Machine-learning phenotypic classification of bicuspid aortopathy

Autor: Jay J. Idrees, John Ehrlinger, Patrick Collier, Ashley M. Lowry, Eric E. Roselli, Yuanjia Zhu, Eugene H. Blackstone, Charles M. Wojnarski, Brian P. Griffin, Lars G. Svensson, Theresa Carnes, Bruce W. Lytle
Rok vydání: 2018
Předmět:
Adult
Male
Pulmonary and Respiratory Medicine
medicine.medical_specialty
Computed Tomography Angiography
Aortic Valve Insufficiency
Heart Valve Diseases
Aorta
Thoracic
030204 cardiovascular system & hematology
Aortography
Pattern Recognition
Automated
Machine Learning
03 medical and health sciences
0302 clinical medicine
Bicuspid aortic valve
Aneurysm
Bicuspid Aortic Valve Disease
Predictive Value of Tests
medicine.artery
Internal medicine
Ascending aorta
medicine
Brachiocephalic artery
Humans
Diagnosis
Computer-Assisted
Aged
Aorta
medicine.diagnostic_test
business.industry
Reproducibility of Results
Magnetic resonance imaging
Aortic Valve Stenosis
Middle Aged
Sinus of Valsalva
medicine.disease
Phenotype
Pathophysiology
Aortic Aneurysm
Cross-Sectional Studies
030228 respiratory system
Aortic Valve
cardiovascular system
Cardiology
Radiographic Image Interpretation
Computer-Assisted
Female
Surgery
Cardiology and Cardiovascular Medicine
business
Zdroj: The Journal of Thoracic and Cardiovascular Surgery. 155:461-469.e4
ISSN: 0022-5223
Popis: Bicuspid aortic valves (BAV) are associated with incompletely characterized aortopathy. Our objectives were to identify distinct patterns of aortopathy using machine-learning methods and characterize their association with valve morphology and patient characteristics.We analyzed preoperative 3-dimensional computed tomography reconstructions for 656 patients with BAV undergoing ascending aorta surgery between January 2002 and January 2014. Unsupervised partitioning around medoids was used to cluster aortic dimensions. Group differences were identified using polytomous random forest analysis.Three distinct aneurysm phenotypes were identified: root (n = 83; 13%), with predominant dilatation at sinuses of Valsalva; ascending (n = 364; 55%), with supracoronary enlargement rarely extending past the brachiocephalic artery; and arch (n = 209; 32%), with aortic arch dilatation. The arch phenotype had the greatest association with right-noncoronary cusp fusion: 29%, versus 13% for ascending and 15% for root phenotypes (P .0001). Severe valve regurgitation was most prevalent in root phenotype (57%), followed by ascending (34%) and arch phenotypes (25%; P .0001). Aortic stenosis was most prevalent in arch phenotype (62%), followed by ascending (50%) and root phenotypes (28%; P .0001). Patient age increased as the extent of aneurysm became more distal (root, 49 years; ascending, 53 years; arch, 57 years; P .0001), and root phenotype was associated with greater male predominance compared with ascending and arch phenotypes (94%, 76%, and 70%, respectively; P .0001). Phenotypes were visually recognizable with 94% accuracy.Three distinct phenotypes of bicuspid valve-associated aortopathy were identified using machine-learning methodology. Patient characteristics and valvular dysfunction vary by phenotype, suggesting that the location of aortic pathology may be related to the underlying pathophysiology of this disease.
Databáze: OpenAIRE
Externí odkaz: