- Advanced X-ray and CT Imaging
- Radiomics and Machine Learning in Medical Imaging
- Cardiac Imaging and Diagnostics
- Cardiovascular Function and Risk Factors
- Image and Signal Denoising Methods
- Photoacoustic and Ultrasonic Imaging
- Medical Imaging Techniques and Applications
- Music and Audio Processing
- Anomaly Detection Techniques and Applications
- Water Systems and Optimization
- Advanced Image Processing Techniques
Norwegian University of Science and Technology
2024-2025
IMEC
2022
KU Leuven
2022
To develop and compare methods to automatically estimate regional ultrasound image quality for echocardiography separate from view correctness. Three estimating were developed: (i) classic pixel-based metric: the generalized contrast-to-noise ratio (gCNR), computed on myocardial segments (region of interest) left ventricle lumen (background), extracted by a U-Net segmentation model; (ii) local coherence: average coherence as predicted model that predicts B-mode images at pixel level; (iii)...
Fully automatic cardiac segmentation can be a fast and reproducible method to extract clinical measurements from an echocardiography examination. The U-Net architecture is the current state-of-the-art deep learning for medical segment structures in real-time with average errors comparable inter-observer variability. However, this still generates large outliers that are often anatomically incorrect. This work uses concept of graph convolutional neural networks predict contour points interest...
Automatic estimation of cardiac ultrasound image quality can be beneficial for guiding operators and ensuring the accuracy clinical measurements. Previous work often fails to distinguish view correctness echocardiogram from quality. Additionally, previous studies only provide a global value, which limits their practical utility. In this work, we developed compared three methods estimate quality: 1) classic pixel-based metrics like generalized contrast-to-noise ratio (gCNR) on myocardial...
Anomaly detection models can help to automatically and proactively detect faults in industrial machines. Microphones are appealing as they generally inexpensive unlike visual inspection, recording sound samples give information about the internals of machine. However, conventional methods based on an AutoEncoder (AE) structure learned from scratch struggle learn how robustly reconstruct with limited available data. This paper addresses this problem by presenting a method for unsupervised...
Fully automatic cardiac segmentation can be a fast and reproducible method to extract clinical measurements from an echocardiography examination. The U-Net architecture is the current state-of-the-art deep learning for medical segment structures in real-time with average errors comparable inter-observer variability. However, this still generates large outliers that are often anatomically incorrect. This work uses concept of graph convolutional neural networks predict contour points interest...