Back to Case Studies
Custom Software

OpenCures

Clinical Decision Platform for Functional and Integrative Medicine

249+

QA Test Cases

31

Conditions Modelled

94

Intervention Subdomains

Phase 2

Status

Project Overview

  • OpenCures is a clinical platform built with Dr. Sanjay Sachdeva for doctors practising functional and integrative medicine in India.
  • It turns a patient's lab reports and history into a timeline, checks for contraindications, and helps the doctor build a treatment protocol from a graded evidence base.
  • Neurabit took the product from prototype to a working MVP, with weekly product reviews alongside the clinical team. Phase 2 is now being planned.
  • The scope was agreed through a formal sign-off sheet covering conditions, intervention domains, the grading rubric and the database design.

Challenges

Lab Data Locked in PDFs

Patient reports arrive as typed and scanned documents, and values had to be read accurately and the same way every time.

Complex Patients

Real patients have several conditions and medicines at once, so the platform had to flag drug and treatment contraindications.

Evidence Across Many Traditions

Interventions span many domains, and each needed a clear tier and rank that a doctor can trust.

Clinical Quality Bar

A tool doctors rely on needs far more testing than a typical app.

Platform Features

  • AI Lab Report Reading: Upload a report and every test value appears on the patient timeline with the right units and dates.
  • Patient Timeline: Clinical history and lab results in one view.
  • Contraindication Checks: Flags interactions between medicines, conditions and proposed interventions.
  • Protocol Builder: Doctors assemble a treatment plan, generate the plan document and schedule follow-ups.
  • Reviewer and Admin Panels: Clinical reviewers validate content, admins control card visibility and data.

A Graded Evidence Base

  • 31 condition nodes and functional patterns, with insulin resistance modelled as a parent node.
  • 11 intervention domains and 94 subdomains, agreed with the clinical lead.
  • A grading rubric that gives every intervention a tier and a rank.

Quality

  • A single QA suite organised by feature, covering both product QA and AI QA, with 249+ test cases.
  • AI tests check that the same report gives identical values every time, across typed and scanned reports.
  • The suite is laid out so automated QA runs and human testers can both read and update it.

Technology Stack

Web PlatformLLM Document ExtractionClinical Knowledge GraphAdmin & Reviewer Panels

Interested in a similar solution?

Let's discuss how we can build something extraordinary for your business.

Book a Call