Healthcare Intelligence Platform: Domain Data Mart Portfolio & Architecture Framework Guide

This release delivers a major architectural update to the Healthcare Intelligence Reference Architecture ecosystem under the permanent identifier DOI: 10.5281/zenodo.22255118. This consolidated handbook unifies the operational consumption tiers, domain-driven data marts, and advanced predictive capabilities into a single, seamless logical reference blueprint. Authored and architected by Chotika Ruangjun, this technical framework is designed strictly at the vendor-neutral Logical Layer. To preserve absolute platform security and corporate data confidentiality, the blueprint is white-labeled and contains zero live physical schemas, database tables, or backend server source codes. Key Architectural Portfolios Contained in This Release: 1. ENTERPRISE DATA MART PORTFOLIO FRAMEWORK: Outlines the subject-oriented paradigm shift from traditional, siloed reporting to an decoupled analytical ecosystem. It details the 5-Layer Analytical Portfolio Blueprint (Clinical, Research, Medication, Financial, and Supply Chain) and introduces the Concentric Shared Dimensions Matrix (Patient, Geography, Disease, Drug, Facility, Time, and Outcome) ensuring absolute metric consistency across all enterprise metrics. 2. NON-COMMUNICABLE DISEASE (NCD) INTELLIGENCE ARCHITECTURE: Provides the end-to-end framework for tracking chronic disease burdens over multiple decades (including Diabetes, Hypertension, CKD, and CVD). This module models individual clinical activities into unbroken longitudinal care vectors, utilizing an automated Slowly Changing Dimension (SCD) Type 2 tracking engine to capture diagnostic and staging state transitions without data leakage. 3. CANCER INTELLIGENCE & PRECISION ONCOLOGY ARCHITECTURE: Establishes a highly specialized blueprint for oncology data modeling across major high-burden tumors and hematological malignancies. It introduces advanced precision medicine matrices by integrating structural molecular alterations and biomarker profiles (such as HER2, EGFR, and PD-L1 assays) directly into longitudinal cancer journey vectors and treatment response lifecycles. 4. AI-READY ARCHITECTURE & SEMANTIC REASONING FOUNDATIONS: Configures the semantic data layer for modern machine learning, feature engineering store optimization, and conversational decision intelligence. The curated logical views within this handbook serve as safe, structured boundaries for Enterprise Large Language Models (LLMs) and Knowledge Layers, establishing the core foundations for hallucination-free Retrieval-Augmented Generation (RAG) applications.