AUDIO MESSAGE

Send me an audio message.

Have a question, reaction or idea? Record up to 3 minutes. Nothing leaves your device until you tick the box and tap Send recording — and it then goes straight to Mark's own storage, never through a third-party service.

0:00 / 3:00

Your recording stays on your device until you tap Send recording or Share another way.

Breakthrough T1D Clinical & Research Congress · 2026

Clinical data records what happened. ESP360° preserves why the decision happened.

Name: ESP360° longitudinal decision-provenance proof-of-concept dataset
Creator: Mark E. Paull

A structured decision-provenance architecture built to represent the human reasoning, uncertainty, context and timing that conventional clinical telemetry leaves out.

Presented by Mark E. Paull at CRC 2026 in Philadelphia: one short oral presentation and two e-posters, all accepted through competitive abstract review. This site is the public, conference-safe presentation layer. Reconstructive engineering detail remains private.

Enter CRC 2026 Explain it simply

Mark E. PaullIndependent Researcher · Decision Systems Architect · 60th year with type 1 diabetes

536annotated records

468logged events

341structured columns

21annotation layers

One subject, 68 consecutive days — 12 October to 18 December 2025 (CGM on every day; decision context on 44 days). Proof of concept, not validation.

Verified post-submission dataset figures are 341 columns across 68 consecutive days.

CRC 2026 · THREE ACCEPTED PRESENTATIONS

EXPLORE THE THREE CRC 2026 PRESENTATIONS

1 ORAL · 2 E-POSTERS · ACCEPTED THROUGH COMPETITIVE ABSTRACT REVIEW

BREAKTHROUGH T1D CLINICAL & RESEARCH CONGRESS · PHILADELPHIA · OCTOBER 9–11, 2026

THREE CRC 2026 PRESENTATIONS

Three presentations, given in Philadelphia. Slides, posters and audio are below; the peer-reviewed #47 paper is also public.

ORAL · #47

The Calendar Is Not Biology.

Temporal Topology in Continuous Glucose Monitoring

Published result — Diabetes Technology & Therapeutics
An 89-day CGM series of 25,536 readings. The originally published 12.1% entropy advantage for circular topology has been withdrawn: reanalysis showed that 36 linear and 36 angular bins are the same partition under θ = 2πt/24. The surviving result is cross-midnight correlation of r ≈ 0.09 under same-calendar-date pairing versus r ≈ 0.97 under physiologically sequential pairing, n = 89. A correction is with the journal.
This is the peer-reviewed temporal-topology study, a separate dataset from the ESP360° annotated workbook of 68 consecutive days shown above. PubMed

Slides (PDF) Download slides Audio (MP3)

E-POSTER · #33

Decision-Aware Insulin Therapy.

A 360-Variable, 21-Layer Decision Manifold for Capturing Moment-of-Action Clinical Reasoning

Presented e-poster.
Poster available below.

Poster (PDF) Download poster

E-POSTER · #51

Selective Ground Truth in Adaptive Medical AI.

Preventing Expert Drift in PCCP-Governed Systems

Presented e-poster.
Poster available below.

Poster (PDF) Download poster Audio (MP3)

ESP360° CONFERENCE SITE

Add the ESP360° conference site to your phone.

Direct access to CRC 2026 presentation information and public research material.

Tap the button to install ESP360° on this device.

BREAKTHROUGH T1D · PHILADELPHIA · OCTOBER 9–11, 2026 CRC 2026 RESOURCES Open the conference materials ↓

Handout (PDF)

ORAL · #47Temporal Topology in Continuous Glucose Monitoring

Slides (PDF)Download slides Audio (MP3)

E-POSTER · #33Decision-Aware Insulin Therapy: A 360-Variable, 21-Layer Decision Manifold for Capturing Moment-of-Action Clinical Reasoning

Poster (PDF)Download poster

E-POSTER · #51Selective Ground Truth in Adaptive Medical AI: Preventing Expert Drift in PCCP-Governed Systems

Poster (PDF)Download poster Audio (MP3)

Open the full CRC 2026 Resources page →

The longitudinal record begins before digital diabetes

1968: one year after diagnosis.

A Canadian Diabetes Association newsletter clipping documents the beginning of the lived record behind the later ESP360° architecture. The research argument is not that longevity proves the model; it explains the unusually long observational substrate from which the decision-layer problem emerged.

See the research lineage →

Choose your depth

One idea. Different explanations for different readers.

The front page stays simple. The deeper material is separated by audience so the site does not try to say everything at once.

Plain language

What ESP360° means without technical vocabulary, and why a glucose number is not the same thing as the decision that produced it.

Read → 02

For clinicians

What the architecture adds to CGM, insulin and outcome review without replacing clinical judgment or making autonomous recommendations.

Read → 03

Technical / research

Study scope, record structure, variable classification, verification status and the limits of what a single-subject dataset can establish.

Read → 04

For engineers & AI teams

The six architectural domains, where decision provenance sits beside existing telemetry, and what is deliberately withheld.

Read →

The core distinction

Inference is not provenance.

A model may infer why a person acted. ESP360° is designed to preserve what the person actually knew, expected, feared, weighed and decided before the outcome was known.