Consiglio Nazionale delle Ricerche

Tipo di prodottoArticolo in rivista
TitoloHierarchical Graph Representations in Digital Pathology
Anno di pubblicazione2021
Autore/iP. Pati, G. Jaume, A. Foncubierta, F. Feroce, A. M. Anniciello, G. Scognamiglio, N. Brancati, M.Fiche, E. Dubruc, D.Riccio, M. Di Bonito, G. De Pietro, G. Botti, J. P. Thiran, M. Frucci, O. Goksel, M. Gabrani
Affiliazioni autoriETH Zurich Zurich, Switzerland; EPFL Lausanne, Switzerland; ETH Zurich Zurich, Switzerland; IRCCS-Fondazione Pascale Naples, Italy; IRCCS-Fondazione Pascale Naples, Italy; IRCCS-Fondazione Pascale Naples, Italy; ICAR-CNR Naples, Italy; Aurigen,Switzerland; CHUV,Switzerland; ICAR-CNR Naples, Italy; IRCCS-Fondazione Pascale Naples, Italy; ICAR-CNR Naples, Italy; IRCCS-Fondazione Pascale Naples, Italy; EPFL ausanne, Switzerland; ICAR-CNR Naples, Italy; ETH Zurich Zurich, Switzerland; IBM Research Zurich, Switzerland;
Autori CNR e affiliazioni
  • inglese
Abstractancer diagnosis, prognosis, and therapy responsepredictions from tissue specimens highly depend on the phe-notype and topological distribution of constituting histologicalentities. Thus, adequate tissue representations for encodinghistological entities is imperative for computer aided cancerpatient care. To this end, several approaches have leveragedcell-graphs that encode cell morphology and cell organizationto denote the tissue information. These allow for utilizing graphtheory and machine learning to map tissue representationsto tissue functionality to help quantify their relationship.Though cellular information is crucial, it is incomplete aloneto comprehensively characterize complex tissue structure. Weherein treat the tissue as a hierarchical composition of multipletypes of histological entities from fine to coarse level, capturingmultivariate tissue information at multiple levels. We proposea novel multi-level hierarchical entity-graph representationof tissue specimens to model hierarchical compositions thatencode histological entities as well as their intra- and inter-entity level interactions. Subsequently, a hierarchical graphneural network is proposed to operate on the hierarchicalentity-graph representation to map the tissue structure to tissuefunctionality. Specifically, for input histology images we utilizewell-defined cells and tissue regions to build HierArchical Cell-to-Tissue (HACT) graph representations, and deviseHACT-Net, a message passing graph neural network, to classify suchHACTrepresentations. As part of this work, we introducethe BReAst Carcinoma Subtyping (BRACS) dataset, a largecohort of Haematoxylin & Eosin stained breast tumor regions-of-interest, to evaluate and benchmark our proposed method-ology against pathologists and state-of-the-art computer-aideddiagnostic approaches. Through comparative assessment andablation studies, our proposed method is demonstrated to yieldsuperior classification results compared to alternative methodsas well as individual pathologists.
Lingua abstractinglese
Altro abstract-
Lingua altro abstract-
Pagine da-
Pagine a-
Pagine totali19
Titolo chiave: arXiv
Numero volume della rivista2102.11057v2
Fascicolo della rivista-
Verificato da refereeNo
Stato della pubblicazionePreprint
Indicizzazione (in banche dati controllate)-
Parole chiaveDigital pathology, Breast cancer classification, Hi-erarchical tissue representation, Hierarchical graph neuralnetwork, Breast cancer datase
Link (URL, URI)
Titolo parallelo-
Scadenza embargo-
Data di accettazione-
Note/Altre informazioni-
Strutture CNR
  • ICAR — Istituto di calcolo e reti ad alte prestazioni
Moduli/Attività/Sottoprogetti CNR
  • DIT.AD022.156.002 : GRUPPI DI RICERCA - Napoli
Progetti Europei-
Hierarchical Graph Representations in Digital Pathology (documento privato )
Descrizione: versione accessibile online
Tipo documento: application/pdf