Principles of Medicine for Engineers

Part 2: Physiology and its Pathological Perturbations

How your body works, and how it can break

Let’s continue our effort to distill medicine into fundamental (engineering) principles — in Part 1 we set the foundations with what a system was and different levels of understanding the system.

How your heart moves when you’re healthy (left) and one way that it can move when you’re not feeling well (right). Source: https://gfycat.com/adoredunknowndeviltasmanian

In this post I’m going to talk about two of the core concepts in medicine:

• physiology = how the parts of a healthy body work together when you’re feeling well
• patho±physiology = how the parts of a body work when you’re not feeling well

Friendly reminder: this is mostly for myself for now — trying to put down to paper things that are fresh in mind.

A quick preliminary

To talk about the basic principles of patho±physiology, we have to have one quick preliminary about dimensions.

Marvel has been awful for movie-making. Just absolute trash. This was cool though.

Dimensions seem magic — but they’re just another word for variable. And each moving part has its own variable.

If we’ve got two variables, we can plot it over two dimensions. Three variables, three dimensions.

Different variables, or dimensions, can be drawn along different axes. This is relatively easy with three variables.

But what happens when we have four or more variables? It becomes tough to plot it. But we need to find a way to visualize it because even simple organs have more than four moving parts.

One way we can visualize more than 3-dimensions is to re-plot each dimensions along a single axis:

Two views of the same data.

With this way, we see that we can plot 5+ dimensions pretty easily! (Even infinity dimensions…)

Here we’re visualizing five dimensions (left). We can take it to the limit and visualize an infinity of dimensions (right)!

This will become important, later here and in future posts.

Physiology is a beautiful, but flawed, dance

Physiology is the coordination that all the parts of our body have with each other when there isn’t any clear symptoms.

Look at that coordination. Their left-right movement is in opposite directions — so that “dimension” could be negative on our plot to convey an anticorrelation.

This dance looks different in each person, but what it always has in common: we feel good when it happens.

So if we looked at how all the interactions in the body behaved when this group of people were feeling good, we’d notice some conserved patterns.

Certain patterns of interactions show up across different people that all feel healthy.

In this plot, think of each x-axis tick as an “interaction”. That interaction has some weight in each person, some range of weights that it can take on as that person goes about their day.

Beyond that, there’s a set of ranges of weights that a whole group of people can have even as they’re feeling and functioning completely normally.

The pattern of interactions we see from a whole group of people, all feeling healthy. There’s variation, but it’s mostly small. Sometimes there are big variations, but since they’re still associated with healthy, we consider them a different “mode” of physiology.

Of course, changes in certain parts of physiology don’t always lead to feeling fine.

When a step goes wrong…

Every dimension of physiology can be perturbed — sometimes that perturbation is small and the rest of the body can absorb its effect to keep you functioning the same, but sometimes it’s large enough that your body can’t deal with it, and you start feeling off.

That’s what we call patho±physiology.

Sometimes one of the interactions can be far away from what’s associated with health — if these perturbations lead to changes in how the person feels, then that’s considered patho±physiology.

An important principle: Patho±physiology tends to be sparse in that very few things are actually deviating from physiology.

Non-sparse deviations tend to be very lethal (think severe trauma or multi-organ failure) or incompatible with life (think genetic disorders that you never see in children because they completely halt fetal development.

Typically, for a given pathology (disorder/disease), there are a small number of problematic interactions. If there are a lot of them, chances are they’re pretty catastrophic.

Anchor in the output

Maybe the most important point here is that pathology is defined by the output space.

The symptoms, and the patient, decide what’s patho±physiology and what’s not.

That means: just because someone has a perturbation to their physiology does not automatically mean there’s a downstream pathology or disorder.

Principle: If a perturbation hurts functioning or causes suffering, then it’s a patho±physiology that should be treated.

There are plenty of variations that don’t lead to problematic outputs — these are not considered pathological perturbations to physiology. Even if it leads to friction with societal pressures. In those cases the only sustainable intervention may be societal change, not focused intervention in individuals. That’s a bigger discussion for another

Summary

All the parts of the body work together to make you feel good — this is physiology. When one part is out of sync it can potentially cause problems (but not always) — this is patho±physiology.

Pathophysiology tends to be sparse in the parts of physiology that are perturbed and, as a result, likely to be manageable with targeted interventions. Not all perturbations to physiology are considered patho±physiology — some are just healthy variation = physiology.

These basic principles apply to every system in the body, every organ in the body, and indeed every pathology we try to infer and address in medicine.