
Submissions will be judged through a combination of human review by 1) AI/ML and clinical experts to ensure usefulness of data and documentation for secondary research, and 2) formula metrics to ensure accuracy, completeness, consistency, and validity of AI-ready dataset.
The repository includes clear, accessible scripts used to generate both the raw and AI-ready datasets, and they were made available by the stated deadline. The code is well organized and sufficiently documented to support reproducibility and independent review.
Both the Raw and AI-Ready datasets were made available in the team’s private GitHub repository by the posted submission deadline, and were clearly labeled and accessible for review.
The submission form is clear, descriptive, and informative. Responses provide sufficient detail to understand the approach, decisions, and outcomes, enabling effective evaluation without unnecessary ambiguity.
The submission demonstrates a high level of accuracy. The data is consistent, well-validated, and free from obvious errors, supporting reliable analysis and downstream use.
The code is well-documented and informative. Structure, naming, and in-line comments clearly convey intent and logic, making the implementation easy to understand, review, and reuse.
The proposed use cases are thoughtful and well articulated. They demonstrate a strong understanding of the dataset, the AI-readiness steps applied, and how those enable meaningful applications across medicine, public health, and data science/AI contexts. The examples emphasize quality and depth over quantity, reflecting clear potential for real-world impact and researcher training.