EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Comparative study of convolutional neural network architectures for gastrointestinal lesions classification

Erik O Cuevas-Rodriguez, Carlos E Galvan-Tejada iD, Valeria Maeda-Gutiérrez, Gamaliel Moreno-Chávez, Jorge I Galván-Tejada, Hamurabi Gamboa-Rosales iD, Huizilopoztli Luna-García, Arturo Moreno-Baez, José María Celaya-Padilla

PeerJ · 16.03.2023 · 11 · e14806

DOI: 10.7717/peerj.14806 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PeerJ, online erschienen am 16.03.2023. Verfasst von Erik O Cuevas-Rodriguez und 8 weiteren Autor:innen. Beteiligte Einrichtung: Unidad Académica de Ingeniería Eléctrica, Universidad Autónoma de Zacatecas, Zacatecas, Zacatecas, México. Hauptthemen laut MEDLINE: Deep Learning. Die Arbeit wurde bisher 17-mal zitiert (Crossref) und nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The gastrointestinal (GI) tract can be affected by different diseases or lesions such as esophagitis, ulcers, hemorrhoids, and polyps, among others. Some of them can be precursors of cancer such as polyps. Endoscopy is the standard procedure for the detection of these lesions. The main drawback of this procedure is that the diagnosis depends on the expertise of the doctor. This means that some important findings may be missed. In recent years, this problem has been addressed by deep learning (DL) techniques. Endoscopic studies use digital images. The most widely used DL technique for image processing is the convolutional neural network (CNN) due to its high accuracy for modeling complex phenomena. There are different CNNs that are characterized by their architecture. In this article, four architectures are compared: AlexNet, DenseNet-201, Inception-v3, and ResNet-101. To determine which architecture best classifies GI tract lesions, a set of metrics; accuracy, precision, sensitivity, specificity, F1-score, and area under the curve (AUC) were used. These architectures were trained and tested on the HyperKvasir dataset. From this dataset, a total of 6,792 images corresponding to 10 findings were used. A transfer learning approach and a data augmentation technique were applied. The best performing architecture was DenseNet-201, whose results were: 97.11% of accuracy, 96.3% sensitivity, 99.67% specificity, and 95% AUC.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningNeural Networks, ComputerGastrointestinal TractEndoscopy, GastrointestinalDiagnosis, Computer-Assisted

Stichwörter der Autor:innen

ClassificationComputer-aided diagnosticConvolutional neural networkDeep learningEndoscopyGastrointestinalGastrointestinal lesions

Autor:innen und Institutionen

  • Unidad Académica de Ingeniería Eléctrica, Universidad Autónoma de Zacatecas, Zacatecas, Zacatecas, México.

ORCID-Profile: Carlos E Galvan-Tejada, Hamurabi Gamboa-Rosales

Zitieren

APA

Cuevas-Rodriguez, E. O., Galvan-Tejada, C. E., Maeda-Gutiérrez, V., Moreno-Chávez, G., Galván-Tejada, J. I., Gamboa-Rosales, H., Luna-García, H., Moreno-Baez, A., & Celaya-Padilla, J. M. (2023). Comparative study of convolutional neural network architectures for gastrointestinal lesions classification. PeerJ, 11, e14806. https://doi.org/10.7717/peerj.14806

Vancouver

Cuevas-Rodriguez EO, Galvan-Tejada CE, Maeda-Gutiérrez V, Moreno-Chávez G, Galván-Tejada JI, Gamboa-Rosales H, et al. Comparative study of convolutional neural network architectures for gastrointestinal lesions classification. PeerJ. 2023;11:e14806. doi:10.7717/peerj.14806

BibTeX
@article{CuevasRodriguez2023,
  author = {Cuevas-Rodriguez, Erik O and Galvan-Tejada, Carlos E and Maeda-Gutiérrez, Valeria and Moreno-Chávez, Gamaliel and Galván-Tejada, Jorge I and Gamboa-Rosales, Hamurabi and Luna-García, Huizilopoztli and Moreno-Baez, Arturo and Celaya-Padilla, José María},
  title = {Comparative study of convolutional neural network architectures for gastrointestinal lesions classification},
  journal = {PeerJ},
  year = {2023},
  volume = {11},
  pages = {e14806},
  doi = {10.7717/peerj.14806}
}
Kostenlos starten →