Computer science researchers help answer an age-old medical science question

Tuesday, July 10, 2018

Why do our veins look blue when blood is red? This is a seemingly elementary science trivia question, and certainly not one that computer science researchers would be expected to be interested in.

But Spencer Van Leeuwen, who recently completed his Master of Mathematics in Computer Science at the University of Waterloo, and his supervisor Professor Gladimir Baranoski, who heads the Natural Phenomena Simulation Group in the David R. Cheriton School of Computer Science, have recently shed light on exactly why veins look blue.

The question is much more complex than it seems, and by tackling it using computer models that simulate organic structures, Van Leeuwen and Baranoski have elucidated the details how light interacts with the cellular and subcellular parts of the skin and blood to clarify how veins get their bluish appearance.

photo of Dr. Gladimir Baranoski and Spencer Van Leeuwen at computer

Professor Gladimir Baranoski (left) and his master's student Spencer Van Leeuwen examine computer-generated swatches depicting the effect of Rayleigh scattering in the papillary dermis on the appearance of a skin specimen without (pink circles) and with (blue circles) a subcutaneous vein.

The monitor on the left shows the typical distinct contrast of skin appearance caused by subcutaneous veins. 

Ultimately, the work that Van Leeuwen and Baranoski have done in clarifying why veins look blue will add to the knowledge base that is needed for technologies that are capable of non-invasive diagnostic testing. There are some instruments today such as devices that measure the amount of oxygen in the blood without pricking the skin, and people have long dreamed of having a Star Trek-like tricorder that can scan for medical problems in a less invasive way. But for those types of instruments to work well, the fundamental data is important to ensure that the results are accurate.

“What we are doing is contributing to the understanding of what is happening inside our tissues,” Baranoski says.

Simulating how materials absorb and propagate light is known as “appearance modelling,” and can be applied to anything from a table, to a flower or to the human skin and blood, Baranoski says. It has long been an aspect of computer graphics used in computer games, but now it has the potential for a much wider range of applications, far beyond entertainment, he adds. “Computer science has to expand. It has strong potential to help in other areas.”

Decades ago, researchers realized that the bluish colour of veins might have something to do with the scattering of light, particularly a type of scattering known as Rayleigh scattering, which is what causes the sky to look blue. Since blue light has shorter, smaller wavelengths, it is more likely to be scattered when it bumps into molecules of the atmosphere, giving the sky its blue appearance. It was strongly suspected that that Rayleigh scattering, named after British physicist John William Strutt, also known as Lord Rayleigh, was also involved in the bluish appearance of veins.

But the question of exactly what is happening at the cellular and subcellular level to cause the scattering eluded researchers. It wasn’t until the mid-1990s that researchers started to look more closely at how the skin and vessels might contribute to the scattering of the light. They found that blue light does not penetrate as deeply into the skin as red and green light, which is more likely to be absorbed within the blood. The blue light therefore has more of a tendency to be scattered away from the blood, giving veins the bluish appearance. This, along with the way humans perceive colour, helped to explain the blue appearance of veins.

But even then, the researchers were not able to pinpoint exactly what structures in the skin were scattering the light and get details of how the scattering worked. This was where Van Leeuwen, working in Baranoski’s group, made a big contribution.

The research involved using two sophisticated modelling tools that Baranoski and his students have developed at Waterloo over the years. One is called HyLIoS (Hyperspectral Light Impingement on Skin) that simulates the interaction of light with skin. The other model, that Van Leeuwen helped improve, is called CLBlood (Cell-Based Light Interaction Model for Human Blood), and it simulates how light is absorbed and propagated by blood flowing in the veins. Both employ computational algorithms known as the Monte Carlo methods, that rely on repeated random sampling to obtain results.

The tools developed at Waterloo are powerful enough to account for many more details in the properties of the skin and blood, such as the way that the distribution of pigment-containing structures known as melanosomes help to scatter the light in the skin and blood, and the location and orientation of cell structures.

Using these tools, it was possible to plug in the parameters for how different types of light coming in at different angles and and simulate how small tissues structures, known as collagen fibrils, scatter light and ultimately affect the colour of the skin and veins. The resulting colour variations are then reproduced using skin swatches generated using techniques employed in computer graphics.

He was then able to validate these models against measurements cited in research literature that describes the physical properties of the skin, vein and blood structures. “If, in the literature, it says that if you reduce the amount of melanin, the skin will appear lighter, we can reduce the melanin and see if the skin appears lighter,” Van Leeuwen says. He was also able to compare the results against how humans see the colours of the skin and veins. “We did every type of validation that we were capable of, given the measurements in the literature.”

Baranoski says using computers, is possible to turn the Rayleigh scattering on or off, for example, and see how the colour changes. It also becomes possible to differentiate between one type of light scattering and another and see the effects.

“These models are tools for us, and you can imagine them as a kind of microscope,” Baranoski says.

Their paper, Elucidating the Contribution of Rayleigh Scattering to the Bluish Appearance of Veins, recently appeared in the Journal of Biomedical Optics.