Imperial College London
Portrait Picture

Professor Jake Baum

Faculty of Natural Sciences, Department of Life Sciences

Visiting Professor
 
 
 
//

Contact

 

+44 (0)20 7594 5420jake.baum Website

 
 
//

Location

 

c/o Baum labSir Alexander Fleming BuildingSouth Kensington Campus

//

Summary

 

Publications

Citation

BibTex format

@article{Davidson:2022:10.1017/S2633903X21000015,
author = {Davidson, M and Andradi-Brown, C and Yahiya, S and Chmielewski, J and O'Donnell, A and Gurung, P and Jeninga, M and Prommana, P and Andrew, D and Petter, M and Uthaipibill, C and Boyle, M and Ashdown, G and Dvorin, J and Reece, S and Wilson, D and Cunningham, K and Ando, DM and Dimon, M and Baum, J},
doi = {10.1017/S2633903X21000015},
journal = {Biological Imaging},
pages = {1--13},
title = {Automated detection and staging of malaria parasites from cytological smears using convolutional neural networks},
url = {http://dx.doi.org/10.1017/S2633903X21000015},
volume = {1},
year = {2022}
}
Download

RIS format (EndNote, RefMan)

TY  - JOUR
AB  - Microscopic examination of blood smears remains the gold standard for laboratory inspection and diagnosis of malaria. Smear inspection is, however, time consuming and dependent on trained microscopists with results varying in accuracy. We sought to develop an automated image analysis method to improve accuracy and standardisation of smear inspection that retains capacity for expert confirmation and image archiving. Here we present a machine-learning method that achieves red blood cell (RBC) detection, differentiation between infected/uninfected cells and parasite life stage categorisation from unprocessed, heterogeneous smear images. Based on a pre-trained Faster Region-Based Convolutional Neural Networks (R-CNN) model for RBC detection, our model performs accurately, with average precision of 0.99 at an intersection-over-union threshold of 0.5. Application of a residual neural network (ResNet)-50 model to infected cells also performs accurately, with an area under the receiver operating characteristic curve of 0.98. Lastly, combining our method with a regression model successfully recapitulates intra-erythrocytic developmental cycle with accurate lifecycle stage categorisation. Combined with a mobile-friendly web-based interface, called PlasmoCount, our method permits rapid navigation through and review of results for quality assurance. By standardising assessment of Giemsa smears, our method markedly improves inspection reproducibility and presents a realistic route to both routine lab but also future field-based automated malaria diagnosis.
AU  - Davidson,M
AU  - Andradi-Brown,C
AU  - Yahiya,S
AU  - Chmielewski,J
AU  - O'Donnell,A
AU  - Gurung,P
AU  - Jeninga,M
AU  - Prommana,P
AU  - Andrew,D
AU  - Petter,M
AU  - Uthaipibill,C
AU  - Boyle,M
AU  - Ashdown,G
AU  - Dvorin,J
AU  - Reece,S
AU  - Wilson,D
AU  - Cunningham,K
AU  - Ando,DM
AU  - Dimon,M
AU  - Baum,J
DO  - 10.1017/S2633903X21000015
EP  - 13
PY  - 2022///
SN  - 2633-903X
SP  - 1
TI  - Automated detection and staging of malaria parasites from cytological smears using convolutional neural networks
T2  - Biological Imaging
UR  - http://dx.doi.org/10.1017/S2633903X21000015
UR  - https://www.cambridge.org/core/journals/biological-imaging/article/automated-detection-and-staging-of-malaria-parasites-from-cytological-smears-using-convolutional-neural-networks/8573173B4952D45CA7618E548977EB50
UR  - http://hdl.handle.net/10044/1/90644
VL  - 1
ER  -
Download