ESP360° · CRC 2026 presentation site · Philadelphia · October 9–11, 2026Email · WhatsApp

Philadelphia · October 9–11, 2026

ESP360° at Breakthrough T1D CRC 2026

One short oral presentation and two e-posters, accepted through competitive abstract review, with an outstanding-abstract travel grant. The full public presentation material will be gathered here when the congress opens on October 9, 2026.

Scheduled — Abstract #47, short oral. Saturday 10 October 2026 · Short Orals 15, Emerging Diabetes Technologies · Station 05 · session 12:50–1:30 PM · presentation 1:20–1:25 PM. Presenters are required at the station ten minutes before the session opens, so arrival is 12:40 PM.
CRC scientific program checked 30 August 2026; official program is subject to change.

E-poster session times for abstracts #33 and #51 have not yet been published. Program details are subject to change until the congress confirms the final schedule.

Official sources: CRC scientific program · abstract guidelines and embargo.

Abstract #47 · Short oral presentationTemporal Topology in Continuous Glucose Monitoring: Midnight Segmentation Introduces Structural Measurement Artifact
Available October 9Available October 9Audio October 9
Abstract #33 · E-posterConference abstract details available October 9, 2026.
Available October 9Available October 9Audio October 9
Abstract #51 · E-posterConference abstract details available October 9, 2026.
Available October 9Available October 9Audio October 9

#47 — The calendar is not biology

Continuous glucose monitoring produces a physiologically continuous signal, but almost every analytical framework cuts that signal at midnight because the calendar says so. Glucose metabolism follows circadian process, not calendar convention.

Peer-reviewed publication: Diabetes Technology & Therapeutics · PubMed

An 89-day CGM series of 25,536 readings was analysed under three temporal representations: conventional 24-hour linear segmentation, higher-resolution linear segmentation, and a circular representation. Post-acceptance reanalysis showed that the reported entropy advantage for circular topology cannot be supported. Under θ = 2πt/24, 36 equal-width linear bins and 36 equal-width angular sectors are the same partition, so their entropies are necessarily identical. Pooled entropy was approximately 4.54 bits at 24 linear bins and 5.12 bits at both 36 linear and 36 angular bins: a bin-resolution effect of about 12.8%, and a topology effect of approximately zero. What survives is narrower. Correlation between the 23:20–00:00 and 00:00–00:40 windows was about r = 0.09 when the two windows were paired within the same calendar date, and about r = 0.97 when the evening was paired with the morning that physiologically followed it, across 89 nights. That is an indexing and pairing effect, not information created by circular coordinates. A correction has been submitted to the journal.

E-posters #33 and #51. Titles, abstracts, presentation files and audio remain withheld until CRC 2026 opens on October 9, 2026.