HEp-Net: a smaller and better deep-learning network for HEp-2 cell classification

Yuexiang Li, Linlin Shen

Research output: Journal PublicationArticlepeer-review

17 Citations (Scopus)


Indirect immunofluorescence of Human Epithelial-2 (HEp-2) cells is a commonly used method for the diagnosis of autoimmune diseases. Traditional approach relies on specialists to observe HEp-2 slides via the fluorescence microscope, which suffers from a number of shortcomings like being subjective and labour intensive. In this paper, we proposed a deep-learning network, namely HEp-Net, to automatically classify HEp-2 cell images. The proposed HEp-Net uses multi-scale convolutional component to extract features from Hep-2 cell images and fuses the features extracted by shallow and deep layers for performance improvement. The proposed model is evaluated on publicly available I3A (Indirect Immunofluorescence Image Analysis) and MIVIA data-sets. Experimental result demonstrates that, compared to the state-of-the-art approaches, our proposed HEp-Net yields better performance with smaller network size.

Original languageEnglish
Pages (from-to)266-272
Number of pages7
JournalComputer Methods in Biomechanics and Biomedical Engineering: Imaging and Visualization
Issue number3
Publication statusPublished - 4 May 2019
Externally publishedYes


  • deep-learning network
  • HEp-2 cells
  • image classification

ASJC Scopus subject areas

  • Computational Mechanics
  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging
  • Computer Science Applications


Dive into the research topics of 'HEp-Net: a smaller and better deep-learning network for HEp-2 cell classification'. Together they form a unique fingerprint.

Cite this