Detail publikace

Deep convolutional networks for OCT image classification

HESKO, B.

Originální název

Deep convolutional networks for OCT image classification

Anglický název

Deep convolutional networks for OCT image classification

Jazyk

en

Originální abstrakt

In this work, OCT (optical coherence tomography) images are classified according to the present pathology into four distinct categories. Three different neural network models are used to classify images, each model is recent and we are achieving exceptional results on the testing dataset, which was unknown to the network during the training. Accuracy on the testing set is higher than 98% and only a few of images are classified into the wrong category. This makes our approach perspective for future automatic use. To further improve results, all three models are using transfer learning.

Anglický abstrakt

In this work, OCT (optical coherence tomography) images are classified according to the present pathology into four distinct categories. Three different neural network models are used to classify images, each model is recent and we are achieving exceptional results on the testing dataset, which was unknown to the network during the training. Accuracy on the testing set is higher than 98% and only a few of images are classified into the wrong category. This makes our approach perspective for future automatic use. To further improve results, all three models are using transfer learning.

Dokumenty

BibTex


@inproceedings{BUT156730,
  author="Branislav {Hesko}",
  title="Deep convolutional networks for OCT image classification",
  annote="In this work, OCT (optical coherence tomography) images are classified according to the present pathology into four distinct categories. Three different neural network models are used to classify images, each model is recent and we are achieving exceptional results on the testing dataset, which was unknown to the network during the training. Accuracy on the testing set is higher than 98% and only a few of images are classified into the wrong category. This makes our approach perspective for future automatic use. To further improve results, all three models are using transfer learning.",
  address="Vysoké učení technické vBrně, Fakulta elektrotechniky a komunikačních technologií",
  booktitle="Proceedings of the 25th Conference STUDENT EEICT 2019",
  chapter="156730",
  howpublished="online",
  institution="Vysoké učení technické vBrně, Fakulta elektrotechniky a komunikačních technologií",
  year="2019",
  month="april",
  pages="437--442",
  publisher="Vysoké učení technické vBrně, Fakulta elektrotechniky a komunikačních technologií",
  type="conference paper"
}