NYUAD Hackathon for Social Good in the Arab World: Focusing on Quantum Computing (QC) and Aritificial Intelligence (AI)
QURAI delivers a full pipeline from cancer detection to treatment using patient data and quantum optimization to personalize safer, low-toxicity radiotherapy that targets cancer cells and minimize risk over healthy tissue.
The workflow is organized into four main stages:
- Input: Biological cell parameters from patients.
- Purpose: Gather necessary data features (e.g., size, texture, perimeter) for diagnosis.
- Input: Patient cell parameters.
- Output:
1→ Cancer detected (Positive)0→ No cancer detected (Healthy)
- Description: A trained ML model classifies whether the input data suggests the presence of cancer.
- Formulate the beam angle selection problem as a QUBO (Quadratic Unconstrained Binary Optimization) problem.
- Solve using Quantum Approximate Optimization Algorithm (QAOA).
- Classical post-processing to refine and validate the results.
- Purpose: Find the optimal radiation beam angles for therapy, minimizing damage to healthy tissues while maximizing impact on the tumor.
- Machine Learning: Cancer classification based on patient cell data.
- Quantum Computing: QUBO formulation and QAOA for optimization tasks.
- Classical Computing: Post-processing optimization results.
- Medical Data Processing: Handling and interpreting biological parameters.
- Expand dataset to include multiple cancer types.
- Integrate more quantum algorithms for other treatment parameters.
- Improve classical-quantum hybrid processing pipeline.
- Add explainability modules for ML predictions to increase trust and transparency.
Supports UN SDGs:
- Good Health and Well-being (3)
- Industry, Innovation, and Infrastructure (9)
- Sustainable Cities and Communities (11)
Presentation link : https://www.canva.com/design/DAGlt6_FInQ/UdR-tRDGpVxuW7hIIK8q5w/edit?utm_content=DAGlt6_FInQ&utm_campaign=designshare&utm_medium=link2&utm_source=sharebutton