Application of machine learning to determine top predictors of noncalcified coronary burden in psoriasis: An observational cohort study

JOURNAL OF THE AMERICAN ACADEMY OF DERMATOLOGY

卷:83

Article

Munger, Eric¹  Choi, Harry²  Dey, Amit K.²  Elnabawi, Youssef A.²  Groenendyk, Jacob W.²  Rodante, Justin²  Keel, Andrew²  Aksentijevich, Milena²  Reddy, Aarthi S.²  Khalil, Noor²  Argueta-Amaya, Jenis²  Playford, Martin P.²  Erb-Alvarez, Julie²  Tian, Xin²  Wu, Colin²  Gudjonsson, Johann E.³  Tsoi, Lam C.³  Jafri, Mohsin Saleet¹  Sandfort, Veit²  Chen, Marcus Y.²  Shah, Sanjiv J.⁴  Bluemke, David A.⁵  Lockshin, Benjamin⁶  Hasan, Ahmed²  Gelfand, Joel M.⁷  Mehta, Nehal N.² 

[1] George Mason Univ, Fairfax, VA 22030 USA

[2] NHLBI, NIH, Bethesda, MD 20892 USA

[3] Univ Michigan, Ann Arbor, MI 48109 USA

[4] Northwestern Univ, Chicago, IL 60611 USA

[5] Univ Wisconsin, Madison, WI USA

[6] DermAssociates, Silver Spring, MD USA

[7] Univ Penn, Philadelphia, PA 19104 USA

关键词: atherosclerosis;    cardiometabolic disease;    coronary artery disease;    machine learning;    psoriasis;    random forest algorithm;   

DOI  :  10.1016/j.jaad.2019.10.060

来源: Elsevier

【 摘 要 】

Background: Psoriasis is associated with elevated risk of heart attack and increased accumulation of subclinical noncalcified coronary burden by coronary computed tomography angiography (CCTA). Machine learning algorithms have been shown to effectively analyze well-characterized data sets. Objective: In this study, we used machine learning algorithms to determine the top predictors of noncalcified coronary burden by CCTA in psoriasis. Methods: The analysis included 263 consecutive patients with 63 available variables from the Psoriasis Atherosclerosis Cardiometabolic Initiative. The random forest algorithm was used to determine the top predictors of noncalcified coronary burden by CCTA. We evaluated our results using linear regression models. Results: Using the random forest algorithm, we found that the top 10 predictors of noncalcified coronary burden were body mass index, visceral adiposity, total adiposity, apolipoprotein A1, high-density lipoprotein, erythrocyte sedimentation rate, subcutaneous adiposity, small low-density lipoprotein particle, cholesterol efflux capacity and the absolute granulocyte count. Linear regression of noncalcified coronary burden yielded results consistent with our machine learning output. Limitation: We were unable to provide external validation and did not study cardiovascular events. Conclusion: Machine learning methods identified the top predictors of noncalcified coronary burden in psoriasis. These factors were related to obesity, dyslipidemia, and inflammation, showing that these are important targets when treating comorbidities in psoriasis.

【 授权许可】

Free