An enterprise LIMS that takes a pathology sample from booking to signed PDF report.
2024 · Healthcare / Diagnostics · role: owner
● status: production·7 services·~2 min read
01 — overview
PathLab digitally transforms pathology lab operations end to end: patients book tests and download secure reports, technicians track samples through a multi-status lifecycle, doctors monitor results, and administrators run the lab on real analytics.
02 — the problem
Pathology labs run on manual, error-prone processes — handwritten registers, long turnaround times, and results patients can't access without a phone call or a second visit.
Lab data is sensitive medical data: it needs strict role separation, audit trails, and secure delivery, not a shared spreadsheet.
03 — architecture
A React frontend talks to a layered Spring Boot backend (controller → service → repository → entity) over PostgreSQL. Caffeine caches hot reads, Flyway versions the schema, and reports render through OpenHTML2PDF + FreeMarker templates.
04 — engineering decisions
01
Samples move through six explicit states — collection pending → collected → in transit → received → tested → discarded — with technician assignment and barcode generation at each step. State is the source of truth, not tribal knowledge.
02
ADMIN, LAB_TECH, DOCTOR, and PATIENT each see exactly their slice: patients their reports, technicians their queue, doctors their patients, admins the whole lab — enforced in Spring Security, not the UI.
03
Parametric results render through FreeMarker templates into OpenHTML2PDF — professional, consistent reports and invoices with bulk-processing support.
04
Caffeine caching on the hot read paths (test catalog, pricing) keeps patient-facing pages fast without complicating the write path.
Explain this project
A Laboratory Information Management System on Java 21 / Spring Boot with a layered controller → service → repository architecture over PostgreSQL. Samples move through an explicit six-state lifecycle — the state is the source of truth, with barcode generation and technician assignment at each step — and four roles (ADMIN, LAB_TECH, DOCTOR, PATIENT) are enforced in Spring Security rather than the UI. Reports render deterministically through FreeMarker → OpenHTML2PDF, Flyway versions the schema, and Caffeine caches hot read paths like the test catalog and pricing.
05 — features
06 — stack
The repository has the code, the commits, and the decisions in context — or reach out and I'll walk you through it.