← BACK halfmarble halfmarble 02 // PK
SteadyDoseTimer — Pharmacokinetic Simulation
Bateman model: C(t) = A(e-ke·t - e-ka·t) — click chart to set NOW, then add doses
FOR EDUCATIONAL AND RESEARCH PURPOSES ONLY. This simulation is not medical advice. It does not account for individual physiology, drug interactions, or clinical context. Always consult your physician before making any changes to your medication schedule.
2:00 PM NOW — drag on chart to move
Activity Demand — click/drag to paint
DYSKINESIA TOLERANCE
Eating Windows — doses during meals use reduced absorption
Meal 1 3:00 PM – 4:00 PM
Meal 2 12:00 PM – 1:00 PM
Meal 3 8:00 AM – 9:00 AM
DOSES TODAY 0
CURRENT LEVEL 0%
TIME IN RANGE
NEXT DOSE IN
VARIABILITY ±50%
Confidence band (population variability in absorption & elimination)
Therapeutic window (effective range without dyskinesia)
Expected blood concentration (Bateman equation)
Fasting dose — auto-detected from eating windows, faster absorption
Circadian zones — morning peak (best absorption), afternoon dip (wearing off risk), wind-down
Activity demand — modulates therapeutic window (higher demand = higher coverage needed)
Eating window — reduced levodopa absorption (protein competes for LNAA transport)
Glass Box // what the AI reads

No black box. Every paper and data source behind the pharmacokinetic model is listed here. You see what the math sees — nothing hidden, nothing invented. The source code will be released for independent review.

ANCHORED The PK model is pinned to published pharmacokinetic parameters — not a drifting LLM. The math cannot silently change between updates.
ON-DEVICE The model lives on your phone. No cloud. No server. No API calls. Your medication data never leaves your device.
LOCAL Every dose log and subjective report stays on your phone. We cannot read it. We cannot sell it. It does not exist anywhere else.
TRACEABLE Every parameter in the Bateman equation traces to a published source below. If it's not in the Glass Box, the model won't use it.
Pharmacokinetic Model Sources
01 Pharmacokinetics of levodopa Nutt & Fellman — Movement Disorders, 1993
02 IPX066 (Rytary): pharmacokinetics of extended-release carbidopa-levodopa Hauser et al. — Neurology, 2013
03 Continuous dopaminergic stimulation in Parkinson's disease Olanow et al. — The Lancet Neurology, 2006
04 Population pharmacokinetics of levodopa in Parkinson's disease patients Clinical Pharmacokinetics, 2020
05 Sinemet CR pharmacokinetics: bioavailability and absorption characteristics Movement Disorders, 2014
06 Effect of food on the pharmacokinetics of levodopa formulations Parkinsonism & Related Disorders, 2016
Clinical Framework
07 Low-fat versus ketogenic diet in Parkinson's disease: A pilot randomized controlled trial Phillips et al. — Movement Disorders, 2018
08 Wearing off and motor complications in Parkinson's disease Stocchi et al. — Neurology, 2010
09 Motor fluctuations and dyskinesia in Parkinson's disease: clinical guide and practical tips Vijayakumar & Jankovic — The Lancet Neurology, 2018
10 Bateman, H. — The solution of a system of differential equations in pharmacokinetics Proc. Cambridge Phil. Soc., 1910
Simulation Parameters Derived From
Sinemet IR: t½ 90 min, Tmax 30-60 min, F=99% — from FDA prescribing label + Nutt & Fellman (1993)
Sinemet CR: t½ 90 min, Tmax 120 min, F=71% — from FDA prescribing label + controlled-release PK studies
Rytary ER: dual-phase absorption (60% fast / 40% slow), Tmax 1h/4h — from Hauser et al. (2013) + FDA label
Food effect: high-protein meals delay Tmax by ~30 min, reduce Cmax — from competitive amino acid transport literature
GLASS_BOX_POLICY: Every parameter in this simulation traces to a published pharmacokinetic source. The Bateman equation is the standard one-compartment oral absorption model used in clinical pharmacology. No proprietary algorithms. No hidden tuning. No synthetic data.