena and Morandini, Pierandrea and Savevski, Victor and Arroyo, Borja and Parras, Juan and Zhao, Lin Pierre and Platzbecker, Uwe and Diez-Campelo, Maria and Santini, Valeria and Fenaux, Pierre and Haferlach, Torsten and Krogh, Anders and Zazo, Santiago and Fariselli, Piero and Sanavia, Tiziana and Della Porta, Matteo Giovanni and Castellani, Gastone},
title = {MOSAIC: An Artificial Intelligence–Based Framework for Multimodal Analysis, Classification, and Personalized Prognostic Assessment in Rare Cancers},
journal = {JCO Clinical Cancer Informatics},
year = {2024},
number = {8},
doi = {10.1200/cci.24.00008}
}
EUVIMED