NidaanPath AI is designed with transparency, safety, and human oversight at its core.
NidaanPath is not an emergency service. A person experiencing urgent or severe symptoms must seek immediate professional medical assistance. Call emergency services or visit the nearest hospital.
โ NidaanPath Does NOT
-
โDiagnose any disease
-
โPredict medical outcomes
-
โPrescribe medication
-
โRecommend specific medical tests
-
โChange or stop medication
-
โReplace healthcare professionals
-
โIdentify medical negligence
-
โGuarantee medical safety
-
โDecide which doctor is correct
-
โMake autonomous clinical decisions
โ NidaanPath ONLY
-
โReconstructs the diagnostic journey timeline
-
โDetects process gaps from confirmed records
-
โAsks clarifying questions about process gaps
-
โMatches new evidence to open gaps
-
โPrepares evidence for safe clinician review
-
โDisplays source fragments for every finding
-
โRequires patient confirmation before data enters journey
-
โUses deterministic rules for stagnation detection
-
โPreserves and displays all clinical uncertainty
-
โGenerates evidence-grounded clinician packets
๐ก Three Mandatory Safety Gates
Patient Confirmation
No AI-extracted information enters the Diagnostic Journey Twin until the patient explicitly confirms or corrects it.
Deterministic Stagnation Rules
Journey state is determined by transparent Python rules only. Gemini cannot independently classify a journey as stagnant.
Clinician Escalation
Any medication uncertainty, contradiction or clinical ambiguity is escalated to the clinician packet. NidaanPath never resolves clinical uncertainty autonomously.
๐ง Technical Safeguards
All demo records are synthetic. No real patient data is used anywhere.
Gemini API key is never exposed to the browser, JavaScript, or logs.
Every extracted finding includes a source document and source fragment.
Extraction confidence is shown for every finding.
Uncertain fields are always preserved and marked, never silently resolved.
Every agent tool call is logged with timestamp and safety-gate status.
Uploaded files are processed server-side and are not shared.
Patient identifiers use synthetic case codes only.
โ Limitations
- NidaanPath is a prototype MVP, not a regulated medical device.
- Extraction accuracy depends on document legibility and format.
- Mock AI mode uses deterministic responses, not real language understanding.
- The system cannot verify document authenticity.
- Journey reconstruction relies on confirmed documents only.
- No Tamil OCR is supported in the current MVP.