acribis-scores 0.3.0

Implementation of several cardiovascular risk scores

Cardiovascular Risk Scores for the ACRIBiS Project

Risk scores

CHA₂DS₂-VASc

(Congestive heart failure/Left ventricular dysfunction, Hypertension, Age ≥ 75 years, Diabetes mellitus, Prior Stroke/Transient ischaemic attack/Thromboembolism, Vascular disease, Age 65-74 years, Sex category)

Original publication: https://doi.org/10.1378/chest.09-1584

Predicts the 1-year risk of thromboembolic events in patients with atrial fibrillation (AF)

HAS-BLED

(Hypertension, Abnormal renal and liver function, Stroke, Bleeding, Labile international normalized ratios, Elderly, Drugs or alcohol)

Original publication: https://doi.org/10.1378/chest.10-0134

Predicts the 1-year risk of major bleeding in AF patients

ABC-AF

(Age, Biomarker, Clinical history)

Stroke

Original publication: https://doi.org/10.1093/eurheartj/ehw054
Recalibration: https://doi.org/10.1161/CIRCULATIONAHA.120.053100

Predicts the 1-year risk of stroke in AF patients

Bleeding

Original publication: https://doi.org/10.1016/S0140-6736(16)00741-8
Recalibration: https://doi.org/10.1161/CIRCULATIONAHA.120.053100

Predicts the 1-year risk of bleeding in AF patients

Death

Original publication: https://doi.org/10.1093/eurheartj/ehx584

Predicts the 1-year risk of death in AF patients

CHARGE-AF

(Cohorts for Heart and Aging Research in Genomic Epidemiology)

Original publication: https://doi.org/10.1161/JAHA.112.000102

Predicts the 5-year incidence of AF

SMART

(Secondary Manifestations of ARTerial disease)

Original publication: https://doi.org/10.1136/heartjnl-2013-303640

Recalibration: https://doi.org/10.1161/CIRCULATIONAHA.116.021314

Assumptions on the effects of antithrombotic treatment: https://doi.org/10.1016/S0140-6736(09)60503-1

Predicts the 10-year risk of recurrent ischaemic events in patients with pre-existing vascular disease

SMART-REACH

(Secondary Manifestations of ARTerial disease-REduction of Atherothrombosis for Continued Health)

Original publication: https://doi.org/10.1161/JAHA.118.009217

Predicts the 10-year risk and lifetime risk (i.e. risk until age 90 years) for (recurrent) myocardial infarction, stroke or vascular death and (recurrent) event free life-expectancy

MAGGIC

(Meta-Analysis Global Group In Chronic Heart Failure )

Original publication: https://doi.org/10.1093/eurheartj/ehs337

[!IMPORTANT] The implementation differs from the formula presented in the article!

Statement on the website of the online calculator:
From 18 September 2013, the integer score will increase by 2 if heart failure was diagnosed > 18 months ago. This may affect a comparison of the current result to risk assessments before this date.

Calculator returns integer risk score, can be used to look up 1- and 3-year risk of death for patients with heart failure (HF)

Barcelona Bio-HF

Original publication (v1): https://doi.org/10.1371/journal.pone.0085466

Recalibration (v2): https://doi.org/10.1002/ejhf.949

Recalibration (v3): https://doi.org/10.1002/ejhf.2752

Predicts risk of all-cause death, HF-related hospital readmission, the combination of both endpoints, and life expectancy

Predicts the risk at 1 to 5 years

Publication

Preliminary results presented at GMDS Jahrestagung 2024: https://doi.org/10.3205/24GMDS112

Installation

Using pip

pip install acribis-scores

From source

# Clone repository
git clone [e-mail]:IMI-HD/acribis_scores_python.git
cd acribis_scores_python

Windows

# Build
py -m build

# Install
py -m pip install dist/acribis_scores-<version>-py3-none-any.whl

# Run simple gui
py src/demo/gui.py

Linux

# Build
python3 -m build

# Install
python3 -m pip install dist/acribis_scores-<version>-py3-none-any.whl

# Run simple gui
python3 src/demo/gui.py

Testing

For automated testing against the R implementation and online/Excel calculators (if available) of these scores, clone the R project and follow the instructions there (https://github.com/IMI-HD/acribis_scores_r). Run the R Shiny app locally on port 80 and keep it running while executing the tests:

# In the acribis_scores_r folder
shiny::runApp("app/demo_gui.R", port=80)

Install the package with additional optional dependencies and run the tests:

# Install with optional test dependencies
pip install acribis-scores[test]

Windows

# In the acribis_scores_python folder
py -m unittest discover tests

Linux

# In the acribis_scores_python folder
python3 -m unittest discover tests