Keyword search (3,936 papers available)


A comparative analysis of deep learning architectures on high variation malaria parasite classification dataset.

Author(s): Rahman A, Zunair H, Reme TR, Rahman MS, Mahdy MRC

Malaria, one of the leading causes of death in underdeveloped countries, is primarily diagnosed using microscopy. Computer-aided diagnosis of malaria is a challenging task owing to the fine-grained variability in the appearance of some uninfected and infect...

Article GUID: 33465520
Title:A comparative analysis of deep learning architectures on high variation malaria parasite classification dataset.
Authors:Rahman A,  Zunair H,  Reme TR,  Rahman MS,  Mahdy MRC
Link:https://www.ncbi.nlm.nih.gov/pubmed/33465520
DOI:10.1016/j.tice.2020.101473
Category:Tissue Cell
PMID:33465520
Dept Affiliation: ENCS
1 Department of Electrical & Computer Engineering, North South University, Bashundhara, Dhaka 1229, Bangladesh. Electronic address: aimon.rahman@northsouth.edu.
2 Concordia University, Montreal, QC, Canada. Electronic address: h_zunair@encs.concordia.ca.
3 Department of Electrical & Computer Engineering, North South University, Bashundhara, Dhaka 1229, Bangladesh. Electronic address: rahman.reme@northsouth.edu.
4 Department of Computer Science & Engineering, Bangladesh University of Engineering and Technology ECE Building, West Palasi, Dhaka 1205, Bangladesh.
5 Department of Electrical & Computer Engineering, North South University, Bashundhara, Dhaka 1229, Bangladesh. Electronic address: mahdy.chowdhury@northsouth.edu.

Description:

A comparative analysis of deep learning architectures on high variation malaria parasite classification dataset.

Tissue Cell. 2020 Dec 31; 69:101473

Authors: Rahman A, Zunair H, Reme TR, Rahman MS, Mahdy MRC

Abstract

Malaria, one of the leading causes of death in underdeveloped countries, is primarily diagnosed using microscopy. Computer-aided diagnosis of malaria is a challenging task owing to the fine-grained variability in the appearance of some uninfected and infected class. In this paper, we transform a malaria parasite object detection dataset into a classification dataset, making it the largest malaria classification dataset (63,645 cells), and evaluate the performance of several state-of-the-art deep neural network architectures pretrained on both natural and medical images on this new dataset. We provide detailed insights into the variation of the dataset and qualitative analysis of the results produced by the best models. We also evaluate the models using an independent test set to demonstrate the model's ability to generalize in different domains. Finally, we demonstrate the effect of conditional image synthesis on malaria parasite detection. We provide detailed insights into the influence of synthetic images for the class imbalance problem in the malaria diagnosis context.

PMID: 33465520 [PubMed - as supplied by publisher]