Informatics Methodology &
Clinical Validation

An overview of the deterministic and probabilistic protocols governing the Renal Companion inference engine.

Dual-Layer Validation Architecture

Renal Companion operates on a hybrid inference model that balances the absolute safety of clinical rules with the contextual flexibility of Large Language Models.

Layer 1: Deterministic Clinical Logic

A hard-coded rule engine based on KDOQI 2024 and KDIGO 2024 practice guidelines. Every dietary entry is validated against a 50-point safety checklist including CKD stage, GFR slope, and serum electrolyte levels.

if (K_intake > daily_limit) { trigger_overload_override(); adjust_safety_buffer(0.85); }
Layer 2: Probabilistic RAG Inference

Retrieval-Augmented Generation (RAG) allows Murshid to synthesize advice within a restricted clinical sandbox. The AI is bounded by the deterministic layer, ensuring it never suggests intake exceeding calculated safety limits.

murshid.reason(query, patient_context) .map(KDOQI_Safe_Zones) .validate(Layer1_Constraints);

Validated Clinical Formulas

CKD-EPI (2021) Equation

Used for eGFR estimation without race variables, complying with the latest international clinical recommendations.

Modified Adjusted Body Weight (AjBW)

Critical for calculating protein requirements in obese or malnourished CKD patients to prevent over-prescription.

Corrected Calcium / PRAL Indices

Automated adjustment for Hypoalbuminemia and Potential Renal Acid Load monitoring.

Informatics Safety Guardrails

To prevent 'hallucinations', the System Prompt incorporates a Clinically-Constrained Grammar (CCG) that restricts output to KDOQI-verified nutrient ranges.

Forensic Clinical Documentation

Technical specifications for internal logic and vision-based auditing

Protocol A: Phosphorus Additive Audit

The vision-based inference engine performs deep-parsing of ingredient lists to detect inorganic phosphorus additives, which have nearly 100% absorption rates in CKD patients.

TARGET_PHOS_ADDITIVES = { 'E338', 'E339', 'E340', 'E341', 'E343', 'E450', 'E451', 'E452', 'Sodium Phosphate', 'Pyrophosphate' }
Protocol B: Weight-Based Recalibration

Protein and fluid dosing utilize the Adjusted Body Weight (AjBW) modification of the Hamwi formula for patients with BMI outliers (>30 or <18.5).< /p>

AjBW = IBW + 0.25 * (ActualDryWeight - IBW) // Prevents over-prescription in obesity

Research & Development Roadmap

The trajectory towards clinical implementation and multi-center validation

Phase I: Informatics Sandbox

Validation of the deterministic rule engine against 10,000+ synthetic patient profiles to ensure zero-hallucination guardrails.

Phase II: Observational Pilot

Monitoring dietary adherence in a controlled cohort of 200 patients (CKD Stage 3-5) using the Renal Companion interface.

Phase III: Clinical Integration

Integration with Hospital Information Systems (HIS) via HL7/FHIR for direct laboratory data synchronization.