Så utvärderar vi hallucinationsrisken för klinisk AI

Notats bevis är inte en magisk noggrannhetsprocent. Det är en transparent metodik: extrahera kliniska fakta först, generera anteckningar från dessa fakta, visa den råa kontexten för klinikern och granska varje output mot bevisen.

Utvärderingsmetod

1. Build the reference facts

A reviewer reads the encounter material and creates a reference set of clinical facts: symptoms, negatives, medication changes, assessment, plan, safety-netting, and coding-relevant details.

2. Compare transcript-direct output

We generate a note from transcript-style input and mark unsupported statements, missing high-salience facts, incorrect attribution, and invented certainty.

3. Compare FactsContext output

We generate documentation from extracted facts and review whether each sentence is supported by a visible fact. The clinician-facing fact list is evaluated as part of the output, not hidden.

4. Record limits, not magic numbers

We do not publish fake accuracy percentages. Each evaluation note includes dataset scope, review date, known limitations, and examples of what the system still requires clinicians to verify.

Vad granskare markerar

Unsupported clinical assertion

Wrong medication, dose, frequency, or route

Missing red-flag negative or safety-net advice

Wrong diagnosis certainty: possible vs established

Wrong speaker attribution

Unsupported ICD-10 suggestion

Clinician-visible evidence for each key statement

Facts reusable for notes, codes, referrals, and patient instructions

Daterad intern anteckning

Aktuell utvärderingsanteckning, 2026-07-01: den här sidan dokumenterar metoden och de kvalitativa exempel som används för att utvärdera FactsContext-arkitekturen. Den hävdar inte extern validering eller universell noggrannhet. Nästa bevismilstolpe bör vara en blindad, specialitetsstratifierad granskning med datasetstorlek, bedömarsamstämmighet och andel obestyrkta påståenden rapporterade öppet.

Faktautvinning kontra direkt generering från transkription

Medication change discussed twice

Risk vid direkt transkription

“Increase amlodipine to 10 mg and stop lisinopril.” The transcript contained a correction: the clinician first considered stopping lisinopril, then decided to continue it after reviewing renal function.

FactsContext-output

Facts: amlodipine increased to 10 mg daily; lisinopril continued; renal function normal; review in 6 weeks. Note generated from those facts only.

Varför det spelar roll

Transcript-direct generation can smooth over corrections. FactsContext preserves the final decision as a discrete fact before writing.

Negative finding matters

Risk vid direkt transkription

“No neurological symptoms.” The actual encounter only documented no saddle anesthesia and no bladder symptoms; leg radiation was present.

FactsContext-output

Facts: left leg radiation to calf; SLR positive left; no saddle anesthesia; no bladder or bowel symptoms. Note keeps the negatives specific.

Varför det spelar roll

Broad negative statements are risky. The fact list keeps the clinical context granular and reviewable.

Code suggestion requires evidence

Risk vid direkt transkription

Suggested J44.1 for COPD exacerbation without showing the symptom or treatment evidence.

FactsContext-output

Facts: increased breathlessness, purulent sputum, prednisolone burst, antibiotics started. Suggested J44.1 with those facts as evidence.

Varför det spelar roll

The code is easier to verify because the reason for the suggestion is visible, not buried in prose.

Kända begränsningar

Clinician review remains mandatory. Notat drafts; clinicians verify and sign.

The method reduces unsupported statements by architecture, but no clinical AI should claim zero hallucinations.

Small internal evaluations are useful for engineering direction, not a substitute for external clinical validation.

Specialty, language, audio quality, speaker overlap, and local coding rules can change performance.

Läs beviset, granska sedan dina egna fakta.

Utvärderingsmetoden är enkel eftersom produkten är designad för att vara inspekterbar: fakta först, anteckning sedan, klinisk granskning alltid.

Prova Notat gratis