Prototype (revised) — slide “The problem” · outputs rail from the datamarts Option C only: connector from Marts · no status badges · uniform intensity · capitalized box texts.
AritmoLab · Policlinico San Donato
The problem04 · The problem
From free text to a research platform

The clinical truth lives in unstructured columns

“Il paziente riferisce sincope ricorrente; ECG basale con sopraslivellamento ST in V1–V3; test provocativo con flecainide positivo per pattern Brugada.”

EN“The patient reports recurrent syncope; baseline ECG with ST-segment elevation in V1–V3; flecainide provocation test positive for a Brugada pattern.”

SourcesCardioref · Genetic data
Omics Portal · ECG
→
Stagingraw replica
→
Integration3NF · text miningAI
→
Data Warehousestar schema
→
Martsdbt · Superset
coming up — how we read the text
01
Audited numbers58,438 letters · 73,389 pathologies · 2,307 Brugada patients
02
Hard clinical NLPnegation · family attribution · two languages
03
Clinical ontologytwo tiers, ordered by clinical precedence
04
Trust by designquality gates on 100 manual reviews
outputs of the datamarts
Dashboardsclinical review on Superset
Machine Learningcohort tables for model training
Predictive Statisticsoutcomes & risk analysis
↑
AIThothII — natural-language → SQL over the DWH, builds the final datamarts
Role of AI in the Analysis of Unstructured Clinical DatabasesDr. Marco Pancotti - MultiPhysixLabDr. Sara Paratico - Gruppo San DonatoSan Donato Milanese, Milan, Italy · 2–3 October 202604 / 14

Option C, revised. The rail drops straight from the Marts stage; the three outputs have equal visual weight and no status labels; texts inside the diagram boxes are capitalized; the EN translation is now the same size as the Italian.