Brian Chapman, Ph.D.

Download Curriculum Vitae

I began my career in medical imaging, focusing on the physical and mathematical processes that allow us to take a person's body and create a computable abstraction of their anatomy. I worked on vascular imaging and abstracted the complexity of the arterial tree down to a graph of nodes (bifurcations) and edges (arterial segments); a tangle of arteries becomes a set of features you can quantify and compare across populations. It works — I built these methods for intracranial aneurysm, pulmonary embolism, and pulmonary hypertension, and they measure what a radiologist cannot easily measure by hand. It also throws nearly everything away: the blood, the pulse, the person the vasculature belongs to. The throwing away — the abstraction — is what makes the process useful.

This early work was motivated by the philosopher-mathematician Alfred North Whitehead, who wrote in Science and the Modern World:

In a sense, Plato and Pythagoras stand nearer to modern physical science than does Aristotle. … The practical counsel to be derived from Pythagoras, is to measure, and thus to express quality in terms of numerically determined quantity. … Aristotle by his Logic throws the emphasis on classification. … Classification is necessary. But unless you can progress from classification to mathematics, your reasoning will not take you very far.

Most of my career has been spent pushing Whitehead's vision of mathematics-based reasoning into biomedical informatics. Now, some three decades after Whitehead persuaded me to measure rather than classify, I find my research interests bending under the influence of another of his ideas. Abstraction is indispensable, but it runs the risk of being mistaken for the thing itself — a hazard Whitehead called "the fallacy of misplaced concreteness," and which Korzybski put more memorably: the map is not the territory. The medical record is not the person. That is trivially obvious in a face-to-face encounter and not obvious at all in the age of algorithmic medicine, where the record is increasingly the only patient the system ever meets.

A second Whitehead fallacy now seems to me just as consequential. In Modes of Thought he named "the fallacy of the perfect dictionary" — the belief "that mankind has consciously entertained all the fundamental ideas which are applicable to its experience," and that our language already expresses them. Clinical terminologies are dictionaries built on that assumption. Whatever a person cannot say in ICD, SNOMED, or a structured field is invisible to every algorithm downstream; for those systems, it did not happen.

My research now asks where biomedical informatics commits or facilitates these fallacies, what the consequences are for the patient experience, and, most importantly, how we might redesign our systems and the abstractions beneath them to guard against both. The work draws on traditional informatics methods, on current developments in artificial intelligence, and on philosophy — particularly epistemic rights and phenomenology. It also draws on ethnography, and there I mean three things at three stages. What I do now is autoethnography: I treat my own long patient record, and the decades of encounters behind it, as data to be analyzed rather than as anecdote. What guides me is the illness-narrative tradition — Arthur Kleinman's distinction between disease, the practitioner's reformulation, and illness, the life it happens to; and Arthur Frank's argument in The Wounded Storyteller that people do not simply tell stories about illness, they think with them. What I want next is collaboration with medical anthropologists, whose fieldwork reaches what a single person's account, however carefully examined, cannot. I believe ethnography is not a repudiation of measurement. It is how we find out what our abstractions have discarded, and how we enlarge the dictionary.

In addition to research, I am a passionate educator. Most recently I have taught courses in the digital transformation of healthcare, machine learning applications for health, and principles of artificial intelligence for public health. In 2017 I was named Outstanding Educator of Health Sciences Graduate Students at the University of Utah, and in 2018 I served as program chair for the AMIA Informatics Educators Forum in New Orleans.