YSTE

Project documentation.

Project Helix documented across five sections: overview, problem and context, research and design, technical development, and reflection.

01

Project overview

What is Project Helix?

A clinical decision-support platform investigating how pancreatic cancer patient information — scans, bloods, histopathology, GP referrals, MDT notes — can be organised into one clearer view. PRISM is the working prototype built to explore that question.

02

Problem & context

The information challenge

Pancreatic cancer care often spans multiple hospitals, departments and systems. That fragmentation can delay access to information, add administrative load, and make multidisciplinary decision-making harder. Project Helix investigates how a secure, clinician-centred platform could improve how that information is organised, accessed and presented — without replacing clinical judgement.

03

Research & design

Decisions, sources, feedback

Development follows an iterative process: clinician consultation, system design, prototyping, testing, and refinement based on feedback. As of July 2026, structured input has come from Dr. Lorraine Walsh, a Consultant [confirm title] treating pancreatic cancer patients — feedback that's directly shaped design decisions. Background research draws on MDT workflow literature, EHR and clinical decision-support research, healthcare interoperability, GDPR, and journals including The Lancet, Nature Medicine and JMIR Medical Informatics.

04

Technical development

Architecture, boundaries, choices

A functional desktop prototype built with Electron, React and Supabase. Work to date covers architecture design, database implementation, authentication, patient management, interface development, version control and deployment. The platform is scoped deliberately: it's decision support, not diagnosis, and it's built to complement hospital systems, not replace them.

05

Reflection

What's next

Ongoing usability testing and feature development, guided by continued clinician feedback ahead of YSTE in January 2027. The open question the project keeps returning to: what does responsible, clinician-shaped software actually look like in a high-stakes setting — and where should it stop?