Self-supervised Motion Descriptor for Cardiac Phase Detection in 4D CMR Based on Discrete Vector Field Estimations

Koehler, Sven; Hussain, Tarique; Hussain, Hamza; Daniel, Young,; Sarikouch, Samir; Pickardt, Thomas; Greil, Gerald; Engelhardt, Sandy

doi:10.1007/978-3-031-23443-9_7

Cited by 5 publications

(15 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Figure 2 ) identifies cardiac key frames based on contraction and relaxation curves (cf. Section 5.3 ), as shown in previous work [ 32 ]. The derived key frame specific strain (ED-MS, MS-ED, ED-PF, PF-MD; explained in Section 5.2 ) enables the comparison of temporally aligned ( ED2K and K2K ) deformation values (cf.…”

Section: Resultsmentioning

confidence: 67%

“…Figure 4 b ) and section 5.3 ). Mostly, the best parameters from the original publications [ 32 , 36 ] were used whenever possible to avoid dataset-driven overfitting of the models. In the Supplemental Material, the main parameters of both models are provided (cf.…”

Section: Resultsmentioning

confidence: 99%

“…With respect to the key frame detection error we report the cyclic key frame difference [ 32 ] for all five key frames on the DMD57 dataset and for the ED and ES key frame on the ACDC dataset. In terms of key frame differences (difference in frames between the derived key frame index and the labelled key frame index), the average cyclic key frame difference on the DMD57 data is 0.81±0.77, with per key frame errors of 0.86 ± 0.77, 0.60 ± 0.75, 0.79 ± 0.67, 0.96 ± 0.76, and 1.23±1.05, corresponding to the order: end diastolic (ED), mid-systolic (MS), end-systolic (ES), peak filling (PF), and mid-diastolic (MD).…”

Section: Resultsmentioning

confidence: 99%

“…The key frames are extracted in a self-supervised manner (cf. Section 5.3 ) [ 32 ], meaning for learning this task no expert labels are required, and we show the robustness of the approach in several experiments (cf. Section 2 ).…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Artificial Intelligence to derive aligned strain in cine CMR to detect patients with myocardial fibrosis: an open and scrutinizable approach

Koehler,

Kuhm,

Huffaker

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

Cine Cardiac Magnetic Resonance (CMR) is the gold standard for cardiac function evaluation, incorporating ejection fraction (EF) and strain as vital indicators of abnormal deformation. Rare pathologies like Duchenne muscular dystrophies (DMD) are monitored with repeated late gadolinium-enhanced (LGE) CMR for identification of myocardial fibrosis. However, it is judicious to reduce repeated gadolinium exposure and rather employ strain analysis from cine CMR. This solution is limited so far since full strain curves are not comparable between individual cardiac cycles and current practice mainly neglects diastolic deformation patterns. Our novel Deep Learning-based approach derives strain values aligned by key frames throughout the cardiac cycle. In a reproducibility scenario (57+82 patients), our results reveal five times more significant differences (22 vs. 4) between patients with scar and without, enhancing scar detection by +30%, improving detection of patients with preserved EF by +61%, with an overall sensitivity/specificity of 82/81%.

show abstract

Section: Resultsmentioning

confidence: 67%

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Artificial Intelligence to derive aligned strain in cine CMR to detect patients with myocardial fibrosis: an open and scrutinizable approach

Koehler,

Kuhm,

Huffaker

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Besonders bei der Erkennung der Herzphasen [15] und der Konturierung der MRT-Bilder kann die KI erheblich unterstützen. Die traditionell manuelle Segmentierung von MRT-Bildern ist mit einer Bearbeitungszeit von bis zu 30 Minuten pro Patienten äußerst zeit-und ressourcenaufwendig.…”

Section: Was Ist Wichtig?unclassified

Künstliche Intelligenz in der kardialen Bildgebung

Engelhardt,

Martin,

Rodríguez Bolanos

et al. 2023

Aktuelle Kardiologie

Self Cite

View full text Add to dashboard Cite

ZusammenfassungVerfahren der künstlichen Intelligenz (KI) haben in der kardialen Bildgebung große Relevanz erlangt. In dieser Kurzübersicht werden die Anwendungsbereiche von KI in der Echokardiografie, Magnetresonanztomografie (MRT) und Computertomografie (CT) beleuchtet. Die Analyse des ventrikulären Volumens und dessen Funktion werden durch KI in der Echokardiografie und in der MRT erweitert und verbessert. In der CT ermöglicht die KI die automatische Quantifizierung der koronaren Kalklast, Risikoklassifizierung und Prognose des Behandlungserfolgs. KI verringert die Kluft zwischen Experten und Anfängern, da sie zuverlässige und reproduzierbare Analysen durchführt bei gesteigerter Effizienz. Neue Konzepte wie das föderierte Lernen, das multizentrisches verteiltes Trainieren ermöglicht, sowie die multimodale KI, die automatisiert Text- und Bildeingaben miteinander verknüpft, werden voraussichtlich die nächsten größeren Meilensteine darstellen.

show abstract

Self-Supervised Representation Learning for Diagnosis of Cardiac Abnormalities on Echocardiograms

Krishnamoorthy,

Agrawal,

Agarwal

2024

2024 International Conference on Optimization Computing and Wireless Communication (ICOCWC)

View full text Add to dashboard Cite

Self-supervised Motion Descriptor for Cardiac Phase Detection in 4D CMR Based on Discrete Vector Field Estimations

Cited by 5 publications

References 20 publications

Artificial Intelligence to derive aligned strain in cine CMR to detect patients with myocardial fibrosis: an open and scrutinizable approach

Artificial Intelligence to derive aligned strain in cine CMR to detect patients with myocardial fibrosis: an open and scrutinizable approach

Künstliche Intelligenz in der kardialen Bildgebung

Self-Supervised Representation Learning for Diagnosis of Cardiac Abnormalities on Echocardiograms

Contact Info

Product

Resources

About