Bring evidence together
Review food, timing, recorded insulin, glucose, context and source provenance as one historical case.
For people living with type 1 diabetes
Make complex meals less of a guess.
Dia-Log is building an evidence-first way to inspect selected past meals, recorded insulin, glucose and context — so you can see whether they are comparable enough to use as personal references.
Explore the fictional historical-case catalogue behind the review model.
Interactive preview
Historical evidence catalogue
Pizza with a pause
Jul 12, 18:30
▲
Sushi (stress)
Jul 15, 19:15
▲
Oatmeal — clean run
Jul 18, 08:05
Open demo →
No sign-in required
When a past personal reference feels questionable
A familiar meal is not always a comparable meal.
Before a recurring difficult meal, an older record can look useful until you inspect whether its evidence and context are still comparable to the situation in front of you.
- A past meal looks familiar, but its timing or context was different.
- Recorded insulin, activity, stress or routine may not line up with the older case.
- You have some numbers, but not a traceable record of everything behind them.
- A reference seems useful until you see what is missing, different or uncertain.
The hard part is not recording another number. It is seeing which past cases can be reviewed together — and which should not.
What Dia-Log is being built to make easier
Inspect which past meal records are comparable enough to review.
Candidate patterns should start with evidence, not an explanation. Dia-Log is being designed to keep selected meal records, timing and context together, while making their differences and limits visible.
Select cases with a reason
Choose past records with relevant similarities, then inspect whether their documented differences still allow comparison.
Keep the limits visible
Keep gaps, corrections, exclusions, contradictions and uncertainty visible instead of filling them with a guess.
The goal is a clearer starting point for reviewing your own meal history, not a treatment conclusion built on incomplete evidence.
Evidence Dia-Log is being built to review
A meal is only one part of a historical case.
These details can support a comparison when they were recorded. If something is missing, it should remain visible rather than being guessed.
- Meal and carbohydrate amount
- Glucose trace
- Timing and recorded insulin use
- Physical activity
- Stress, sleep or illness
- Other recorded context and sources
One missing field can make a comparison ineligible
Source
REC-2026-0712-07
Mobile app export
Field state
Physical activity: missing
Incomplete context remains visible for review
Comparison outcome
No conclusion
The missing source field is not silently filled with a guess
How Dia-Log is being built to work
From historical evidence to a candidate pattern — or no conclusion.
Dia-Log is being designed as a historical-review workflow: selected cases can be inspected together without losing the provenance, differences or limits that make them less comparable.
Open demo- 1
Bring evidence together
Review the meal, glucose, recorded insulin, timing, context and source details as one historical case.
- 2
Select historical cases
Choose past records with relevant similarities to inspect alongside a difficult meal.
- 3
Inspect provenance and context
See what was recorded, where it came from and which details differ between selected cases.
- 4
Check comparison eligibility
Keep missing context, corrections, exclusions and contradictions visible before reviewing cases together.
- 5
Inspect candidate patterns
Review selected comparable history while keeping the supporting evidence and uncertainty attached.
- 6
Retain no conclusion when needed
When the evidence is not sufficient, keep an explicit no-conclusion state instead of creating a rule.
The public demo is a fictional historical-case catalogue. It does not use personal health data or provide personal analysis or treatment guidance.
Why the context matters
A number alone cannot show whether past cases are comparable.
When evidence is hidden
When evidence stays visible
Two meals look similar on the surface.
Timing, insulin and other recorded context stay attached to each case.
Only a glucose result is visible.
You can inspect the meal, source and recorded circumstances behind it.
A missing detail disappears behind a summary.
A missing, conflicting or excluded field remains part of the review.
A familiar result starts to look like a pattern.
Uncertainty stays visible, so the review can retain no conclusion.
What Dia-Log is being built around
Make historical meal evidence easier to inspect.
Bring evidence together
Keep the food, timing, recorded insulin, glucose, context and sources together in one historical case.
Compare with a reason
Inspect whether selected cases are comparable enough to use as personal references — or why they are not.
Keep limits in the picture
Keep differences, gaps, exclusions, contradictions and uncertainty visible throughout the review.
Retain no conclusion
When the evidence is not sufficient, keep no conclusion explicit instead of turning incomplete history into treatment guidance.
Explore the fictional review model and its visible evidence limits.
Open demoFor people living with type 1 diabetes
Help test a historical-review workflow for difficult meals.
Join the wait list if a past personal reference feels questionable before a recurring difficult meal and you want to inspect its evidence and context more closely.
The wait list is for future research and limited pilot opportunities. It does not promise immediate access, personal analysis or medical guidance.
Join wait listStart by exploring the fictional demo catalogue.
Questions, answered
What is Dia-Log building?
An evidence-first historical-review workflow for people with type 1 diabetes. It brings selected past meals, recorded insulin, glucose, context and source details together so users can inspect candidate patterns — or retain no conclusion.
Who is it for?
People living with type 1 diabetes when a past personal reference feels questionable before a recurring difficult meal.
What can I explore in the demo?
A fictional historical-case catalogue that demonstrates the review model. You can browse scripted records and inspect how source details, timing, field state and missing context stay visible. The demo does not use your personal health data.
Does Dia-Log recommend what I should do?
No. Dia-Log does not provide a treatment conclusion, dose, glucose prediction or recommended future action. It is not medical advice and does not replace the guidance of your healthcare team.
What happens when I join the wait list?
You may hear about future research and limited pilot opportunities as we build Dia-Log. Joining does not promise immediate access, personal analysis or medical guidance.
Inspect historical evidence before the next difficult meal.
Explore the fictional review model Dia-Log is building for people with type 1 diabetes, then help shape future research and pilot opportunities.