Synthetic data only. Never enter patient information; nothing is sent or stored. For clinicians. Not FDA cleared.For licensed clinicians. Educational demonstration on synthetic data. Never enter real patient identifiers or protected health information: everything runs in your browser and nothing is sent or stored. Clinical decision support only; not FDA cleared; physician review is required.
CPE-001 · version 1.1.0
Thyroid Pattern Engine: test how TSH and FT4 become a clinical pattern
Functional state from TSH and FT4, kept separate from etiology. Central disease and decompensation never hide.
The Thyroid Pattern Engine reads TSH and FT4 against the laboratory's own reference interval, decides which functional pattern fits, and then asks a second, separate question: what context could explain it.
That separation is the point. Anti-TPO positivity never makes someone hypothyroid, a normal TSH never excludes central disease, and a low FT4 in a patient with a pituitary history triggers an adrenal-safety alert before anything else.
Playground
Load a synthetic case or type your own values. The result recomputes as you type, and every rule links to its page.
Thyroid-function testsMove the sliders or type values
- below range
- within range
- above range
- below range
- within range
- above range
Zones are drawn from your reference intervals. The engine, not the map, decides the pattern.
Laboratory values and reference rangesUse your lab's own interval
µIU/mL and mIU/L are numerically identical.
Interpreted against the reporting assay's upper limit.
Only used in the thyrotoxicosis pathway.
Patient context
Medications and exposures
Symptoms and physical findingsSupportive only: never classify on their own
Symptoms
Signs and decompensation red flags
What it does
- Classifies overt primary, possible subclinical, possible central and thyrotoxicosis patterns using the lab's reference range.
- Reports etiology (autoimmune, postpartum, medication/exposure, pituitary) separately from the functional state.
- Routes pregnancy, fertility, postpartum, medication and assay-interference contexts to dedicated modules.
- Raises an urgent alert for decompensation red flags, and an adrenal-safety alert whenever central hypothyroidism is possible.
- Shows exactly which rule fired and which input values triggered it.
What it never does
- Produce a probability or confidence score.
- Recommend a levothyroxine dose or an automatic treatment.
- Diagnose Graves disease or Hashimoto thyroiditis from an antibody alone.
- Use reverse T3 or an unvalidated symptom score to classify.
Safety gates
These gates run before the pattern is shown, and urgent findings are displayed first.
- Possible central hypothyroidism raises an adrenal-insufficiency check before thyroid hormone replacement.
- Overt hypothyroidism with altered mental status, hypothermia, hypoventilation, instability, marked bradycardia, severe hyponatremia or hypoglycemia raises an urgent escalation alert.
- Biotin or acute severe illness routes the result through an interference review.
Inputs
- TSH and free T4, each with the laboratory reference range
- Anti-TPO / Anti-TG, optional
- Pregnancy, fertility, postpartum and pituitary context
- Medications and exposures (amiodarone, lithium, biotin, …)
- Decompensation signs
Scope
In scope
- Adult-acquired thyroid function patterns
- Documented adult clinical context
Routed out
- Pediatric thyroid disease
- Thyroid nodules and cancer
- Complete pregnancy, postpartum or drug-induced management without a dedicated module
The full rule catalog
7 active rules and 5 sandbox or context entries. Each has its own page with conditions, outputs and evidence. 12 of 12 active rules are exercised by at least one synthetic case.
Functional-state patterns
Safety patterns
Etiology patterns
Context modules
Synthetic cases
Every case is synthetic. Open one to see the inputs, the full result and the rule trace; then change the values.
FX-001
Overt primary hypothyroidism with autoimmune support
A 54-year-old with fatigue, cold intolerance and a TSH of 18.4.
FX-002
Antibody positive with euthyroid function
Strongly positive antibodies, but TSH and FT4 are both normal.
FX-003
Mild subclinical pattern with positive Anti-TPO
Mildly high TSH, normal FT4, positive Anti-TPO.
FX-004
TSH at or above 10 with normal FT4
TSH of 12 with a normal FT4 in a woman of 57.
FX-005
Pregnancy context with mild TSH elevation and Anti-TPO positivity
Pregnant, TSH 3.1 against a pregnancy range, Anti-TPO positive.
FX-006
Preconception infertility context
Trying to conceive, TSH 4.8, Anti-TPO positive.
FX-007
Central hypothyroidism after traumatic brain injury
Fatigue and low libido after a head injury. TSH 1.4, FT4 0.5.
FX-008
Low FT4 with inappropriately mild TSH elevation and pituitary history
Pituitary history. FT4 is low and TSH is only slightly high.
FX-009
Thyrotoxicosis with Anti-TPO positivity
Palpitations and weight loss. TSH 0.02, FT4 2.6, Anti-TPO positive.
FX-010
Postpartum overt hypothyroid pattern with iron deficiency context
Eight months postpartum, TSH 12.4, low FT4, mild anemia.
FX-011
LDL and CK pattern without thyroid labs
No thyroid tests yet, but LDL 168 and CK 285 on a statin.
FX-012
Biotin use with discordant apparent thyrotoxicosis labs
Low TSH, high FT4, and a daily biotin supplement.
FX-013
Severe decompensation risk alert
A 78-year-old, confused and cold, with TSH 38.5 and sodium 128.
FX-014
Amiodarone-associated overt biochemical pattern
On amiodarone: TSH 11.2 and FT4 0.6.
FX-015
No biochemical thyroid dysfunction with depression-like symptoms
Low mood, weight gain and fatigue, with TSH 3.2 and FT4 1.3.
Evidence
Guidelines and references the rules cite. Each rule page lists the ones that support it.
EV-AACE-ATA-2012Garber JR et al. Clinical Practice Guidelines for Hypothyroidism in Adults. Endocr Pract. 2012;18(6):988-1028.
Supports: Subclinical hypothyroidism below TSH 10 requires individualized treatment decision.
EV-ATA-HYPO-2014Jonklaas J et al. Guidelines for the Treatment of Hypothyroidism. Thyroid. 2014;24(12):1670-1751.
Supports: Levothyroxine management framework for hypothyroidism.
EV-ENDO-HYPOPIT-2016Fleseriu M et al. Hormonal Replacement in Hypopituitarism in Adults. J Clin Endocrinol Metab. 2016;101(11):3888-3921.
Supports: FT4 low with low, normal or mildly elevated TSH in pituitary disease supports central hypothyroidism; adrenal safety review.
EV-ATA-PREGAmerican Thyroid Association. Hypothyroidism in Pregnancy; based on ATA pregnancy guidance.
Supports: TPO-positive pregnancy: treat when TSH above 4.0; consider treatment at 2.5-4.0; special-population routing.
EV-ASRM-2024ASRM Practice Committee. Subclinical hypothyroidism in the infertile female population: a guideline. Fertil Steril. 2024;121:765-782.
Supports: Fertility-specific uncertainty and avoidance of blanket treatment logic.
EV-ATA-HASHIMOTOAmerican Thyroid Association. Hashimoto's Thyroiditis.
Supports: Positive antibodies with normal TSH and FT4 do not alone require thyroid hormone treatment.
EV-ATA-THYROIDITISAmerican Thyroid Association. Thyroiditis.
Supports: Destructive/postpartum thyroiditis can have thyrotoxic then hypothyroid phases; etiologic differentiation needed.
EV-ATA-TFTAmerican Thyroid Association. Thyroid Function Tests.
Supports: Reverse T3 is not clinically useful for determining hypothyroidism in healthy nonhospitalized adults.
Frequently asked questions
Does the engine diagnose hypothyroidism?
No. It reports a biochemical pattern, such as the pattern compatible with overt primary hypothyroidism, and every output requires physician confirmation. The diagnosis stays with the clinician.
Why is TSH above 2.5 not flagged as abnormal in a non-pregnant adult?
The engine uses each laboratory's own reference interval. A normal adult, non-pregnant TSH above 2.5 is not labeled abnormal on that number alone. Pregnancy has its own context module with trimester- and assay-specific handling.
How does the engine handle central hypothyroidism?
A low FT4 with a low, normal or only mildly elevated TSH is reported as a possible central pattern. In that case the engine raises an alert to assess adrenal function before thyroid hormone replacement planning.
Does a positive Anti-TPO mean Hashimoto thyroiditis or treatment?
A positive Anti-TPO supports an autoimmune etiology. With normal TSH and FT4 it does not by itself indicate thyroid hormone treatment. With thyrotoxicosis it does not distinguish Graves disease from destructive thyroiditis.
Why does the engine not output probabilities?
No probability calibration exists for these rules, so the engine does not display one. Outputs are states, matched rules and the evidence that triggered them.
Specification status: Specification v1.1.0. Reviewed by Dr. Vinicius Amaral, D.O. Not for clinical production use.
Clinical review: Dr. Vinicius Amaral, D.O., Medical reviewer, 2026-10-03.
Keep going
CPE-008
Adrenal Insufficiency Pattern Engine
Morning cortisol bands, ACTH and stimulation results, glucocorticoid exposure and the adrenal-crisis situations that cannot wait.
CPE-004
Iron Deficiency Pattern Engine
Iron status from ferritin and transferrin saturation, read correctly when inflammation, heart failure or kidney disease changes the rules.
CPE-010
PCOS Pattern Engine
Two of three criteria in adults, irregular cycles plus hyperandrogenism in adolescents, and PCOS only after other causes are excluded.
Resources
Chronic low-grade inflammation and biomarker patterns
Game
Pattern challenge
Guess what the engine will report.
Try Biod in your practice
The Clinical Intelligence Layer — transforming fragmented clinical information into structured intelligence for clinician review and action.