Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Lumen centerline detection in complex coronary angiograms

M Sonka1, M D Winniford, X Zhang

  • 1Department of Electrical and Computer Engineering, University of Iowa, Iowa City 52242.

IEEE Transactions on Bio-Medical Engineering
|June 1, 1994
PubMed
Summary

This study introduces a new way to detect the centerline of blood vessels in complex coronary angiograms. Traditional methods often struggle with accuracy in these cases, especially when vessels have irregular shapes or poor contrast. The new approach mimics how doctors visually identify the centerline by detecting both vessel borders at the same time. The researchers tested this method against two conventional techniques using 89 complex images. They compared the results using five objective measures and a subjective evaluation by an experienced cardiologist. The new method outperformed the others in all five measures and was rated higher in quality. This suggests that the new approach could improve the accuracy of automated diagnostics in coronary imaging.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Risk Stratification and Coronary Optical Coherence Tomography Findings in Asymptomatic Patients With Type 1 Diabetes Mellitus.

Physiological research·2025
Same author

Absolute emission intensities of the gamma rays from the decay of <sup>224</sup>Ra and <sup>212</sup>Pb progenies and the half-life of the <sup>212</sup> Pb decay.

Applied radiation and isotopes : including data, instrumentation and methods for use in agriculture, industry and medicine·2024
Same author

A new determination of the <sup>68</sup>Ga half-life and evaluation of literature data.

Applied radiation and isotopes : including data, instrumentation and methods for use in agriculture, industry and medicine·2023
Same author

Determination of the Terbium-152 half-life from mass-separated samples from CERN-ISOLDE and assessment of the radionuclide purity.

Applied radiation and isotopes : including data, instrumentation and methods for use in agriculture, industry and medicine·2023
Same author

Analysis of a neutron-induced conversion electron spectrum of gadolinium.

Applied radiation and isotopes : including data, instrumentation and methods for use in agriculture, industry and medicine·2023
Same author

Half-life determination of <sup>155</sup>Tb from mass-separated samples produced at CERN-MEDICIS.

Applied radiation and isotopes : including data, instrumentation and methods for use in agriculture, industry and medicine·2022

Area of Science:

  • Medical imaging analysis
  • Cardiovascular diagnostics
  • Computer-aided detection in radiology

Background:

Accurate lumen centerline detection in coronary angiograms is essential for diagnostic accuracy in cardiovascular imaging. Prior research has shown that traditional methods often struggle with complex anatomical structures and overlapping vessel segments. These limitations may hinder precise measurements of vessel dimensions and stenosis severity. Established techniques typically rely on sequential border detection or single-point estimation strategies. However, such approaches may fail in regions with poor contrast or irregular vessel shapes. This gap motivated the development of a new method that mimics clinical visual interpretation. The need for a more reliable automated system is evident in clinical settings where precision is critical. No prior work had resolved the challenge of simultaneous border identification in complex coronary images.

Purpose Of The Study:

The goal of this research was to introduce a novel centerline detection method that mirrors clinician judgment in coronary angiograms. The specific problem addressed is the inaccuracy of conventional automated methods in complex vessel segments. The motivation stems from the limitations observed in traditional sequential border detection techniques. The study aimed to compare the new approach with two established methods using both objective and subjective criteria. The primary objective was to assess whether simultaneous border detection improves centerline accuracy. The secondary aim was to validate the method's performance in a large dataset of complex images. The researchers proposed that this approach could enhance diagnostic reliability in clinical practice. The study's design focused on evaluating both quantitative and qualitative aspects of centerline detection.

Keywords:
coronary angiogramcenterline detectionmedical imagingcardiovascular diagnostics

Frequently Asked Questions

The new method simultaneously detects left and right coronary borders to estimate the centerline, whereas conventional methods use sequential or single-point estimation strategies.

Five indices of centerline position and orientation were used to evaluate the accuracy of the detected centerlines in the 89 complex coronary images.

The researchers propose that simultaneous border detection improves accuracy by mimicking how clinicians visually identify the midline between vessel borders.

An experienced cardiologist evaluated the centerlines without knowing the detection method used, confirming the new method's superior quality (p < 0.001).

Related Experiment Videos

Main Methods:

The new method simultaneously identifies left and right coronary borders to estimate the lumen centerline. This approach emulates how clinicians visually determine the midline between vessel borders. The algorithm was tested on 89 complex coronary angiograms with observer-identified centerlines. Two conventional centerline detection methods were used for comparison purposes. Five objective indices measured centerline position and orientation accuracy. An experienced cardiologist performed a subjective quality assessment of the centerlines. The observer was blinded to the detection method used in each case. The performance of the three methods was compared using statistical analysis of the indices.

Main Results:

The new method produced centerlines that were significantly more accurate than conventional approaches. Objective indices showed statistically significant improvements across all five parameters (p < 0.001). The mean deviation from observer-defined centerlines was lowest for the new method. Subjective evaluation by a cardiologist confirmed the superior quality of the new method's centerlines. The conventional methods exhibited higher variability in complex vessel segments. The new approach demonstrated consistent performance across all 89 images tested. The results suggest that simultaneous border detection enhances centerline accuracy in challenging cases. The statistical significance of the findings supports the method's clinical potential.

Conclusions:

The authors propose that their new method improves centerline detection accuracy in complex coronary angiograms. The results suggest that simultaneous border identification enhances diagnostic reliability. The study's findings indicate that this approach outperforms conventional methods in both objective and subjective assessments. The researchers propose that this method could be integrated into clinical imaging systems. The performance improvements observed in complex cases support the method's practical value. The authors suggest that the approach better reflects clinical visual interpretation. The results may inform future developments in automated coronary imaging analysis. The study's conclusions are based on direct comparisons with established methods.

The new method was tested on 89 complex coronary angiograms with observer-defined centerlines for comparison.

The authors propose that the method could enhance diagnostic accuracy in clinical settings by providing more reliable centerline detection in complex vessel segments.