Related Experiment Video
Updated: May 5, 2026
Implantation of Total Artificial Heart in Congenital Heart Disease
Published on: July 18, 2014
Algorithmic complexity stratification for congenital heart disease patients
Jason Chami1, Geoff Strange2,3, David Baker4
1Sydney Medical School, The University of Sydney, Camperdown, Australia.
Insights
An algorithm can now automatically score the complexity of congenital heart disease (CHD) using diagnosis lists. This automates a process previously requiring manual input, enabling large-scale CHD research and improved health service planning.
Area of Science:
- Cardiology
- Medical Informatics
- Health Services Research
Background:
- Congenital Heart Disease (CHD) comprises diverse diagnoses requiring complexity stratification for clinical guidelines, health services planning, and research.
- Current methods for assigning CHD complexity scores necessitate manual specialist input, limiting scalability and data completeness in large databases.
- Existing scoring systems are prone to becoming outdated with revised clinical guidelines.
Purpose of the Study:
- To develop and validate an algorithm for automatically assigning complexity scores to Congenital Heart Disease patients based solely on their diagnosis lists.
- To enable efficient and consistent complexity scoring for large-scale Congenital Heart Disease datasets.
- To facilitate research into the management, outcomes, and burden of Congenital Heart Disease.
Main Methods:
- Utilized up-to-date European Society of Cardiology guidelines to construct an algorithm for CHD complexity scoring.
- Algorithmically assigned complexity scores using patient diagnosis lists.
- Validated algorithm performance against manual complexity scores assigned by two independent Congenital Heart Disease specialists.
Main Results:
- The algorithm achieved 96% accuracy when compared against specialist consensus on CHD complexity scores.
- Specialist agreement on complexity scores occurred in 68% of cases overall, and 79% for moderate or complex CHD.
- Algorithm limitations were identified in cases with non-specific diagnoses (e.g., unspecified septal defects) or procedure-dependent complexity.
Conclusions:
- An algorithm can successfully determine CHD complexity scores from diagnosis lists for a majority of patients.
- Automated complexity scoring can be implemented in large CHD databases, such as the Registry of the Congenital Heart Alliance of Australia and New Zealand.
- This approach will significantly enhance targeted research on CHD management, outcomes, and patient burden.
Background:
Congenital Heart Disease (CHD) encompasses a huge variety of rare diagnoses that range in complexity and comorbidity. To help build clinical guidelines, plan health services and conduct statistically powerful research on such a disparate set of diseases there have been various attempts to group pathologies into mild, moderate, or severe disease. So far, however, these complexity scores have required manual specialist input for every case, and are therefore missing in large databases where this is impractical, or quickly outdated when guidelines are revised.
Methods:
We used the up-to-date European Society of Cardiology guidelines to create an algorithm to assign complexity scores to CHD patients using only their diagnosis list. Two CHD specialists then independently assigned complexity scores to a random sample of patients.
Results:
Our algorithm was 96% accurate where both specialists agreed on a complexity score; this occurred 68% of the time overall, and 79% of the time in moderate or complex CHD. The algorithm "failed" mainly when diagnoses were insufficiently specific, usually for septal defects (where size was unspecified), or where complexity depends on the procedure performed (e.g. atrial/arterial switch for transposition of the great arteries).
Conclusions:
We were able to algorithmically determine the complexity scores of a majority of patients with CHD based on their diagnosis list alone. This could allow for automatic complexity scoring of most patients in large CHD databases, for example our own Registry of the Congenital Heart Alliance of Australia and New Zealand. This will facilitate targeted research into the management, outcomes and burden of CHD.
Related Concept Videos
01:30Pulmonary Hypertension: Classification and Pathogenesis
There are various classifications for PH, each relating to different underlying causes and also...
01:23Mitral Stenosis II: Clinical features and Diagnostic Tests
01:30Heart Failure IV: Classification and Diagnostic Evaluation
01:30Cardiomyopathy II: Dilated Cardiomyopathy
01:29Cardiomyopathy III: Hypertrophic Cardiomyopathy
01:29Cardiomyopathy V: Interprofessional Care