A total of 16.5 µl of whole blood stained with antibodies specific for CD4 and CD3 markers is delivered to the flow cell after 8 min, and an image of the same region of the membrane is obtained with two different emission filters.

(A) AlexaFluor-488-conjugated anti-CD4 antibody stains CD4⁺ cells (T lymphocytes and monocytes) green.

(B) AlexaFluor-647-conjugated anti-CD3 antibody stains CD3⁺ T lymphocytes red.

(C) By digitally merging the two images, CD3⁺CD4⁺ T lymphocytes (i.e., “CD4 cells”) appear yellow and are distinguished from CD4⁺CD3^- monocytes (green) and CD3⁺CD4^- T lymphocytes (red).

(D) A lymphocyte selection algorithm is applied to the merged image, based on a lymphocyte profile as defined by size, shape, and uniformity. Objects not fitting the lymphocyte profile are deleted while remaining objects are selected and ultimately counted. A similar protocol to count CD8 cells is used in each participant.

Boxed region indicates two CD4⁺ cells (yellow in [C]) in the original (A and B), merged (C), and processed (D) images. Large green and red objects seen in some images represent aggregates of fluorescent antibody.

From: Rodriguez WR, Christodoulides N, Floriano PN, Graham S, Mohanty S, et al. (2005) A Microchip CD4 Counting Method for HIV Monitoring in Resource-Poor Settings. PLoS Med 2(7): e182