Ensembles of convolutional neural network models for pediatric pneumonia diagnosis

Liz, Helena and Sánchez-Montañés, Manuel and Tagarro, Alfredo and Domínguez-Rodríguez, Sara and Dagan, Ron and Camacho, David (2021). Ensembles of convolutional neural network models for pediatric pneumonia diagnosis. "Future Generation Computer Systems-the International Journal of Escience", v. 122 ; pp. 220-233. ISSN 0167-739X. https://doi.org/10.1016/j.future.2021.04.007.


Title: Ensembles of convolutional neural network models for pediatric pneumonia diagnosis
  • Liz, Helena
  • Sánchez-Montañés, Manuel
  • Tagarro, Alfredo
  • Domínguez-Rodríguez, Sara
  • Dagan, Ron
  • Camacho, David
Item Type: Article
Título de Revista/Publicación: Future Generation Computer Systems-the International Journal of Escience
Date: September 2021
ISSN: 0167-739X
Volume: 122
Freetext Keywords: Ensembles of Convolutional Neural Networks: eXplainable Artificial Intelligence; Heatmaps; Pneumonia; Pediatrics
Faculty: E.T.S.I. de Sistemas Informáticos (UPM)
Department: Sistemas Informáticos
Creative Commons Licenses: Recognition - No derivative works - Non commercial

Full text

[img] PDF - Requires a PDF viewer, such as GSview, Xpdf or Adobe Acrobat Reader
Download (16MB)


Pneumonia is a lung infection that causes 15% of childhood mortality (under 5 years old), over 800,000 children under five every year, around 2,200 every day, all over the world. This pathology is mainly caused by viruses or bacteria. X-rays imaging analysis is one of the most used methods for pneumonia diagnosis. These clinical images can be analyzed using machine learning methods such as convolutional neural networks (CNN), which learn to extract critical features for the classification. However, the usability of these systems is limited in medicine due to the lack of interpretability, because of these models cannot be used to generate an understandable explanation (from a human-based perspective), about how they have reached those results. Another problem that difficults the impact of this technology is the limited amount of labeled data in many medicine domains. The main contributions of this work are two fold: the first one is the design of a new explainable artificial intelligence (XAI) technique based on combining the individual heatmaps obtained from each model in the ensemble. This allows to overcome the explainability and interpretability problems of the CNN ?black boxes?, highlighting those areas of the image which are more relevant to generate the classification. The second one is the development of new ensemble deep learning models to classify chest X-rays that allow highly competitive results using small datasets for training. We tested our ensemble model using a small dataset of pediatric X-rays (950 samples of children between one month and 16 years old) with low quality and anatomical variability (which represents one of the biggest challenges addressed in this work). We also tested other strategies such as single CNNs trained from scratch and transfer learning using CheXNet. Our results show that our ensemble model clearly outperforms these strategies obtaining highly competitive results. Finally we confirmed the robustness of our approach using another pneumonia diagnosis dataset.

Funding Projects

Government of SpainTIN2017-85727-C4-3-PDeepBioUniversidad Autónoma de MadridNuevos modelos de computo bioinspirado para entornos masivamente complejos
Universidad Politécnica de MadridPCI2019-103623CHIST-ERA 2017 BDSI PACMELUnspecifiedProcess-aware Analytics Support based on Conceptual Models for Event Logs (PACMEL)
Government of SpainPGC2018-095895-B-I00UnspecifiedUnspecifiedDinámica neuronal secuencial exploratoria: experimentos, formalismo teórica y aplicaciones
Madrid Regional GovernmentS2018/TCS-4566CYNAMONUnspecifiedCybersecurity, Network Analysis and Monitoring for the Next Generation Internet
Madrid Regional GovernmentS2017/BMD-3688UnspecifiedUnspecifiedUnspecified

More information

Item ID: 67830
DC Identifier: https://oa.upm.es/67830/
OAI Identifier: oai:oa.upm.es:67830
DOI: 10.1016/j.future.2021.04.007
Official URL: https://www.sciencedirect.com/science/article/pii/S0167739X2100128X
Deposited by: Memoria Investigacion
Deposited on: 22 Feb 2022 15:10
Last Modified: 30 Nov 2022 09:00
  • Logo InvestigaM (UPM)
  • Logo GEOUP4
  • Logo Open Access
  • Open Access
  • Logo Sherpa/Romeo
    Check whether the anglo-saxon journal in which you have published an article allows you to also publish it under open access.
  • Logo Dulcinea
    Check whether the spanish journal in which you have published an article allows you to also publish it under open access.
  • Logo de Recolecta
  • Logo del Observatorio I+D+i UPM
  • Logo de OpenCourseWare UPM