Documents

  • 112470F

    Final published version, 3 MB, PDF-document

DOI

We present a simple method for the diagnosis of urinary schistosomiasis using an in-line lensless holographic microscope combined with flow cytometry technique. Using simple image processing algorithms and binary image classifier, our system provides automated detection of Schistosoma haematobium eggs in infected urine samples. Registered hologram is reconstructed by applying backpropagation from sensor to sample plane and reconstructed image is automatically analysed for the presence of S. haematobium eggs. Designed for use in a resource-poor laboratory setting, our proposed method has been implemented using a Raspberry Pi computer. From pre-clinical test performed with human urine samples spiked with S. haematobium eggs (approximately 200 eggs per 12 ml of urine), we achieved a sensitivity and specificity of 50.6% and 98.6% respectively. Our proposed method requires no complex sample preparation methods making the system simple to operate and useable in point-of-care diagnosis of urinary schistosomiasis.This method can be optimized to complement existing diagnostic procedures for the detection of S. haematobium eggs and can be deployed to inaccessible remote areas.
Original languageEnglish
Title of host publicationProceedings of SPIE
Subtitle of host publicationOptical Diagnostics and Sensing XX: Toward Point-of-Care Diagnostics
EditorsGerard L. Coté
Place of PublicationBellingham, WA, USA
PublisherSPIE
Number of pages10
Volume11247
ISBN (Electronic)9781510632578
DOIs
Publication statusPublished - 2020
EventSPIE BiOS: Optical Diagnostics and Sensing XX: Toward Point-of-Care Diagnostics - The Moscone Center, San Francisco,California, United States
Duration: 1 Feb 20206 Feb 2020
https://spie.org/conferences-and-exhibitions/photonics-west/bios?SSO=1

Publication series

NameProceedings of SPIE
Volume11247
ISSN (Electronic)0277-786X

Conference

ConferenceSPIE BiOS
CountryUnited States
CitySan Francisco,California
Period1/02/206/02/20
Internet address

ID: 70197758