An automated parasite infection diagnostic system via open-source microscopy and deep learning (Conference Presentation)

Yaning Li; Rui Zheng; Yizhen Wu; Kaiqin Chu; Qianming Xu; Mingzhai Sun; Zachary J. Smith

doi:10.1117/12.2506571

4 March 2019 An automated parasite infection diagnostic system via open-source microscopy and deep learning (Conference Presentation)

Yaning Li, Rui Zheng, Yizhen Wu, Kaiqin Chu, Qianming Xu, Mingzhai Sun, Zachary J. Smith

Author Affiliations +

Proceedings Volume 10869, Optics and Biophotonics in Low-Resource Settings V; 108690G (2019) https://doi.org/10.1117/12.2506571
Event: SPIE BiOS, 2019, San Francisco, California, United States

Abstract

We report the development of a cost-effective, automated parasite diagnostic system that does not require special sample preparation or a trained user. It is composed of a cost-effective, portable microscope that can automatically auto-focus and scan over the size of an entire McMaster chamber (100 mm2) and capture high resolution (~1 µm) bright field images without need for user intervention. Fecal samples prepared using the McMaster flotation method were imaged, with the imaging region comprising the entire McMaster chamber. A convolutional neural network (CNN) automatically segments and analyzes the images to robustly separate eggs from background debris. The performance of the CNN is high despite the challenging, unbalanced nature of the images, where >95% of images contain no eggs and thus the potential for false-positives is high. Simple post-processing of the CNN output yields both egg species and egg counts. The system was validated by comparing hand counts with automated counts of samples containing eggs from ascarid, strongyle, and Trichuris nematodes, along with Eimeria oocysts. The system shows excellent performance, even on challenging Eimeria parasites whose small size is similar to fecal debris. The R2 values between hand and automated counts are >0.95 for both Eimeria and nematode parasites. Further, the diagnostic accuracy of our system for recommending antibiotic treatment is 100% for nematode parasites and 96% for Eimeria. As a further demonstration of utility, the system was used to conveniently quantify drug response over time, showing residual disease due to antibiotic resistance after 2 weeks.

Conference Presentation

Citation Download Citation

Yaning Li, Rui Zheng, Yizhen Wu, Kaiqin Chu, Qianming Xu, Mingzhai Sun, and Zachary J. Smith "An automated parasite infection diagnostic system via open-source microscopy and deep learning (Conference Presentation)", Proc. SPIE 10869, Optics and Biophotonics in Low-Resource Settings V, 108690G (4 March 2019); https://doi.org/10.1117/12.2506571

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
PRESENTATION

WATCH
PRESENTATION SAVE TO MY LIBRARY

GET CITATION

KEYWORDS

Diagnostics

Microscopy

Image segmentation

Convolutional neural networks

Image analysis

Image resolution

Microscopes

Show All Keywords

Keywords/Phrases

Search In:

Publication Years