Multiple instance learning for lung pathophysiological findings detection using CT scans

Data de publicação: Data Ahead of Print:

Autores da FMUP

  • Cláudia Susana Soares De Freitas

    Autor

  • Venceslau José Coelho Pinto Hespanhol

    Autor

Participantes de fora da FMUP

  • Frade, J
  • Pereira, T
  • Morgado, J
  • Silva, F
  • Mendes, J
  • Negrao, E
  • de Lima, BF
  • da Silva, MC
  • Madureira, A.
  • Ramos, I
  • Costa, JL
  • Cunha, A
  • Oliveira, HP

Unidades de investigação

Abstract

Lung diseases affect the lives of billions of people worldwide, and 4 million people, each year, die prematurely due to this condition. These pathologies are characterized by specific imagiological findings in CT scans. The traditional Computer-Aided Diagnosis (CAD) approaches have been showing promising results to help clinicians; however, CADs normally consider a small part of the medical image for analysis, excluding possible relevant information for clinical evaluation. Multiple Instance Learning (MIL) approach takes into consideration different small pieces that are relevant for the final classification and creates a comprehensive analysis of pathophysiological changes. This study uses MIL-based approaches to identify the presence of lung pathophysiological findings in CT scans for the characterization of lung disease development. This work was focus on the detection of the following: Fibrosis, Emphysema, Satellite Nodules in Primary Lesion Lobe, Nodules in Contralateral Lung and Ground Glass, being Fibrosis and Emphysema the ones with more outstanding results, reaching an Area Under the Curve (AUC) of 0.89 and 0.72, respectively. Additionally, the MIL-based approach was used for EGFR mutation status prediction - the most relevant oncogene on lung cancer, with an AUC of 0.69. The results showed that this comprehensive approach can be a useful tool for lung pathophysiological characterization.

Dados da publicação

ISSN/ISSNe:
1741-0444, 0140-0118

MEDICAL & BIOLOGICAL ENGINEERING & COMPUTING  Springer Verlag

Tipo:
Article
Páginas:
1569-1584
Link para outro recurso:
www.scopus.com

Citações Recebidas na Web of Science: 1

Citações Recebidas na Scopus: 8

Documentos

  • Não há documentos

Métricas

Filiações mostrar / ocultar

Keywords

  • Multiple instance learning; Computer-aided diagnosis; Computed tomography; Lung disease detection; Lung cancer characterization

Financiamento

Proyectos asociados

Doença do Refluxo Gastro-esofágico e Doença Pulmonar

Investigador Principal: Venceslau José Coelho Pinto Hespanhol

Estudo Clínico Académico (Refluxo Gastro-esofágic) . 2020

A importância do diagnóstico na Fibrose Pulmonar Idiopática

Investigador Principal: Venceslau José Coelho Pinto Hespanhol

Estudo Clínico Académico (FPI) . 2020

Imunoterapia no cancro do pulmão: PD-L1, biomarcador preditivo?

Investigador Principal: Venceslau José Coelho Pinto Hespanhol

Estudo Clínico Académico . 2020

Citar a publicação

Partilhar a publicação