Top-ranked capstone project
Mohamed Amin El-lithy · Graduate project
- Client
- Mohamed Amin El-lithy
- Industry
- Graduate project
- Deployment status
- Prototype
Inside the project
Solution
A #1-ranked graduate capstone demonstrating the complete machine-learning workflow — problem framing, data cleaning, exploratory analysis, feature preparation, model comparison, validation, explainability, documentation and presentation. It is an educational prototype, not a clinical system: it cannot support diagnosis or treatment.
Our approach
Judged #1 in its cohort, the project is evidence of applied capability — analytical reasoning, a disciplined end-to-end ML process, and clear communication — the portfolio-ready outcome Epsilon programs are built to produce. As an educational prototype its limitations are explicit: dataset representativeness, threshold choice, false positives/negatives, and the need for external validation and qualified medical review before any real-world use.
What this case does not show
An educational capstone prototype, not a clinical system: it cannot support diagnosis or treatment, and was not deployed to a customer environment.
Proposed Performance Targets
Proposed target
50%
Target relative increase in capstone rubric scores
- What is measured
- Evaluate project development from the first formal review to final submission using consistent criteria for data preparation, methodology, evaluation and presentation.
- How it would be evaluated
- First formal project review to final assessment of the same projects within a defined cohort.
- What this figure is not
- A capstone rubric score is not model accuracy, clinical validation, a deployment-success rate or commercial ROI.