Research
The refusal to fill a gap is the same refusal in both jobs.
Peer-reviewed clinical work on decision provenance — what happens to a measurement when nobody records the context it was taken in.
This page is here for one reason. If you are deciding whether to hand your family’s history to a stranger, it is worth knowing that the same person spends the rest of his working life arguing, in journals, that data without its context is not evidence.
A blood glucose reading of 4.1 means one thing after a meal and another thing after two hours of shovelling snow, and the number is identical in both cases. Clinical systems record the number and discard the afternoon. My research is about what that costs, and about building a way to keep the afternoon attached to the number.
A family history has exactly the same failure mode. A date without its context is not a memory. It is a number somebody will inherit and be unable to read.
The work
- ESP360° / decision provenance. A framework for recording not just what a clinical decision was, but the state and context in which it was made.
- Peer-reviewed clinical publications. Temporal Discontinuity in Continuous Glucose Monitoring in Diabetes Technology & Therapeutics, and Seventeen Alarms at 3 A.M. in Diabetes Care.
- Two United States provisional patent filings. Clinical decision architecture.
- ESP360° annotated dataset. 536 records, 341 structured variables and 21 human-state annotation layers, pairing physiological signals with decision context.
- Living with type 1 diabetes since 1967. The lived experience underneath the research questions.
Why it belongs on this website
Because the discipline transfers exactly, and because it is checkable. Almost everyone offering to write your family history asks you to trust their sincerity. I would rather you looked at a body of work where being wrong is expensive and being caught is public.
The habit I am arguing for in the journals is the habit on the front page of this site: when the record does not know, the record must say so. Filling a gap smoothly is how you lose the thing you were trying to preserve — in a dataset and in a family.
Recognition
The work has drawn interest across domains outside diabetes care — decision provenance is not a problem unique to medicine. I keep the specifics on LinkedIn rather than listing them here, because a page of one’s own commendations is not evidence of anything and I would rather you read the writing.
Where to start
Send me 250 words about one person.
No fee, no obligation. I will shape what you send into a finished passage, return it, and tell you honestly what your material could become. If the answer is that you do not need me, I will say that too.