Assessing the diagnostic accuracy and prognostic utility of artificial intelligence detection and grading of coronary artery calcification on nongated computed tomography (CT) thorax

B. Shear; J. Graby; D. Murphy; K. Strong; A. Khavandi; T. A. Burnett; P. F.P. Charters; J. C.L. Rodrigues

doi:10.1016/j.crad.2025.106909

Assessing the diagnostic accuracy and prognostic utility of artificial intelligence detection and grading of coronary artery calcification on nongated computed tomography (CT) thorax

B. Shear, J. Graby, D. Murphy, K. Strong, A. Khavandi, T. A. Burnett, P. F.P. Charters, J. C.L. Rodrigues

Department for Health

Research output: Contribution to journal › Article › peer-review

Abstract

Aims: This study assessed the diagnostic accuracy and prognostic implications of an artificial intelligence (AI) tool for coronary artery calcification (CAC) assessment on nongated, noncontrast thoracic computed tomography (CT).

Materials and Methods: A single-centre retrospective analysis of 75 consecutive patients per age group (<40, 40-49, 50-59, 60-69, 70-79, 80-89, and ≥90 years) undergoing non-gated, non-contrast CT (January-December 2015) was conducted. AI analysis reported CAC presence and generated an Agatston score, and the performance was compared with baseline CT reports and a dedicated radiologist re-review. Interobserver variability between AI and radiologist assessments was measured using Cohen's κ. All-cause mortality was recorded, and its association with AI-detected CAC was tested.

Results: A total of 291 patients (mean age: 64 ± 19, 51% female) were included, with 80% (234/291) of AI reports passing radiologist quality assessment. CAC was reported on 7% (17/234) of initial clinical reports, 58% (135/234) on radiologist re-review, and 57% (134/234) by AI analysis. After manual quality assurance (QA) assessment, the AI tool demonstrated high sensitivity (96%), specificity (96%), positive predictive value (95%), and negative predictive value (97%) for CAC detection compared with radiologist re-review. Interobserver agreement was strong for CAC prevalence (κ = 0.92) and moderate for severity grading (κ = 0.60). AI-detected CAC presence and severity predicted all-cause mortality (p < 0.001).

Conclusion: The AI tool exhibited feasible analysis potential for non-contrast, non-gated thoracic CTs, offering prognostic insights if integrated into routine practice. Nonetheless, manual quality assessment remains essential. Advances in knowledge: This AI tool represents a potential enhancement to CAC detection and reporting on routine noncardiac chest CT.

Original language	English
Article number	106909
Journal	Clinical Radiology
Volume	85
Early online date	22 Mar 2025
DOIs	https://doi.org/10.1016/j.crad.2025.106909
Publication status	Published - 30 Jun 2025

ASJC Scopus subject areas

Radiology Nuclear Medicine and imaging

Access to Document

10.1016/j.crad.2025.106909

Cite this

Shear, B., Graby, J., Murphy, D., Strong, K., Khavandi, A., Burnett, T. A., Charters, P. F. P., & Rodrigues, J. C. L. (2025). Assessing the diagnostic accuracy and prognostic utility of artificial intelligence detection and grading of coronary artery calcification on nongated computed tomography (CT) thorax. Clinical Radiology, 85, Article 106909. https://doi.org/10.1016/j.crad.2025.106909

Shear, B, Graby, J , Murphy, D, Strong, K, Khavandi, A, Burnett, TA, Charters, PFP & Rodrigues, JCL 2025, 'Assessing the diagnostic accuracy and prognostic utility of artificial intelligence detection and grading of coronary artery calcification on nongated computed tomography (CT) thorax', Clinical Radiology, vol. 85, 106909. https://doi.org/10.1016/j.crad.2025.106909

@article{92eb1bc3880d441cb2c6b915ec749bab,

title = "Assessing the diagnostic accuracy and prognostic utility of artificial intelligence detection and grading of coronary artery calcification on nongated computed tomography (CT) thorax",

abstract = "Aims: This study assessed the diagnostic accuracy and prognostic implications of an artificial intelligence (AI) tool for coronary artery calcification (CAC) assessment on nongated, noncontrast thoracic computed tomography (CT). Materials and Methods: A single-centre retrospective analysis of 75 consecutive patients per age group (<40, 40-49, 50-59, 60-69, 70-79, 80-89, and ≥90 years) undergoing non-gated, non-contrast CT (January-December 2015) was conducted. AI analysis reported CAC presence and generated an Agatston score, and the performance was compared with baseline CT reports and a dedicated radiologist re-review. Interobserver variability between AI and radiologist assessments was measured using Cohen's κ. All-cause mortality was recorded, and its association with AI-detected CAC was tested. Results: A total of 291 patients (mean age: 64 ± 19, 51% female) were included, with 80% (234/291) of AI reports passing radiologist quality assessment. CAC was reported on 7% (17/234) of initial clinical reports, 58% (135/234) on radiologist re-review, and 57% (134/234) by AI analysis. After manual quality assurance (QA) assessment, the AI tool demonstrated high sensitivity (96%), specificity (96%), positive predictive value (95%), and negative predictive value (97%) for CAC detection compared with radiologist re-review. Interobserver agreement was strong for CAC prevalence (κ = 0.92) and moderate for severity grading (κ = 0.60). AI-detected CAC presence and severity predicted all-cause mortality (p < 0.001). Conclusion: The AI tool exhibited feasible analysis potential for non-contrast, non-gated thoracic CTs, offering prognostic insights if integrated into routine practice. Nonetheless, manual quality assessment remains essential. Advances in knowledge: This AI tool represents a potential enhancement to CAC detection and reporting on routine noncardiac chest CT.",

author = "B. Shear and J. Graby and D. Murphy and K. Strong and A. Khavandi and Burnett, {T. A.} and Charters, {P. F.P.} and Rodrigues, {J. C.L.}",

year = "2025",

month = jun,

day = "30",

doi = "10.1016/j.crad.2025.106909",

language = "English",

volume = "85",

journal = "Clinical Radiology",

issn = "0009-9260",

publisher = "W.B. Saunders Ltd",

}

TY - JOUR

T1 - Assessing the diagnostic accuracy and prognostic utility of artificial intelligence detection and grading of coronary artery calcification on nongated computed tomography (CT) thorax

AU - Shear, B.

AU - Graby, J.

AU - Murphy, D.

AU - Strong, K.

AU - Khavandi, A.

AU - Burnett, T. A.

AU - Charters, P. F.P.

AU - Rodrigues, J. C.L.

PY - 2025/6/30

Y1 - 2025/6/30

N2 - Aims: This study assessed the diagnostic accuracy and prognostic implications of an artificial intelligence (AI) tool for coronary artery calcification (CAC) assessment on nongated, noncontrast thoracic computed tomography (CT). Materials and Methods: A single-centre retrospective analysis of 75 consecutive patients per age group (<40, 40-49, 50-59, 60-69, 70-79, 80-89, and ≥90 years) undergoing non-gated, non-contrast CT (January-December 2015) was conducted. AI analysis reported CAC presence and generated an Agatston score, and the performance was compared with baseline CT reports and a dedicated radiologist re-review. Interobserver variability between AI and radiologist assessments was measured using Cohen's κ. All-cause mortality was recorded, and its association with AI-detected CAC was tested. Results: A total of 291 patients (mean age: 64 ± 19, 51% female) were included, with 80% (234/291) of AI reports passing radiologist quality assessment. CAC was reported on 7% (17/234) of initial clinical reports, 58% (135/234) on radiologist re-review, and 57% (134/234) by AI analysis. After manual quality assurance (QA) assessment, the AI tool demonstrated high sensitivity (96%), specificity (96%), positive predictive value (95%), and negative predictive value (97%) for CAC detection compared with radiologist re-review. Interobserver agreement was strong for CAC prevalence (κ = 0.92) and moderate for severity grading (κ = 0.60). AI-detected CAC presence and severity predicted all-cause mortality (p < 0.001). Conclusion: The AI tool exhibited feasible analysis potential for non-contrast, non-gated thoracic CTs, offering prognostic insights if integrated into routine practice. Nonetheless, manual quality assessment remains essential. Advances in knowledge: This AI tool represents a potential enhancement to CAC detection and reporting on routine noncardiac chest CT.

AB - Aims: This study assessed the diagnostic accuracy and prognostic implications of an artificial intelligence (AI) tool for coronary artery calcification (CAC) assessment on nongated, noncontrast thoracic computed tomography (CT). Materials and Methods: A single-centre retrospective analysis of 75 consecutive patients per age group (<40, 40-49, 50-59, 60-69, 70-79, 80-89, and ≥90 years) undergoing non-gated, non-contrast CT (January-December 2015) was conducted. AI analysis reported CAC presence and generated an Agatston score, and the performance was compared with baseline CT reports and a dedicated radiologist re-review. Interobserver variability between AI and radiologist assessments was measured using Cohen's κ. All-cause mortality was recorded, and its association with AI-detected CAC was tested. Results: A total of 291 patients (mean age: 64 ± 19, 51% female) were included, with 80% (234/291) of AI reports passing radiologist quality assessment. CAC was reported on 7% (17/234) of initial clinical reports, 58% (135/234) on radiologist re-review, and 57% (134/234) by AI analysis. After manual quality assurance (QA) assessment, the AI tool demonstrated high sensitivity (96%), specificity (96%), positive predictive value (95%), and negative predictive value (97%) for CAC detection compared with radiologist re-review. Interobserver agreement was strong for CAC prevalence (κ = 0.92) and moderate for severity grading (κ = 0.60). AI-detected CAC presence and severity predicted all-cause mortality (p < 0.001). Conclusion: The AI tool exhibited feasible analysis potential for non-contrast, non-gated thoracic CTs, offering prognostic insights if integrated into routine practice. Nonetheless, manual quality assessment remains essential. Advances in knowledge: This AI tool represents a potential enhancement to CAC detection and reporting on routine noncardiac chest CT.

UR - http://www.scopus.com/inward/record.url?scp=105003713684&partnerID=8YFLogxK

U2 - 10.1016/j.crad.2025.106909

DO - 10.1016/j.crad.2025.106909

M3 - Article

AN - SCOPUS:105003713684

SN - 0009-9260

VL - 85

JO - Clinical Radiology

JF - Clinical Radiology

M1 - 106909

ER -

Assessing the diagnostic accuracy and prognostic utility of artificial intelligence detection and grading of coronary artery calcification on nongated computed tomography (CT) thorax

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this