MétaCan
Menu
Back to cohort
Record W4292775080 · doi:10.1093/ehjci/jeac141.012

The utility of a fully automated cardiac magnetic resonance post-processing tool and radiomics algorithm to non-invasively classify patients with or without significant coronary artery stenosis

2022· article· en· W4292775080 on OpenAlexaffabout
Elizabeth Hillier, Mitchel Benovoy, Matthias G. Friedrich

Bibliographic record

VenueEuropean Heart Journal - Cardiovascular Imaging · 2022
Typearticle
Languageen
FieldMedicine
TopicCardiac Imaging and Diagnostics
Canadian institutionsMcGill University Health Centre
Fundersnot available
KeywordsMedicineCoronary artery diseaseCardiologyInternal medicineStenosisMagnetic resonance imagingArteryRadiologyAlgorithm

Abstract

fetched live from OpenAlex

Abstract Funding Acknowledgements Type of funding sources: Public hospital(s). Main funding source(s): Research Institute of the McGill University Health Centre. Background Oxygenation-Sensitive Cardiac Magnetic Resonance (OS-CMR) has emerged as a powerful tool to investigate the underlying physiology of a number of disease states through the assessment of tissue oxygenation status with myocardial oxygenation reserve and functional kinetics of the myocardium with strain. Recently, the analysis of CMR scans with radiomics algorithms has demonstrated to have superior diagnostic accuracy over standard analysis and reporting methods. As up to half of patients undergoing coronary angiography are found to have ischemia with no significant coronary artery obstruction, a non-invasive diagnostic test that can help to more accurately stratify patients presenting with symptoms of ischemia as having significant or no significant coronary artery disease (CAD) would be of great clinical use. Methods We analyzed 49 patients (38 with significant and 15 without significant obstructive CAD) with a positive stress test and coronary angiography. All participants underwent a non-contrast CMR exam on a clinical 3T MRI system (Magnetom Skyra™, Siemens Healthineers, Erlangen, Germany) within one week of the coronary angiography. Long axis cine CMR for ventricular morphology, volumes, function including strain, and short axis OS-CMR images were acquired (total image acquisition time less than 15min). The images were imported and analyzed with a fully automated analysis package including an advanced machine learning algorithm (cvi42™ Cardiom prototype (Circle Cardiovascular Imaging, Alberta, Canada). Per participant, 602 discrete data points per participant are extracted. A 75% or higher degree of coronary artery stenosis on Quantitative Coronary Angiography (QCA) was used as the ground truth and classified as either 1 vessel disease (VD), 2VD, 3VD, or no significant coronary artery obstruction. Results Fig. 1 shows the top discriminative features as identified by the algorithm: OS-CMR derived marker: 1) myocardial oxygen saturation (LV SVO2), 2) myocardial oxygenation in response to hyperventilation stress (MORS), and 3) epicardial myocardial oxygenation reserve (MORE). Other predictive markers were: Peak Systolic Radial Strain, treatment with calcium channel blockers, presence of cerebrovascular disease, and hypertension. The algorithm showed a 73% classification accuracy of identifying patients with or without obstructive coronary artery stenosis. Conclusion In this proof-of-concept analysis, a fully automated post-processing tool and radiomics algorithm has demonstrated the potential to accurately predict clinical classification in patients with and without significant CAD with a non-invasive, contrast-free CMR protocol. Further training and refinement of analysis algorithms are likely to further enhance the predictive value.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.003
metaresearch head score (Gemma)0.007
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.004
Threshold uncertainty score0.015

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0030.007
Meta-epidemiology (narrow)0.0010.000
Meta-epidemiology (broad)0.0010.001
Bibliometrics0.0030.001
Science and technology studies0.0000.000
Scholarly communication0.0010.001
Open science0.0010.001
Research integrity0.0010.001
Insufficient payload (model declined to judge)0.0040.002

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.013
GPT teacher head0.235
Teacher spread0.222 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designObservational
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2022
Admission routes2
Has abstractyes

Explore more

Same venueEuropean Heart Journal - Cardiovascular ImagingSame topicCardiac Imaging and DiagnosticsFrench-language works237,207