A glance at xRAPID in action

The year is almost over, and what an eventful year it has been for us! In 2014, we have created a company, validated our models, and advanced a prototype to the point we can now release a product on the market. More details will follow, but we thought it would be a good idea to at least describe where we are headed to for 2015. In particular, we do have two models for xRAPID now:

  • xRAPID Classic: we have refined the app so that it can work on most standard upright microscopes. Even though we were not planning to do this one first, the majority of professionals in the bio-medical field asked for this solution, and we were only too happy to oblige. It consists in a simple attachment to the microscope eyepiece, modular in diameter so that it should be compatible with most existing microscopes.
  • xRAPID Mobile: this was the first version we developed, and it is nearly done (we are closing fast on QA and optimization). Here, a standalone attachment with lens and a ring of LED lights provides the ultimate portable solution to malaria detection. That’s the piece of technology we have mostly featured here, and we would like to show a bit more of « Classic » in this piece.

Both will start shipping in January-February, and you can get in touch with us to pre-order or request further information.

Now let’s have a look at xRAPID Classic. This is the standard setup we are using to produce the picture and the movie shown below. Here, an iPhone 5S is attached to an Olympus BX400 microscope, and we capture the images and perform an analysis live. The magnification on the picture above is 400 times plus electronic zoom (we needed that to show some nice red blood cells for the picture), and 100 times only for xRAPID . We need less magnification for xRAPID as the iPhone camera captures more details than the human eye: we can therefore grab more red blood cells in one single picture, thereby increasing the probability of detection of infected RBCs in each field we measure. For now, let’s look at a few parasites before showing you xRAPID in action.


Using the xRAPID Classic setup we do see some infected red blood cells without too much effort at 400x magnification. The preparation is simple: we stain a blood film using Giemsa, and the malaria parasites appear darker than the red blood cells. Here we clearly see some young rings (center for example, it looks like an « engagement ring »), and a schizont (later stage, darker in the lower left). We have four young rings in the top half. All parasites are of type plasmodium Vivax. We just wanted to show that the infected RBCs are present, and from the picture you can figure out that it’s a quite high concentration. Now let’s see xRAPID in action…

We are presenting an unedited movie here, acquired using Reflector on a Mac, which mirrors the iPhone screen and is used to record the movie. We used the same slide as the previous picture, with a magnification of 100x this time. We can clearly see some late stages of the parasites, with half a dozen red blood cells fully dark blue: those are schizonts or gametocytes. Since we are at a 100x, we cannot see much otherwise — though I promise when you replay the movie a second time, after seeing where the young rings are, you will pick most of them wih the naked eye. Upon running the movie, we catch all the other cells infected: they are clearly quite invisible to the untrained eye, and the detection method built into xRAPID shows them in less than a second.

To keep in the mood of the season, the slide lit up like a Christmas tree.

Finally let’s talk about a few numbers: we characterized about 250 red blood cells in that capture, in one second. Three more pictures, ten more seconds roughly if we are not playing with a movie recorder at the same time, and we have about 1000 RBCs characterized, from which we get a first meaningful statistic of the populations. In practice, in about a minute we have captured 40 fields, for a total number of RBCs larger than 10’000, from which we get the proportion of each stages of development within 0.1% error.