A paramedic kneels next to an injured person by the side of the road. The patient is losing blood and the hospital is far away. The doctor who will take over in fifteen minutes would like to know two things: the patient's blood type and how much blood has already been lost. Today both answers have to wait until a sample reaches a hospital laboratory. A team from National Taiwan University (NTU) has shown that a first clue can come in a minute, from a drop of finger-prick blood, with no electricity and no analyser. All it takes is a plastic chip and a hand-cranked centrifuge.
The work appears in Biosensors and Bioelectronics. The authors are Yuh-Shiuan Chien, Chun-Yen Chang and Ping-Yeh Chiu, who contributed equally, and Prof. Chien-Fu Chen of NTU's Institute of Applied Mechanics, who led the team. It is still a prototype tested on 21 patient samples. It is worth taking apart, though, because the idea is simple and each element solves a specific engineering problem.
The brief: two results from one drop
Blood type and hematocrit are the two parameters at the start of any decision about a transfusion. A mistake in blood type can trigger a severe haemolytic reaction, in which the body destroys the transfused red cells. Hematocrit tells you what share of the blood's volume is taken up by red blood cells. When it falls, it may signal anaemia or bleeding. When it rises, the blood thickens, which happens for example with dehydration.
Hospitals measure both parameters accurately, but they need equipment to do it. Hematocrit comes from automated blood-count analysers that count cells with electrical pulses or laser light. The older method spins blood in a thin capillary and reads the height of the red-cell column. Blood type is determined in gel columns or on plates coated with antibodies. As the authors write, these methods need a stable power supply, trained staff and usually more than a millilitre of blood. In an ambulance, a disaster zone or a field medical post, those conditions are missing.
So the Taiwanese team set itself requirements like a design brief for an engineer. The device has to work without electricity, from a few microlitres of blood, in roughly a minute and without a specialist to interpret the result. And it has to give both results at once.
Part one: a centrifuge with a crank
The drive is a hand-powered centrifuge, which the authors call hpFuge. Turning the crank spins the chip at around 3,000 revolutions per minute. Centrifugal force pushes the blood through the channels, so the chip has no pumps, valves or batteries.
Hand-powered spinning is not a new idea in itself. In 2017 a Stanford University team described in Nature Biomedical Engineering a paper centrifuge modelled on a children's toy spun with a string. That device separated blood, but did not determine blood type. What is new in the NTU work is what happens inside the chip.
A hand centrifuge has one drawback: it produces a weaker and less even force than a laboratory machine. The authors compensated for this with the shape of the channels, as described below.
Part two: a winding channel with reagent
The chip has three channels, one each for the A, B and Rh antigens. Each holds a reagent with antibodies. If the red cells carry the matching antigen, the antibodies stick them together into clumps. Laboratory staff have been watching this same reaction, agglutination, on glass slides and in test tubes for more than a hundred years.
To give the cells time to mix with the reagent and clump, the channel winds like a serpentine. This keeps the blood in it longer, even though the hand centrifuge pushes it only gently. The team tested how channel length, bend radius and depth affect mixing, steady flow and how readable the result is. These trials produced the design parameters at which the reaction repeats reliably from sample to sample.
Part three: a slit that catches clumps
The key element is a micro-slit at the end of each channel. It is about 7.5 micrometres wide, roughly a tenth of the thickness of a human hair. The authors formed it by chemically modifying the surface of the channel.
The slit works like a sieve with carefully chosen mesh. Single red blood cells squeeze through it and flow on. Clumps of agglutinated cells are too large, so they stop in front of the slit and form a clear red band. If a band appears in the A channel, the blood carries antigen A. If it appears in the Rh channel, the blood is Rh positive. The result is visible to the naked eye, with no reader and no smartphone app.
Part four: a ruler printed on the chip
The same band gives the second result. The more red cells in the blood, the longer the column of clumps in front of the slit. A scale is printed on the chip's cover, so the band's length is read like a ruler and converted to hematocrit.
It is a clever combination. The reaction that settles the blood type is also a measure of the amount of red cells. In a conventional laboratory these are two separate tests on two separate machines.
Stress test: what the trials showed
First the team checked whether band length really tracks hematocrit. Between 10 and 60 percent the relationship turned out to be linear, with a goodness of fit R² of 0.99. That range covers both severe anaemia and markedly thickened blood.
Next came 21 samples from hospital patients. The chip's results were compared with automated analysers. The mean difference in hematocrit was 0.126 percentage points. In 95 percent of cases the difference fell between minus 0.63 and plus 0.88 points. According to the university's press release, the chip's blood types also matched the hospital results. The reading did not depend on how fast someone turned the crank, which is crucial for a hand-powered drive. The whole test took about a minute and used 3 microlitres of blood in total.
"Our goal is to bring two essential blood tests out of the laboratory and into the hands of first responders, with no electricity or specialized training needed," says Prof. Chen in the NTU press release.
Where the design still has slack
An engineer also looks at what has not been tested. Twenty-one samples are a good first test, but too few to judge rare cases. There are more than 30 blood group systems and more than 300 antigens on red cells. The chip checks the three most important: A, B and Rh. The available abstract and press release do not say whether difficult samples were tested, such as weak variants of the D antigen or newborns' blood.
It is also unknown how the chip will cope with the conditions it is meant for. Paramedics use equipment in rain, frost and heat, often wearing gloves and under time pressure. The antibodies in the reagents must survive months in storage and in a medical bag. Reading a band by eye is simple, but in poor light it is easy to be off by a few scale marks. Only field trials, which have not yet taken place, can settle these questions.
That is why we rate this technology at TRL 4: a working prototype tested in the laboratory on real clinical samples, but with no ambulance trials, no mass production and no market approval.
It also has to be said plainly what the chip will not replace. Before a transfusion, the hospital still confirms the blood type in a serology laboratory and checks the compatibility of donor and recipient blood. Until the type is known, emergency cases receive group O Rh-negative blood, which can usually be given to anyone. A quick field test will not change these rules. It could, however, give the doctor a hint about what they are dealing with before the patient reaches hospital, and a hematocrit reading helps estimate blood loss.
Others have tried too
The Taiwanese team is not the first to try to take these tests out of the laboratory. In 2020 researchers at the University of Cincinnati described a strip test for hematocrit, in which blood soaks into paper and a smartphone app reads the length of the red streak. In 2021 a University of Toronto team showed in Clinical Chemistry a platform that determined blood type, donor compatibility and hematocrit on a single chip. That system, however, moved droplets with electrodes, so it needed power.
Against this background, the NTU chip stands out for its combination of features. It has no electronics, needs no phone, gives two results at once, and the answer appears in a minute. That set of features suits a paramedic's bag better than a hospital laboratory.
The Polish perspective: ambulances, the military and regulation
For Poland, the most obvious applications are emergency medical services and the military. Ambulance crews carry equipment that has to be light, rugged and independent of power. Battlefield medicine has for years been looking for ways to transfuse blood as close as possible to the point of injury. Add to that disasters, floods and places far from the nearest laboratory.
The road to the ambulance, however, runs through regulation. In the European Union, in vitro diagnostic devices are governed by Regulation 2017/746, known as the IVDR. Tests that determine ABO and Rh blood groups fall into the highest risk class, D. That means assessment by a notified body and the involvement of EU reference laboratories. For a manufacturer this means years of work and clinical studies. For the user it is a guarantee that the test will not mix up blood groups more often than the standards allow.
The chip itself, on the other hand, is cheap to make. It is a piece of plastic with channels, reagent and printing. Polish companies and academic centres working on microfluidics and diagnostics could find a role here as component suppliers or partners in validation studies. The first step, though, belongs to the authors, who must show that the prototype works outside the laboratory.
What comes next
The next stages are predictable: more samples, including difficult ones, a reagent shelf-life test, trials in real ambulances and medical posts, and then certification. Each of these stages may expose a weakness that 21 samples in a laboratory do not reveal.
Even so, the design holds up through its simplicity. The crank replaces a motor, the winding channel replaces a mixer, the slit replaces a reader, and the printed scale replaces an analyser. If the prototype passes field trials, a paramedic will get two results from a drop of blood before the ambulance leaves the scene of an accident.
Bibliography
- Chien Y.-S., Chang C.-Y., Chiu P.-Y., Chen C.-F. Hand-powered micro-slit–gated microfluidics for concurrent blood typing and hematocrit estimation. Biosensors and Bioelectronics 315, 119215 (2027). DOI: 10.1016/j.bios.2026.119215
- Bhamla M. S., Benson B., Chai C., Katsikis G., Johri A., Prakash M. Hand-powered ultralow-cost paper centrifuge. Nature Biomedical Engineering 1, 0009 (2017). DOI: 10.1038/s41551-016-0009
- Frantz E., Li H., Steckl A. J. Quantitative hematocrit measurement of whole blood in a point-of-care lateral flow device using a smartphone flow tracking app. Biosensors and Bioelectronics 163, 112300 (2020). DOI: 10.1016/j.bios.2020.112300
- Sklavounos A. A., Lamanna J., Modi D., Gupta S., Mariakakis A., Callum J., Wheeler A. R. Digital Microfluidic Hemagglutination Assays for Blood Typing, Donor Compatibility Testing, and Hematocrit Analysis. Clinical Chemistry 67, 1699–1708 (2021). DOI: 10.1093/clinchem/hvab180
- Goodnough L. T., Levy J. H., Murphy M. F. Concepts of blood transfusion in adults. The Lancet 381, 1845–1854 (2013). DOI: [10.1016/S0140-6736(13)60650-9](https://doi.org/10.1016/S0140-6736(13)60650-9)
- Regulation (EU) 2017/746 of the European Parliament and of the Council on in vitro diagnostic medical devices (IVDR). EUR-Lex
- National Taiwan University. Hand-cranked device tells blood type and red cell levels in 1 minute. Phys.org (17.10.2026). Phys.org
Comments· 0
No comments yet. Be the first.