tom/HealthLog
1
0 Fork 0
mirror of https://github.com/MBombeck/HealthLog.git synced 2026-11-10 05:12:14 +00:00
Code Issues Packages Projects Releases Wiki Activity

Derive glucose isSpotEstimate from reading density

Browse Source 
Hard-coding isSpotEstimate: true mislabels a dense Nightscout CGM stream
(~288 readings/day) as a spot estimate and undersells the data in the
clinical panel and coach. Compute readings-per-day over the covered span
and treat the series as continuous below the spot caveat once it clears a
~hourly bar (24 readings/day); sparse spot fingersticks keep the caveat.
The panel only renders the spot-reading line when isSpotEstimate, and the
OpenAPI field relaxes from a literal-true to a boolean.
This commit is contained in:
Marc-André Bombeck committed 2026-06-14 20:54:32 +02:00
1 parent c5a7e06e35
commit c268ace58c
6 files changed
+98 -31
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

src/lib/openapi/routes/insights.ts
+2 -2
View File
@@ -587,12 +587,12 @@ const glucoseClinicalSchema = z
hbgi: z.number(),
})
.nullable(),
isSpotEstimate: z.literal(true),
isSpotEstimate: z.boolean(),
})
.meta({
id: "GlucoseClinicalMetrics",
description:
"Server-authoritative glucose clinical panel over the trailing 30-day window. Every figure is a SPOT-READING ESTIMATE (a % of readings), not a CGM time-in-range AGP — `isSpotEstimate` is always true and `stillLearning` gates assertion when the window has too few readings or too short a span. `distribution` carries the Battelino 2019 TIR/TBR/TAR fractions (level-2 nested in level-1) plus minutes-of-a-day equivalents; `gmi` (Bergenstal 2018) + `estimatedA1c` (Nathan 2008 ADAG) derive from the mean; `variability` is SD + CV% with the Monnier 2017 ≥36% instability flag; `advanced` is the disclosure tier — J-index (Wojcicki 1995) + LBGI/HBGI (Kovatchev hypo/hyper risk). All blocks are null when there are no readings; `advanced.jIndex` is null for a single-reading window.",
"Server-authoritative glucose clinical panel over the trailing 30-day window. Figures from sparse spot data are a SPOT-READING ESTIMATE (a % of readings), not a CGM time-in-range AGP; `isSpotEstimate` is derived from reading density (true below ~hourly, false for a continuous CGM stream such as Nightscout) and `stillLearning` gates assertion when the window has too few readings or too short a span. `distribution` carries the Battelino 2019 TIR/TBR/TAR fractions (level-2 nested in level-1) plus minutes-of-a-day equivalents; `gmi` (Bergenstal 2018) + `estimatedA1c` (Nathan 2008 ADAG) derive from the mean; `variability` is SD + CV% with the Monnier 2017 ≥36% instability flag; `advanced` is the disclosure tier — J-index (Wojcicki 1995) + LBGI/HBGI (Kovatchev hypo/hyper risk). All blocks are null when there are no readings; `advanced.jIndex` is null for a single-reading window.",
});
const dashboardSnapshotResponse = z