Engineering the Heart
“No man-made structure is designed like a heart. Considering the highly sophisticated engineering evidenced in the heart, it is not surprising that our understanding of it comes so slowly.” - Daniel. D. Streeter Jr.
Throughout history, scientists have always wondered about the heart. We have dissected it and probed it, shocked it and sliced it into thin sections to peer at under a microscope. Everything we have tried has taught us something new, but nothing has taught us enough to fully understand the whole-scale functioning of the organ itself. Now, cardiovascular diseases account for ~30% of all human deaths worldwide, edging out parasites and infectious disease by ~7%; sedentary lifestyles and massive meals have certainly left developed countries worse for wear in the heart department. In the face of cardiovascular crisis, scientists are using wholly new approaches to try and tease something new out of the heart again - something that could be radically useful in clinical practise.
Somewhere To Start
Successful physiological analysis of any organ system requires understanding of the key functional relationships between the components: Cells (the parts), organs (the whole) and networks (signaling cascades, charge propagation, etc. - the method of communication between the parts to help form the whole functioning organ structure), and how these components change in a disease state. Some of this information may reside in the genome, but GWAS studies have identified a relatively low percentage of explained heritability in cardiovascular disease; some may also reside in the proteins the genes encode for, but to me that’s too simplistic an explanation. Rather, many scientists believe that it is the interactions between proteins - within the context of subcellular, cellular, tissue, organ, and whole-human-body-system structures - that drive the disease state.
In order to understand these complex interactions, and to attempt to make sense of the vast amount of data being generated through experimentation, physiologists have begun to explore these protein interactions in the heart in a quantitative and computational manner, with the long-term goal of acheiving patient-by-patient personalised medicine with their algorithms. Starting from tiny ions and scaling up to model the entire human heart with startling accuracy across thousands of processors, systems physiology is translatable, innovative, and exciting.
Building From the Bottom Up
In reductionist biology, to model an organ means to first model its constituent parts. The cells of the heart - cardiomyocytes - are highly specialised, and are able to sustain electrical propagation from the Sinoatrial Node (SA) to the Atrioventricular Node (AV), while baseline physiology facilitates further propagation to the ventricle and Purkinje neuron fibres, thus causing the timed contraction necessary to pump blood through the system. Scientists were precisely able to model both the action potential in each individual myocardial cell and that action potential’s propogation between cells, forming the foundation for further research into cardiac arrythmia and providing a wholly different “dissection of the heart.”
Cardiac excitation, very generally, involves both the generation of an action potential by individual cardiac cells and the propagation of that action potential through intracellular gap junctions. In the equation below, the derivative of the transmembrane potential with respect to time can be related to the capacitance multiplied by the total transmembrane ionic current:
The equation above states simply that changes in the transmembrane potential - the difference in charge across the heart cell membrane - occur due to the movement of ions across that membrane. This occurs via a variety of pumps and channels, primarily transporting sodium, calcium, and potassium. The action potential is generated by the selective opening and closing of these channels, which brings the membrane potential from negative up to sharply positive.
The simplest model for the propagation of the action potential relies upon a continuous chain of excitable elements. In this chain - imagine a series of cardiomyocytes - the current will flow from a depolarised cell to its less depolarised neighbours via intercellular resistive gap junctions. Cells in a system are always more complicated than cells in a clamp changing the charge on the membrane capacitance, and therefore a slightly more complicated equation is needed to model the propagation of the action potential between excitable heart cells.
This is a common reaction-diffusion equation, here used to relate the transmembrane potential in a single cell to the axial current that flows between cells in a linear chain. The left side of the second equation gives the total transmembrane current, and the right side provides computes the net gain or loss of the axial current as it flows through the system. Basically, it states conservation: The net change in axial current must be accounted for by the current that crosses the cell membrane.
Those of you keener at maths might realise that the second equation presented can be very simply thought of as an extension of the first equation. As shown above, the transmembrane potential is dependent upon both time and space, but in our friendly single-cell example, the transmembrane potential is independent of space and thus if only one cell is present in the chain the second propagation equation reduces to the first.
And thus, we have a propagating current and our building blocks of the heart.
Creating the Heartbeat
Of course, the heart is not a perfect sphere - it has an architecture, a shape. The activation sequence of different parts of the heart has been found to be strongly influenced by the fibrous-sheet architecture of the myocardium, causing non-uniform excitation of those connected excitable heart cells and the boom-boom sound you hear when your heart beats. Simply, different parts of the heart are activated at different times.
Supercomputing the Heart
Modeling the whole-organ system is the most complicated task, requiring data input from multiple levels of complexity and multiple scales, from the ions present inside each myocardial cell to the whole organism. The Alya Red project is trying to use 10,000 processors to model a whole heart using similar* mathematics to those I have outlined briefly here.
Most striking from the video above is the precise geometry and organisation of the muscle fibres designed to propagate the axial current, as well as the diffusion tensor imaging used to validate the model. The scientists behind Ayla Red hope that it will be useful for medical professionals to understand the human body better, diagnose pathologies and even plan surgeries.
Impact on Modern Medicine: Does All This Maths Add Up?
Well, does it? That’s the big question, of course - are all these hours of simulating and 10,000 processors worth hard, cold net life years to patients? While these systems are beautiful and fascinating, the primary goal in the study of human organs must always be clinical. The euHeart Project thinks that these models can be easily adapted into ‘personalised algorithms’ by taking precise biophysical measurements from patients and running code to simulate their individual heart. Models of the heart can connect the dots surrounding pathophysiology; they can allow doctors to peer into underlying causes in a way they couldn’t before without extensive surgery, and assist treatment decisions. However, there’s currently a “significant translational barrier” - models are based too much on data in animal models and controlled conditions, and not enough on real-life human data, which is significantly harder to come by.
The euHeart Project will, over the next four years, attempt to modify existing frameworks so they’re useful in a clinical setting; they hope to allow them to rely on non-invasive measurements and simulate the human heart accurately. Will that ever happen, or is all this modeling a waste of time? Personally, I think it’s feasible; we have a long way to go, but never say never.
Science is all about seeing something that hasn’t been seen before, or figuring something out that connects the dots. While this new methodology in physiology isn’t perfect, it’s certainly a different way of looking at the human heart and valuable research about arrhythmia and cardiomyopathies has already come out of it.
*Similar is a strong word. Try vaguely connected; they’re at the organ level and I primarily spoke about the cellular level. Plus their mathematics will be loads more complex than my simple example.
All figures were taken from references cited in-line via links or made myself. Referencing is done in-line as well via links; special mentions go to Science, vol. 295 no. 5560 “Modeling the heart — from genes to cells to the whole organ” and Physiology Review, vol. 84 “Basic Mechanisms of Cardiac Impulse Propagation and Associated Arrhythmia.”
