PRedicting with Artificial Intelligence riSk aftEr acute coronary syndrome

The methodology for calculating the PRAISE risk score is described in:
D'Ascenzo F, De Filippo O, Gallone G, et al, Machine learning-based prediction of adverse events following an acute coronary syndrome (PRAISE): a modelling study of pooled datasets, The Lancet, Volume 397, Issue 10270, 2021, Pages 199-207, ISSN 0140-6736.

Single patient analysis

In order to run a single patient anlysis with PRAISE it is necessary provide all the clinical, therapeutic, angiographic and procedural data available for the patient, then press the SUBMIT button. The result will be shown at the bottom of the page, showing the calculated score for death, ReAMI and BARC MB events with the corresponding risk class (low/intermediate/high). The score is calculated as a probability, so it is always included between 0 and 1.

Note that the score will be calculated independently from the number of variables provided; nonetheless it is worth noting that the more information are provided the more accurate the prediction will be.

Clinical variables

Therapeutic variables

Angiographic variables

Procedural variables

Multiple patients analysis

In order to run a multiple patient analysis with PRAISE it is necessary to provide a correctly formatted CSV file containing the patients data. Every row of the file identifies a different patient by its its medical data; the correct header of for the CSV file is downlodable below togheter with an example of well-formetted file.

Once the CSV file is ready you have to upload it: click on the 'Browse' button below, navigate to your CSV file, selct it and click 'Open'. Alternatively you can drag and drop it in the file field below.

Now you can click the SUBMIT button: in this way the file will be sent for the evaluation. The result will be shown belown togheter with a link for downloading the dataset augmented with the scores.

Download header
Download example