Automated segmentation and quantification of aortic calcification at abdominal CT: application of a deep learning-based algorithm to a longitudinal screening cohort

ObjectiveTo investigate an automated aortic calcium segmentation and scoring tool at abdominal CT in an adult screening cohort.MethodsUsing instance segmentation with convolutional neural networks (Mask R-CNN), a fully automated vascular calcification algorithm was applied to a data set of 9914 non-contrast CT scans from 9032 consecutive asymptomatic adults (mean age, 57.5 ± 7.8 years; 4467 M/5447F) undergoing colonography screening. Follow-up scans were performed in a subset of 866 individuals (mean interval, 5.4 years). Automated abdominal aortic calcium volume, mass, and Agatston score were assessed. In addition, comparison was made with a separate validated semi-automated approach in a subset of 812 cases.ResultsMean values were significantly higher in males for Agatston score (924.2 ± 2066.2 vs. 564.2 ± 1484.2, p < 0.001), aortic calcium mass (222.2 ± 526.0 mg vs. 144.5 ± 405.4 mg, p < 0.001) and volume (699.4 ± 1552.4 ml vs. 426.9 ± 1115.5 HU, p < 0.001). Overall age-specific Agatston scores increased an average of 10%/year for the entire cohort; males had a larger Agatston score increase between the ages of 40 to 60 than females (91.2% vs. 75.1%, p < 0.001) and had significantly higher mean Agatston scores between ages 50 and 80 (p < 0.001). For the 812-scan subset with both automated and semi-automated methods, median difference in Agatston score was 66.4 with an r2 agreement value of 0.84. Among the 866-patient cohort with longitudinal follow-up, the average Agatston score change was 524.1 ± 1317.5 (median 130.9), reflecting a mean increase of 25.5% (median 73.6%).ConclusionThis robust, fully automated abdominal aortic calcification scoring tool allows for both individualized and population-based assessment. Such data could be automatically derived at non-contrast abdominal CT, regardless of the study indication, allowing for opportunistic assessment of cardiovascular risk.

[1]  B. Berman,et al.  Coronary artery calcium quantification using contrast-enhanced dual-energy computed tomography scans in comparison with unenhanced single-energy scans , 2018, Physics in medicine and biology.

[2]  David H. Kim,et al.  CT colonography to screen for colorectal cancer and aortic aneurysm in the Medicare population: cost-effectiveness analysis. , 2009, AJR. American journal of roentgenology.

[3]  David H. Kim,et al.  Hepatic steatosis (fatty liver disease) in asymptomatic adults identified by unenhanced low-dose CT. , 2010, AJR. American journal of roentgenology.

[4]  P. Pickhardt,et al.  Opportunistic Screening for Osteoporosis Using Abdominal Computed Tomography Scans Obtained for Other Indications , 2013, Annals of Internal Medicine.

[5]  P. Pickhardt.  Imaging and Screening for Colorectal Cancer with CT Colonography. , 2017, Radiologic clinics of North America.

[6]  Ian Graham,et al.  Relationships between body mass index, cardiovascular mortality, and risk factors: a report from the SCORE investigators , 2011, European journal of cardiovascular prevention and rehabilitation : official journal of the European Society of Cardiology, Working Groups on Epidemiology & Prevention and Cardiac Rehabilitation and Exercise Physiology.

[7]  Lawrence H. Kushi,et al.  Recent Trends in Cardiovascular Mortality in the United States and Public Health Goals. , 2016, JAMA cardiology.

[8]  Bram van Ginneken,et al.  Automatic detection of calcifications in the aorta from CT scans of the abdomen1 , 2004 .

[9]  P. Pickhardt,et al.  Does Nonenhanced CT-based Quantification of Abdominal Aortic Calcification Outperform the Framingham Risk Score in Predicting Cardiovascular Events in Asymptomatic Adults? , 2019, Radiology.

[10]  P. Pickhardt,et al.  Future Osteoporotic Fracture Risk Related to Lumbar Vertebral Trabecular Attenuation Measured at Routine Body CT , 2018, Journal of bone and mineral research : the official journal of the American Society for Bone and Mineral Research.

[11]  R. Hajar.  Framingham Contribution to Cardiovascular Disease , 2016, Heart views : the official journal of the Gulf Heart Association.

[12]  P. Mittal,et al.  Changing Abdominal Imaging Utilization Patterns: Perspectives From Medicare Beneficiaries Over Two Decades. , 2016, Journal of the American College of Radiology : JACR.

[13]  B. Chow,et al.  Quantifying Aortic Valve Calcification using Coronary Computed Tomography Angiography. , 2017, Journal of cardiovascular computed tomography.

[14]  Xin Yang,et al.  Automatic Detection and Quantification of Abdominal Aortic Calcification in Dual Energy X-ray Absorptiometry , 2016, KES.

[15]  Jackson T. Wright,et al.  2014 evidence-based guideline for the management of high blood pressure in adults: report from the panel members appointed to the Eighth Joint National Committee (JNC 8). , 2014, JAMA.

[16]  Ronald M. Summers,et al.  Pelvic artery calcification detection on CT scans using convolutional neural networks , 2017, Medical Imaging.

[17]  V. Ribeiro,et al.  Association of body mass index and visceral fat with aortic valve calcification and mortality after transcatheter aortic valve replacement: the obesity paradox in severe aortic stenosis , 2017, Diabetology & Metabolic Syndrome.

[18]  P. Pickhardt,et al.  Visceral adiposity and hepatic steatosis at abdominal CT: association with the metabolic syndrome. , 2012, AJR. American journal of roentgenology.

[19]  Perry J Pickhardt,et al.  Predicting Future Hip Fractures on Routine Abdominal CT Using Opportunistic Osteoporosis Screening Measures: A Matched Case-Control Study. , 2017, AJR. American journal of roentgenology.

[20]  P. Pickhardt,et al.  Does Nonenhanced CT-based Quantification of Abdominal Aortic Calcification Outperform the Framingham Risk Score in Predicting Cardiovascular Events in Asymptomatic Adults? , 2019, Radiology.

[21]  Jianhua Yao,et al.  Fully automated segmentation and quantification of visceral and subcutaneous fat at abdominal CT: application to a longitudinal adult screening cohort. , 2018, The British journal of radiology.

[22]  D. Lloyd‐Jones,et al.  Association of Body Mass Index With Lifetime Risk of Cardiovascular Disease and Compression of Morbidity , 2018, JAMA cardiology.

[23]  M. Joffe,et al.  Aortic calcification predicts cardiovascular events and all-cause mortality in renal transplantation. , 2008, Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association - European Renal Association.

[24]  Seong Ho Park,et al.  Natural history of hepatic steatosis: observed outcomes for subsequent liver and cardiovascular complications. , 2014, AJR. American journal of roentgenology.

[25]  Thomas Trieb,et al.  A method for calcium quantification by means of CT coronary angiography using 64-multidetector CT: very high correlation with agatston and volume scores , 2009, European Radiology.

[26]  Kaiming He,et al.  Mask R-CNN , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[27]  R. Kronmal,et al.  Thoracic aortic calcification and coronary heart disease events: the multi-ethnic study of atherosclerosis (MESA). , 2011, Atherosclerosis.

[28]  W. Zoghbi.  Cardiovascular imaging: a glimpse into the future. , 2014, Methodist DeBakey cardiovascular journal.

[29]  Nathan Lay,et al.  Population-based opportunistic osteoporosis screening: Validation of a fully automated CT tool for assessing longitudinal BMD changes. , 2019, The British journal of radiology.

[30]  Francesco Maisano,et al.  Quantification of aortic valve calcification on contrast-enhanced CT of patients prior to transcatheter aortic valve implantation. , 2017, EuroIntervention : journal of EuroPCR in collaboration with the Working Group on Interventional Cardiology of the European Society of Cardiology.

[31]  K. Bailey,et al.  Computed tomographic coronary artery calcium assessment for evaluating chest pain in the emergency department: long-term outcome of a prospective blind study. , 2010, Mayo Clinic proceedings.

[32]  Ronald M. Summers,et al.  Automated spinal column extraction and partitioning , 2006, 3rd IEEE International Symposium on Biomedical Imaging: Nano to Macro, 2006..

[33]  Nikolas Lessmann,et al.  Automatic quantification of calcifications in the coronary arteries and thoracic aorta on radiotherapy planning CT scans of Western and Asian breast cancer patients. , 2018, Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology.

[34]  M. Fornage,et al.  Heart Disease and Stroke Statistics—2017 Update: A Report From the American Heart Association , 2017, Circulation.

[35]  J. Mckenney,et al.  Executive Summary of The Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III). , 2001, JAMA.

[36]  R. Kronmal,et al.  Carotid-artery intima and media thickness as a risk factor for myocardial infarction and stroke in older adults. Cardiovascular Health Study Collaborative Research Group. , 1999, The New England journal of medicine.

[37]  Bram van Ginneken,et al.  Automatic detection of calcifications in the aorta from CT scans of the abdomen. 3D computer-aided diagnosis. , 2004, Academic radiology.

[38]  Ronald M. Summers,et al.  Atherosclerotic vascular calcification detection and segmentation on low dose computed tomography scans using convolutional neural networks , 2017, 2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017).

[39]  P. Pickhardt,et al.  Opportunistic Screening for Osteoporosis Using Body CT Scans Obtained for Other Indications: the UW Experience , 2017, Clinical Reviews in Bone and Mineral Metabolism.

[40]  J. Rumberger,et al.  A rosetta stone for coronary calcium risk stratification: agatston, volume, and mass scores in 11,490 individuals. , 2003, AJR. American journal of roentgenology.

[41]  Raúl San José Estépar,et al.  Automated quantitative 3D analysis of aorta size, morphology, and mural calcification distributions. , 2015, Medical physics.

[42]  Michael Pignone,et al.  Using the coronary artery calcium score to predict coronary heart disease events: a systematic review and meta-analysis. , 2004, Archives of internal medicine.

[43]  Ronald M. Summers,et al.  A Semi-Supervised CNN Learning Method with Pseudo-class Labels for Atherosclerotic Vascular Calcification Detection , 2019, 2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019).