Top

20-11-2023 | Imaging Informatics and Artificial Intelligence

Machine learning detects symptomatic patients with carotid plaques based on 6-type calcium configuration classification on CT angiography

Authors: Francesco Pisu, Hui Chen, Bin Jiang, Guangming Zhu, Marco Virgilio Usai, Martin Austermann, Yousef Shehada, Elias Johansson, Jasjit Suri, Giuseppe Lanzino, John Benson, Valentina Nardi, Amir Lerman, Max Wintermark, Luca Saba

Published in: European Radiology

Abstract

Objectives

While the link between carotid plaque composition and cerebrovascular vascular (CVE) events is recognized, the role of calcium configuration remains unclear. This study aimed to develop and validate a CT angiography (CTA)–based machine learning (ML) model that uses carotid plaques 6-type calcium grading, and clinical parameters to identify CVE patients with bilateral plaques.

Material and methods

We conducted a multicenter, retrospective diagnostic study (March 2013–May 2020) approved by the institutional review board. We included adults (18 +) with bilateral carotid artery plaques, symptomatic patients having recently experienced a carotid territory ischemic event, and asymptomatic patients either after 3 months from symptom onset or with no such event. Four ML models (clinical factors, calcium configurations, and both with and without plaque grading [ML-All-G and ML-All-NG]) and logistic regression on all variables identified symptomatic patients. Internal validation assessed discrimination and calibration. External validation was also performed, and identified important variables and causes of misclassifications.

Results

We included 790 patients (median age 72, IQR [61–80], 42% male, 64% symptomatic) for training and internal validation, and 159 patients (age 68 [63–76], 36% male, 39% symptomatic) for external testing. The ML-All-G model achieved an area-under-ROC curve of 0.71 (95% CI 0.58–0.78; p < .001) and sensitivity 80% (79–81). Performance was comparable on external testing. Calcified plaque, especially the positive rim sign on the right artery in older and hyperlipidemic patients, had a major impact on identifying symptomatic patients.

Conclusion

The developed model can identify symptomatic patients using plaques calcium configuration data and clinical information with reasonable diagnostic accuracy.

Clinical relevance

The analysis of the type of calcium configuration in carotid plaques into 6 classes, combined with clinical variables, allows for an effective identification of symptomatic patients.

Key Points

• While the association between carotid plaques composition and cerebrovascular events is recognized, the role of calcium configuration remains unclear.

• Machine learning of 6-type plaque grading can identify symptomatic patients. Calcified plaques on the right artery, advanced age, and hyperlipidemia were the most important predictors.

• Fast acquisition of CTA enables rapid grading of plaques upon the patient’s arrival at the hospital, which streamlines the diagnosis of symptoms using ML.

Graphical Abstract

Available only for authorised users

Tsao CW, Aday AW, Almarzooq ZI et al (2022) Heart disease and stroke statistics-2022 update: a report from the American Heart Association. Circulation 145:e153–e639. https://doi.org/10.1161/CIR.0000000000001052 CrossRefPubMed

Kleindorfer DO, Towfighi A, Chaturvedi S et al (2021) 2021 guideline for the prevention of stroke in patients with stroke and transient ischemic attack: a guideline from the American heart association/American stroke association. Stroke 52:e364–e467. https://doi.org/10.1161/STR.0000000000000375 CrossRefPubMed

Saba L, Saam T, Jäger HR et al (2019) Imaging biomarkers of vulnerable carotid plaques for stroke risk prediction and their potential clinical implications. Lancet Neurol 18:559–572. https://doi.org/10.1016/S1474-4422(19)30035-3 CrossRefPubMed

Fisher M, Martin A, Cosgrove M, Norris JW (1993) The NASCET-ACAS plaque project. North American Symptomatic Carotid Endarterectomy Trial. Asymptomatic Carotid Atherosclerosis Study. Stroke 24:I24-5 (discussion I31-2) PubMed

Waksman R, Torguson R (2021) The vulnerable plaque detected: time to consider treatment. Lancet 397:943–945. https://doi.org/10.1016/S0140-6736(21)00504-3 CrossRefPubMed

Schindler A, Schinner R, Altaf N et al (2020) Prediction of stroke risk by detection of hemorrhage in carotid plaques: Meta-analysis of individual patient data. JACC Cardiovasc Imaging 13:395–406. https://doi.org/10.1016/j.jcmg.2019.03.028 CrossRefPubMed

Saba L, Nardi V, Cau R et al (2022) Carotid artery plaque calcifications: lessons from histopathology to diagnostic imaging. Stroke 53:290–297. https://doi.org/10.1161/STROKEAHA.121.035692 CrossRefPubMed

Nandalur KR, Hardie AD, Raghavan P et al (2007) Composition of the stable carotid plaque: insights from a multidetector computed tomography study of plaque volume. Stroke 38:935–940. https://doi.org/10.1161/01.STR.0000257995.74834.92 CrossRefPubMedPubMedCentral

Benson JC, Nardi V, Madhavan AA et al (2022) Reassessing the carotid artery plaque “rim sign” on CTA: a new analysis with histopathologic confirmation. AJNR Am J Neuroradiol 43:429–434. https://doi.org/10.3174/ajnr.A7443 CrossRefPubMedPubMedCentral

10.

Saba L, Chen H, Cau R et al (2022) Impact analysis of different CT configurations of carotid artery plaque calcifications on cerebrovascular events. AJNR Am J Neuroradiol 43:272–279. https://doi.org/10.3174/ajnr.A7401 CrossRefPubMedPubMedCentral

11.

Touboul P-J, Hennerici MG, Meairs S et al (2012) Mannheim Carotid Intima-Media Thickness and Plaque Consensus (2004–2006–2011). Cerebrovasc Dis 34:290–296. https://doi.org/10.1159/000343145 CrossRefPubMed

12.

Ferguson GG, Eliasziw M, Barr HWK et al (1999) The North American Symptomatic Carotid Endarterectomy Trial. Stroke 30:1751–1758. https://doi.org/10.1161/01.STR.30.9.1751 CrossRefPubMed

13.

Lou Y, Caruana R, Gehrke J, Hooker G (2013) Accurate intelligible models with pairwise interactions. In: KDD’ 13: The 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. ACM, pp 623–631

14.

Yao H, Golbus JR, Gryak J et al (2022) Identifying potential candidates for advanced heart failure therapies using an interpretable machine learning algorithm. J Heart Lung Transplant S105324982202109X. https://doi.org/10.1016/j.healun.2022.08.028

15.

Qu Y, Deng X, Lin S et al (2022) Using innovative machine learning methods to screen and identify predictors of congenital heart diseases. Front Cardiovasc Med 8:797002. https://doi.org/10.3389/fcvm.2021.797002 CrossRefPubMedPubMedCentral

16.

Friedman JH (2001) Greedy function approximation: a gradient boosting machine. Ann Stat 29:1189–1232. https://doi.org/10.1214/aos/1013203451 CrossRef

17.

Molinaro AM, Simon R, Pfeiffer RM (2005) Prediction error estimation: a comparison of resampling methods. Bioinformatics 21:3301–3307. https://doi.org/10.1093/bioinformatics/bti499 CrossRefPubMed

18.

DeLong ER, DeLong DM, Clarke-Pearson DL (1988) Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 44:837–845. https://doi.org/10.2307/2531595 CrossRefPubMed

19.

Eisenmenger LB, Aldred BW, Kim S-E et al (2016) Prediction of carotid intraplaque hemorrhage using adventitial calcification and plaque thickness on CTA. AJNR Am J Neuroradiol 37:1496. https://doi.org/10.3174/ajnr.A4765 CrossRefPubMedPubMedCentral

20.

Yang J, Pan X, Zhang B et al (2018) Superficial and multiple calcifications and ulceration associate with intraplaque hemorrhage in the carotid atherosclerotic plaque. Eur Radiol 28:4968–4977. https://doi.org/10.1007/s00330-018-5535-7 CrossRefPubMedPubMedCentral

21.

Cademartiri F, Meloni A, Pistoia L et al (2023) Dual source photon-counting computed tomography—Part II: clinical overview of neurovascular applications. J Clin Med 12. https://doi.org/10.3390/jcm12113626

22.

Futoma J, Simons M, Panch T et al (2020) The myth of generalisability in clinical research and machine learning in health care. Lancet Digit Health 2:e489–e492. https://doi.org/10.1016/S2589-7500(20)30186-2 CrossRefPubMedPubMedCentral

Title: Machine learning detects symptomatic patients with carotid plaques based on 6-type calcium configuration classification on CT angiography
Authors: Francesco Pisu
Hui Chen
Bin Jiang
Guangming Zhu
Marco Virgilio Usai
Martin Austermann
Yousef Shehada
Elias Johansson
Jasjit Suri
Giuseppe Lanzino
John Benson
Valentina Nardi
Amir Lerman
Max Wintermark
Luca Saba
Publication date: 20-11-2023
Publisher: Springer Berlin Heidelberg
Published in: European Radiology
Print ISSN: 0938-7994
Electronic ISSN: 1432-1084
DOI: https://doi.org/10.1007/s00330-023-10347-2

Keynote Webinar | Spotlight on Diet and Diabetes

Springer Medicine

Machine learning detects symptomatic patients with carotid plaques based on 6-type calcium configuration classification on CT angiography

Abstract

Objectives

Material and methods

Results

Conclusion

Clinical relevance

Key Points

Graphical Abstract

Keynote Webinar | Spotlight on Diet and Diabetes

Springer Medicine

Abstract

Objectives

Material and methods

Results

Conclusion

Clinical relevance

Key Points

Graphical Abstract

Please log in to get access to this content