Validation Dashboard

Deterministic acceptance suite. Each fixture is replayed through the CKD engine and compared against canonical expected output, including AKI safety gates, CGA classification, and chronicity confirmation.

← CPE-002 validation

20 of 20 fixtures pass

Fixtures Tested

Passing

Rules Exercised

29 / 35

Safety Alerts Covered

Fixture	Expected	Rules fired	Status	Action
CKD-FX-001 Persistent moderately increased albuminuria with preserved GFR Detect confirmed early CKD without reduced eGFR	DET-001CHR-001GFR-001CGA-001CGA-002CGA-004 CKD G1A2	GFR-001CGA-002DET-001CHR-001CGA-001CGA-004KFRE-002	Pass	Open
CKD-FX-002 Persistent severely increased albuminuria with mildly decreased GFR Ensure G2A3 is not collapsed into A2	DET-001CHR-001GFR-001CGA-001CGA-002CGA-004 CKD G2A3	GFR-001CGA-002DET-001CHR-001CGA-001CGA-004KFRE-002	Pass	Open
CKD-FX-003 Single reduced eGFR without chronicity Prevent false-positive CKD confirmation	DET-001CHR-003GFR-003	GFR-001GFR-003CGA-002DET-001CHR-003	Pass	Open
CKD-FX-004 Acute creatinine rise before CKD staging Run AKI gate first	SAFE-AKI-001	SAFE-AKI-001GFR-001GFR-003CGA-002DET-001	Pass	Open
CKD-FX-005 Isolated suspected G3a requiring cystatin C confirmation support Confirm eGFRcr-cys workflow	GFR-003	GFR-001GFR-003CGA-002DET-001CHR-003	Pass	Open
CKD-FX-006 Combined eGFR reclassification support Use combined equation when cystatin C available	GFR-002	GFR-001GFR-002CGA-002KFRE-002	Pass	Open
CKD-FX-007 Confirmed CKD G3bA3 with diabetes Detect high-risk diabetic kidney disease pattern	CAUSE-001KFRE-001REF-001 CKD G3bA3	GFR-001CGA-002DET-001CHR-001CGA-001CGA-004CAUSE-001KFRE-001REF-001	Pass	Open
CKD-FX-008 Active glomerular disease red flag Flag urgent glomerular pattern rather than benign CKD stage only	SAFE-GN-001REF-001	SAFE-GN-001GFR-001CGA-002DET-001CHR-003REF-001	Pass	Open
CKD-FX-009 Anemia in CKD context without overdiagnosing EPO deficiency Require workup before causal claim	ANEMIA-001	DET-001CHR-002CGA-001CGA-005ANEMIA-001REF-001	Pass	Open
CKD-FX-010 Vitamin D low alone does not establish CKD-MBD Prevent inappropriate CKD-MBD treatment output		DET-001CHR-002CGA-001CGA-005	Pass	Open
CKD-FX-011 Severe potassium elevation with eGFR unknown Fire global safety alert independently of CKD status	SAFE-K-001	SAFE-K-001DQ-003REF-001	Pass	Open
CKD-FX-012 Metformin contraindication threshold Apply FDA renal impairment safety logic	MED-MET-001	GFR-001DET-001CHR-003MED-MET-001REF-001	Pass	Open
CKD-FX-013 Kidney failure range does not automatically mandate dialysis Avoid unsafe KRT statement	REF-001KFRE-001 CKD G5A3	GFR-001CGA-002DET-001CHR-001CGA-001CGA-004KFRE-001REF-001	Pass	Open
CKD-FX-014 Possible monoclonal protein-related kidney disease Use UACR/uPCR discrepancy and systemic features	CAUSE-003CAUSE-004REF-001	CGA-002DET-001CHR-003CAUSE-003CAUSE-004REF-001	Pass	Open
CKD-FX-015 ACE inhibitor monitoring rather than automatic cessation Support guideline-concordant medication stewardship	MED-RASI-001	DET-001CHR-002MED-RASI-001	Pass	Open
CKD-FX-016 ACE inhibitor creatinine rise requiring evaluation Flag >30% change without automatic stop statement	MED-RASI-001MED-RASI-002	DET-001CHR-002MED-RASI-001MED-RASI-002	Pass	Open
CKD-FX-017 UACR positive random sample requires confirmation Implement sample-specific confirmation logic	ACR-001	GFR-001CGA-002ACR-001DET-001KFRE-002	Pass	Open
CKD-FX-018 Transient UACR confounder Prevent albuminuria diagnosis during active UTI	ACR-001ACR-002	GFR-001CGA-002ACR-001ACR-002DET-001KFRE-002	Pass	Open
CKD-FX-019 Possible obstructive kidney process Detect reversible etiologic concern	SAFE-OBS-001REF-001	SAFE-OBS-001DET-001CHR-003REF-001	Pass	Open
CKD-FX-020 SGLT2i early eGFR dip Prevent misclassification as drug harm without red flags	MED-SGLT2-001	DET-001CHR-002MED-SGLT2-001	Pass	Open

Validation-only · Synthetic clinical cases · Physician adjudication required. CPE-003 never confirms CKD from a single abnormal lab, never bypasses AKI safety gates, and never autonomously recommends dialysis, ESA, calcitriol, phosphate binders, or medication discontinuation. Pattern Evidence Strength scores are not diagnostic probabilities.