DiagnostiQ AI: Medical Report Intelligence and Simplification Platform

Laboratory reports contain clinically useful measurements but are often difficult for patients to interpret because they combine abbreviations, units and laboratory-specific reference intervals. This paper presents DiagnostiQ AI, a web-based healthinformatics platform that converts digital or scanned laboratory reports into structured biomarker results and patient-oriented explanations. The system uses PyMuPDF for digital PDF text extraction and Tesseract OCR as a fallback for scanned pages and images. A deterministic regular-expression parser, alias dictionary and reference-range knowledge base identify biomarkers, values and units, prioritising ranges printed on the source report. Abnormal markers and extracted alerts are combined using a transparent risk score, while the FLAN-T5 language model rewrites verified facts into plain-language explanations. The platform also supports English, Hindi and Tamil output, speech narration, report comparison, biomarker inspection, medication-interference screening, consultation preparation and guided breathing. A React 19 frontend communicates with a Flask REST backend. Functional testing covers document ingestion, parsing, classification, translation caching, comparison trends, fallback behaviour and endpoint responses. The paper describes the architecture, decision logic, implementation and observed functional behaviour, while identifying limitations in document layout handling, reference-range personalisation and online translation. DiagnostiQ AI is designed as an educational aid and does not diagnose disease or prescribe treatment.