Repeating an experiment following the exact same procedures and using the exact same data and tooling should always give the same results. This is referred to as replicability, a key principle within the scientific research community. To support replicability in a clinical context, recording and reporting detailed information about data collection procedures and machine settings are standard procedures. However, since data is increasingly used for machine learning purposes it also becomes essential to extend this recording and reporting to the machine learning models that are created. Within this study we attempted to replicate a machine learning model that identifies patients with a high chance of a heart attack from a published article to identify what information is key for replicating machine learning models.