Contact person: Business Client

Phone:Show

Email:Show

Location: NORRKÖPING, Sweden

Budget: Recommended by industry experts

Time to start: As soon as possible

Project description:
"I’m building a standalone, AI-powered healthcare platform that combines digital anamnesis with in-clinic workflows. The first milestone is a functional prototype whose core task is diagnosis support: the system collects a patient’s symptom history, runs it through an explainable AI model, and hands clinicians a concise, ranked list of possible diagnoses.

Key requirements
• Stand-alone architecture – no EMR/HIS integration for now; everything runs in its own Dockerized environment.
• Multilingual capability – the NLP pipeline must work in several languages (English plus at least two additional major languages) and be easy to extend.
• Dual-facing web UI – patient questionnaire on one side, clinician dashboard on the other, built with a modern JavaScript framework.
• Explainability & audit trail – each suggested diagnosis should link back to the data points and model weights that influenced it.
• Security & privacy – encrypted data in transit and at rest, HIPAA/GDPR-aligned design patterns, role-based access.
• Admin basics – user management, data export, configurable questionnaires.

Deliverables
1. Architecture and data-schema document.
2. Trained diagnosis-support model with well-documented inference API (Python + TensorFlow/PyTorch or similar).
3. Working web prototype packaged in Docker (patient & clinician views).
4. Deployment guide/README for local or AWS setup.

Feel free to suggest the exact tech stack; I’m comfortable with Python (FastAPI), React/Vue, and standard healthcare data libraries but open to your expert recommendation.

When you respond, please include:
• A brief overview of similar healthcare or NLP projects you’ve completed.
• Your preferred tools/libraries for this build.
• A rough timeline broken into major phases.

The idea: Patients in the waiting room, via an AI-integrated tablet, fill in their symptoms and receive dynamic follow-up questions. The answers are sent directly to the physician, who can prepare diagnostics and a treatment plan before the consultation.
The physical visit will then focus only on:
Medical examinations & tests

Physical assessment

Any remaining questions from the physician

Vision:
Reduce waiting times by up to 50% by cutting down unnecessary physician time.


Create a hybrid model where AI optimizes anamnesis and triage, but the final medical decision-making remains at the physical visit." (client-provided description)