The missing layer is not another sensor.
It is the reasoning that sits between a physiological signal and the action taken in response — and it is almost never recorded.
Mark E. Paull is an independent researcher and decision systems architect working at the intersection of clinical artificial intelligence, behavioral decision science and narrative medicine. The question underneath all of it is the same one the memoir asks: when a system learns from outcomes, what happens to the human reasoning that produced them?
ESP360° Decision-Aware CGM Dataset
A novel annotated dataset pairing continuous glucose signals with real-time cognitive decision context: one subject, 68 consecutive days without gaps, 536 records, 468 decision events, 340 structured variables and 21 human-state annotation layers. Proof of concept, not validation. It captures behavioral context that standard CGM training datasets do not represent.
Decision Shadow / Decision Provenance
Formalizes how prior clinical decisions generate persistent interpretive bias in AI-assisted diagnosis, with measurable properties — depth, overlap, distortion — applicable across CGM, otolaryngology, ophthalmology and anaesthesia.
The Missing Constraint Layer
Identifies the systematic absence of behavioral and cognitive context as a structural flaw in medical machine-learning pipelines, with direct implications for patient safety when clinician judgment is iteratively encoded as error.
- Peer reviewer
- Diabetes Care, American Diabetes Association · Clinical Diabetes
- Published author
- Diabetes Care (2026) · Diabetes Technology & Therapeutics (2026) · STAT News (2025)
- CRC 2026
- Three abstracts accepted through competitive review for the Breakthrough T1D Clinical & Research Congress, Philadelphia, 9–11 October 2026: one short oral presentation and two e-posters
- Continuing medical education
- 7.75 CME hours completed in 2025, including diabetes technology and CGM education
- Affiliation
- None. Independent.
- ORCID
- 0009-0002-7658-0737
