PlasmoCount is an online web tool for the automated detection and staging of malaria parasites from Giemsa smears. It takes as input multiple raw images of a Giemsa-stained thin blood film and outputs measures of parasitaemia and parasite life stage development.
Disclaimer: PlasmoCount is currently in beta-testing. It should be tested on a user-to-user basis and not be used for medical diagnosis. To use the server please use the login details of username: giemsa & password: malaria123
Uploading your data
To upload your data, please fill out your email address so that we can send you your results. We currently only support Plasmodium falciparum data, but you can test the tool with your own data for the purpose of cell counting or detection of infection. Our gametocytes tool is currently still in development and we hope to release future versions for other parasite species (watch this space).
For the image upload, please upload images in either .jpg or .png format preferably (otherwise it'll be very slow). PlasmoCount works best with:
Images taken at 100x magnification
< 200 cells in FOV
Low cell density (few overlapping RBCs)
Submitting your data
Every submission will have a unique job ID associated to it. When you submit your job, you will be redirected to your results page. You do not need to stay on this page; you can come back at any time.
Interpreting your results
Summary: The Summary section consists of two pie charts that report on parasitaemia and parasite life stage distribution and an interactive asexual stage distribution histogram.
Life stage distribution histogram: The histogram shows the life stages as computed by the model. The current cut-offs used to count the ring, trophozoite, and schizont stages are 1.5 and 2.5. You can change the bin size to group different cells together. The histogram provides a good opportunity for you to check the model performance. You can click on a bin to display the associated infected RBCs.
Table: The table displays the results for each image individually. You can select an image by clicking on a table row. This will display your image with overlaying predictions.
Export: You can export your results in CSV format by clicking the Export button.
Mobile Device Compatibility
The user interface has been designed with mobile devices in mind. It has been tested with microscopy images taken down the microscope with an iPhone at 2x zoom. After submitting, you can view your results on your desktop by navigating to the unique job link.