EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Negative Samples for Improving Object Detection—A Case Study in AI-Assisted Colonoscopy for Polyp Detection

Alba Nogueira-Rodríguez iD, Daniel Glez-Peña, Miguel Reboiro-Jato iD, Hugo López-Fernández iD

Diagnostics · 03.03.2023 · 13 (5) · 966

DOI: 10.3390/diagnostics13050966 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Diagnostics, online erschienen am 03.03.2023. Verfasst von Alba Nogueira-Rodríguez und 3 weiteren Autor:innen. Beteiligte Einrichtung: CINBIO, Department of Computer Science, ESEI-Escuela Superior de Ingeniería Informática, Universidade de Vigo, 32004 Ourense, Spain. Die Arbeit wurde bisher 19-mal zitiert (Crossref) und nennt 26 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministerio de Ciencia y Competitividad, Gobierno de España, Ministerio de Ciencia e Innovación, Gobierno de España, Consellería de Educación, Universidades e Formación Profesional (Xunta de Galicia).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Deep learning object-detection models are being successfully applied to develop computer-aided diagnosis systems for aiding polyp detection during colonoscopies. Here, we evidence the need to include negative samples for both (i) reducing false positives during the polyp-finding phase, by including images with artifacts that may confuse the detection models (e.g., medical instruments, water jets, feces, blood, excessive proximity of the camera to the colon wall, blurred images, etc.) that are usually not included in model development datasets, and (ii) correctly estimating a more realistic performance of the models. By retraining our previously developed YOLOv3-based detection model with a dataset that includes 15% of additional not-polyp images with a variety of artifacts, we were able to generally improve its F1 performance in our internal test datasets (from an average F1 of 0.869 to 0.893), which now include such type of images, as well as in four public datasets that include not-polyp images (from an average F1 of 0.695 to 0.722).

Schlagwörter

Stichwörter der Autor:innen

colorectal cancerconvolutional neural network (CNN)deep learningpolyp detectionpolyp localization

Autor:innen und Institutionen

  • CINBIO, Department of Computer Science, ESEI-Escuela Superior de Ingeniería Informática, Universidade de Vigo, 32004 Ourense, Spain.
  • SING Research Group, Galicia Sur Health Research Institute (IIS Galicia Sur), SERGAS-UVIGO, 36213 Vigo, Spain.

ORCID-Profile: Alba Nogueira-Rodríguez, Miguel Reboiro-Jato, Hugo López-Fernández

Förderung

  • Ministerio de Ciencia y Competitividad, Gobierno de España
  • Ministerio de Ciencia e Innovación, Gobierno de España
  • Consellería de Educación, Universidades e Formación Profesional (Xunta de Galicia)
  • Ministerio de Ciencia y Competitividad and Ministerio de Ciencia e Innovación, Gobierno de España

Zitieren

APA

Nogueira-Rodríguez, A., Glez-Peña, D., Reboiro-Jato, M., & López-Fernández, H. (2023). Negative Samples for Improving Object Detection—A Case Study in AI-Assisted Colonoscopy for Polyp Detection. Diagnostics, 13(5), 966. https://doi.org/10.3390/diagnostics13050966

Vancouver

Nogueira-Rodríguez A, Glez-Peña D, Reboiro-Jato M, López-Fernández H. Negative Samples for Improving Object Detection—A Case Study in AI-Assisted Colonoscopy for Polyp Detection. Diagnostics. 2023;13(5):966. doi:10.3390/diagnostics13050966

BibTeX
@article{NogueiraRodrguez2023,
  author = {Nogueira-Rodríguez, Alba and Glez-Peña, Daniel and Reboiro-Jato, Miguel and López-Fernández, Hugo},
  title = {Negative Samples for Improving Object Detection—A Case Study in AI-Assisted Colonoscopy for Polyp Detection},
  journal = {Diagnostics},
  year = {2023},
  volume = {13},
  number = {5},
  pages = {966},
  doi = {10.3390/diagnostics13050966}
}
Kostenlos starten →