Showing 1–11 of 11 results for author: Cornelissen, S

AngioMoCo: Learning-based Motion Correction in Cerebral Digital Subtraction Angiography

Authors: Ruisheng Su, Matthijs van der Sluijs, Sandra Cornelissen, Wim van Zwam, Aad van der Lugt, Wiro Niessen, Danny Ruijters, Theo van Walsum, Adrian Dalca

Abstract: Cerebral X-ray digital subtraction angiography (DSA) is the standard imaging technique for visualizing blood flow and guiding endovascular treatments. The quality of DSA is often negatively impacted by body motion during acquisition, leading to decreased diagnostic value. Time-consuming iterative methods address motion correction based on non-rigid registration, and employ sparse key points and no… ▽ More Cerebral X-ray digital subtraction angiography (DSA) is the standard imaging technique for visualizing blood flow and guiding endovascular treatments. The quality of DSA is often negatively impacted by body motion during acquisition, leading to decreased diagnostic value. Time-consuming iterative methods address motion correction based on non-rigid registration, and employ sparse key points and non-rigidity penalties to limit vessel distortion. Recent methods alleviate subtraction artifacts by predicting the subtracted frame from the corresponding unsubtracted frame, but do not explicitly compensate for motion-induced misalignment between frames. This hinders the serial evaluation of blood flow, and often causes undesired vasculature and contrast flow alterations, leading to impeded usability in clinical practice. To address these limitations, we present AngioMoCo, a learning-based framework that generates motion-compensated DSA sequences from X-ray angiography. AngioMoCo integrates contrast extraction and motion correction, enabling differentiation between patient motion and intensity changes caused by contrast flow. This strategy improves registration quality while being substantially faster than iterative elastix-based methods. We demonstrate AngioMoCo on a large national multi-center dataset (MR CLEAN Registry) of clinically acquired angiographic images through comprehensive qualitative and quantitative analyses. AngioMoCo produces high-quality motion-compensated DSA, removing motion artifacts while preserving contrast flow. Code is publicly available at https://github.com/RuishengSu/AngioMoCo. △ Less

Submitted 9 October, 2023; originally announced October 2023.

MedShapeNet -- A Large-Scale Dataset of 3D Medical Shapes for Computer Vision

Authors: Jianning Li, Zongwei Zhou, Jiancheng Yang, Antonio Pepe, Christina Gsaxner, Gijs Luijten, Chongyu Qu, Tiezheng Zhang, Xiaoxi Chen, Wenxuan Li, Marek Wodzinski, Paul Friedrich, Kangxian Xie, Yuan **, Narmada Ambigapathy, Enrico Nasca, Naida Solak, Gian Marco Melito, Viet Duc Vu, Afaque R. Memon, Christopher Schlachta, Sandrine De Ribaupierre, Rajnikant Patel, Roy Eagleson, Xiaojun Chen , et al. (132 additional authors not shown)

Abstract: Prior to the deep learning era, shape was commonly used to describe the objects. Nowadays, state-of-the-art (SOTA) algorithms in medical imaging are predominantly diverging from computer vision, where voxel grids, meshes, point clouds, and implicit surface models are used. This is seen from numerous shape-related publications in premier vision conferences as well as the growing popularity of Shape… ▽ More Prior to the deep learning era, shape was commonly used to describe the objects. Nowadays, state-of-the-art (SOTA) algorithms in medical imaging are predominantly diverging from computer vision, where voxel grids, meshes, point clouds, and implicit surface models are used. This is seen from numerous shape-related publications in premier vision conferences as well as the growing popularity of ShapeNet (about 51,300 models) and Princeton ModelNet (127,915 models). For the medical domain, we present a large collection of anatomical shapes (e.g., bones, organs, vessels) and 3D models of surgical instrument, called MedShapeNet, created to facilitate the translation of data-driven vision algorithms to medical applications and to adapt SOTA vision algorithms to medical problems. As a unique feature, we directly model the majority of shapes on the imaging data of real patients. As of today, MedShapeNet includes 23 dataset with more than 100,000 shapes that are paired with annotations (ground truth). Our data is freely accessible via a web interface and a Python application programming interface (API) and can be used for discriminative, reconstructive, and variational benchmarks as well as various applications in virtual, augmented, or mixed reality, and 3D printing. Exemplary, we present use cases in the fields of classification of brain tumors, facial and skull reconstructions, multi-class anatomy completion, education, and 3D printing. In future, we will extend the data and improve the interfaces. The project pages are: https://medshapenet.ikim.nrw/ and https://github.com/Jianningli/medshapenet-feedback △ Less

Submitted 12 December, 2023; v1 submitted 30 August, 2023; originally announced August 2023.

Comments: 16 pages

MSC Class: 68T01

Voxel-wise classification for porosity investigation of additive manufactured parts with 3D unsupervised and (deeply) supervised neural networks

Authors: Domenico Iuso, Soumick Chatterjee, Sven Cornelissen, Dries Verhees, Jan De Beenhouwer, Jan Sijbers

Abstract: Additive Manufacturing (AM) has emerged as a manufacturing process that allows the direct production of samples from digital models. To ensure that quality standards are met in all manufactured samples of a batch, X-ray computed tomography (X-CT) is often used combined with automated anomaly detection. For the latter, deep learning (DL) anomaly detection techniques are increasingly, as they can be… ▽ More Additive Manufacturing (AM) has emerged as a manufacturing process that allows the direct production of samples from digital models. To ensure that quality standards are met in all manufactured samples of a batch, X-ray computed tomography (X-CT) is often used combined with automated anomaly detection. For the latter, deep learning (DL) anomaly detection techniques are increasingly, as they can be trained to be robust to the material being analysed and resilient towards poor image quality. Unfortunately, most recent and popular DL models have been developed for 2D image processing, thereby disregarding valuable volumetric information. This study revisits recent supervised (UNet, UNet++, UNet 3+, MSS-UNet) and unsupervised (VAE, ceVAE, gmVAE, vqVAE) DL models for porosity analysis of AM samples from X-CT images and extends them to accept 3D input data with a 3D-patch pipeline for lower computational requirements, improved efficiency and generalisability. The supervised models were trained using the Focal Tversky loss to address class imbalance that arises from the low porosity in the training datasets. The output of the unsupervised models is post-processed to reduce misclassifications caused by their inability to adequately represent the object surface. The findings were cross-validated in a 5-fold fashion and include: a performance benchmark of the DL models, an evaluation of the post-processing algorithm, an evaluation of the effect of training supervised models with the output of unsupervised models. In a final performance benchmark on a test set with poor image quality, the best performing supervised model was UNet++ with an average precision of 0.751 $\pm$ 0.030, while the best unsupervised model was the post-processed ceVAE with 0.830 $\pm$ 0.003. The VAE/ceVAE models demonstrated superior capabilities, particularly when leveraging post-processing techniques. △ Less

Submitted 9 June, 2023; v1 submitted 13 May, 2023; originally announced May 2023.

CAVE: Cerebral Artery-Vein Segmentation in Digital Subtraction Angiography

Authors: Ruisheng Su, P. Matthijs van der Sluijs, Yuan Chen, Sandra Cornelissen, Ruben van den Broek, Wim H. van Zwam, Aad van der Lugt, Wiro Niessen, Danny Ruijters, Theo van Walsum

Abstract: Cerebral X-ray digital subtraction angiography (DSA) is a widely used imaging technique in patients with neurovascular disease, allowing for vessel and flow visualization with high spatio-temporal resolution. Automatic artery-vein segmentation in DSA plays a fundamental role in vascular analysis with quantitative biomarker extraction, facilitating a wide range of clinical applications. The widely… ▽ More Cerebral X-ray digital subtraction angiography (DSA) is a widely used imaging technique in patients with neurovascular disease, allowing for vessel and flow visualization with high spatio-temporal resolution. Automatic artery-vein segmentation in DSA plays a fundamental role in vascular analysis with quantitative biomarker extraction, facilitating a wide range of clinical applications. The widely adopted U-Net applied on static DSA frames often struggles with disentangling vessels from subtraction artifacts. Further, it falls short in effectively separating arteries and veins as it disregards the temporal perspectives inherent in DSA. To address these limitations, we propose to simultaneously leverage spatial vasculature and temporal cerebral flow characteristics to segment arteries and veins in DSA. The proposed network, coined CAVE, encodes a 2D+time DSA series using spatial modules, aggregates all the features using temporal modules, and decodes it into 2D segmentation maps. On a large multi-center clinical dataset, CAVE achieves a vessel segmentation Dice of 0.84 ($\pm$0.04) and an artery-vein segmentation Dice of 0.79 ($\pm$0.06). CAVE surpasses traditional Frangi-based K-means clustering (P<0.001) and U-Net (P<0.001) by a significant margin, demonstrating the advantages of harvesting spatio-temporal features. This study represents the first investigation into automatic artery-vein segmentation in DSA using deep learning. The code is publicly available at https://github.com/RuishengSu/CAVE_DSA. △ Less

Submitted 1 May, 2024; v1 submitted 3 August, 2022; originally announced August 2022.

Comments: Published in Computerized Medical Imaging and Graphics

autoTICI: Automatic Brain Tissue Reperfusion Scoring on 2D DSA Images of Acute Ischemic Stroke Patients

Authors: Ruisheng Su, Sandra A. P. Cornelissen, Matthijs van der Sluijs, Adriaan C. G. M. van Es, Wim H. van Zwam, Diederik W. J. Dippel, Geert Lycklama, Pieter Jan van Doormaal, Wiro J. Niessen, Aad van der Lugt, Theo van Walsum

Abstract: The Thrombolysis in Cerebral Infarction (TICI) score is an important metric for reperfusion therapy assessment in acute ischemic stroke. It is commonly used as a technical outcome measure after endovascular treatment (EVT). Existing TICI scores are defined in coarse ordinal grades based on visual inspection, leading to inter- and intra-observer variation. In this work, we present autoTICI, an auto… ▽ More The Thrombolysis in Cerebral Infarction (TICI) score is an important metric for reperfusion therapy assessment in acute ischemic stroke. It is commonly used as a technical outcome measure after endovascular treatment (EVT). Existing TICI scores are defined in coarse ordinal grades based on visual inspection, leading to inter- and intra-observer variation. In this work, we present autoTICI, an automatic and quantitative TICI scoring method. First, each digital subtraction angiography (DSA) acquisition is separated into four phases (non-contrast, arterial, parenchymal and venous phase) using a multi-path convolutional neural network (CNN), which exploits spatio-temporal features. The network also incorporates sequence level label dependencies in the form of a state-transition matrix. Next, a minimum intensity map (MINIP) is computed using the motion corrected arterial and parenchymal frames. On the MINIP image, vessel, perfusion and background pixels are segmented. Finally, we quantify the autoTICI score as the ratio of reperfused pixels after EVT. On a routinely acquired multi-center dataset, the proposed autoTICI shows good correlation with the extended TICI (eTICI) reference with an average area under the curve (AUC) score of 0.81. The AUC score is 0.90 with respect to the dichotomized eTICI. In terms of clinical outcome prediction, we demonstrate that autoTICI is overall comparable to eTICI. △ Less

Submitted 7 May, 2021; v1 submitted 3 October, 2020; originally announced October 2020.

Comments: Accepted for publication in IEEE Transactions on Medical Imaging (IEEE TMI)

Laboratory demonstration of a cryogenic deformable mirror for wavefront correction of space-borne infrared telescopes

Authors: Aoi Takahashi, Keigo Enya, Kanae Haze, Hirokazu Kataza, Takayuki Kotani, Hideo Matsuhara, Tomohiro Kamiya, Tomoyasu Yamamuro, Paul Bierden, Steven Cornelissen, Charlie Lam, Michael Feinberg

Abstract: This paper demonstrates a cryogenic deformable mirror (DM) with 1,020 actuators based on micro-electrical mechanical systems (MEMS) technology. Cryogenic space-borne infrared telescopes can experience a wavefront error due to a figure error of their mirror surface, which makes the imaging performance worse. For on-orbit wavefront correction as one solution, we developed a MEMS-processed electro-st… ▽ More This paper demonstrates a cryogenic deformable mirror (DM) with 1,020 actuators based on micro-electrical mechanical systems (MEMS) technology. Cryogenic space-borne infrared telescopes can experience a wavefront error due to a figure error of their mirror surface, which makes the imaging performance worse. For on-orbit wavefront correction as one solution, we developed a MEMS-processed electro-static DM with a special surrounding structure for use under the cryogenic temperature. We conducted a laboratory demonstration of its operation in three cooling cycles between 5 K and 295 K. Using a laser interferometer, we detected the deformation corresponding to the applied voltages under the cryogenic temperature for the first time. The relationship between voltages and displacements was qualitatively expressed by the quadratic function, which is assumed based on the principle of electro-static DMs. We also found that it had a high operating repeatability of a few nm RMS and no significant hysteresis. Using the measured values of repeatability, we simulated the improvement of PSF by wavefront correction with our DM. These results show that our developed DM is effective in improving imaging performance and PSF contrast of space-borne infrared telescopes. △ Less

Submitted 29 September, 2017; originally announced October 2017.

Comments: 29 pages, 23 figures

Journal ref: Applied Optics, Vol.56, Issue 23, pp. 6694-6708 (2017)

MEMS practice, from the lab to the telescope

Authors: Katie M. Morzinski, Andrew P. Norton, Julia Wilhelmson Evans, Layra Reza, Scott A. Severson, Daren Dillon, Marc Reinig, Donald T. Gavel, Steven Cornelissen, Bruce A. Macintosh, Claire E. Max

Abstract: Micro-electro-mechanical systems (MEMS) technology can provide for deformable mirrors (DMs) with excellent performance within a favorable economy of scale. Large MEMS-based astronomical adaptive optics (AO) systems such as the Gemini Planet Imager are coming on-line soon. As MEMS DM end-users, we discuss our decade of practice with the micromirrors, from inspecting and characterizing devices to ev… ▽ More Micro-electro-mechanical systems (MEMS) technology can provide for deformable mirrors (DMs) with excellent performance within a favorable economy of scale. Large MEMS-based astronomical adaptive optics (AO) systems such as the Gemini Planet Imager are coming on-line soon. As MEMS DM end-users, we discuss our decade of practice with the micromirrors, from inspecting and characterizing devices to evaluating their performance in the lab. We also show MEMS wavefront correction on-sky with the "Villages" AO system on a 1-m telescope, including open-loop control and visible-light imaging. Our work demonstrates the maturity of MEMS technology for astronomical adaptive optics. △ Less

Submitted 7 February, 2012; originally announced February 2012.

Comments: 14 pages, 15 figures, Invited Paper, SPIE Photonics West 2012

Direct experimental measurement of phase-amplitude coupling in spin torque oscillators

Authors: L. Bianchini, S. Cornelissen, Joo-Von Kim, T. Devolder, W. van Roy, L. Lagae, C. Chappert

Abstract: We study spin-torque induced oscillations of MgO magnetic tunnel junctions in the time domain. By using the Hilbert transform on the time traces, we obtain for the first time a direct experimental measure of the coupling between the power and the phase fluctuations. We deduce the power restoration rate and we obtain low values for the coupling strength, which is consistent with the weak frequency… ▽ More We study spin-torque induced oscillations of MgO magnetic tunnel junctions in the time domain. By using the Hilbert transform on the time traces, we obtain for the first time a direct experimental measure of the coupling between the power and the phase fluctuations. We deduce the power restoration rate and we obtain low values for the coupling strength, which is consistent with the weak frequency dependence on the applied voltage. △ Less

Submitted 21 July, 2010; v1 submitted 29 April, 2010; originally announced April 2010.

Comments: 3 pages, 3 figures

Journal ref: Appl. Phys. Lett. 97, 032502 (2010)

Quantized spin wave modes in magnetic tunnel junction nanopillars

Authors: A. Helmer, S. Cornelissen, T. Devolder, J. -V. Kim, W. van Roy, L. Lagae, C. Chappert

Abstract: We present an experimental and theoretical study of the magnetic field dependence of the mode frequency of thermally excited spin waves in rectangular shaped nanopillars of lateral sizes 60x100, 75x150, and 105x190 nm2, patterned from MgO-based magnetic tunnel junctions. The spin wave frequencies were measured using spectrally resolved electrical noise measurements. In all spectra, several indep… ▽ More We present an experimental and theoretical study of the magnetic field dependence of the mode frequency of thermally excited spin waves in rectangular shaped nanopillars of lateral sizes 60x100, 75x150, and 105x190 nm2, patterned from MgO-based magnetic tunnel junctions. The spin wave frequencies were measured using spectrally resolved electrical noise measurements. In all spectra, several independent quantized spin wave modes have been observed and could be identified as eigenexcitations of the free layer and of the synthetic antiferromagnet of the junction. Using a theoretical approach based on the diagonalization of the dynamical matrix of a system of three coupled, spatially confined magnetic layers, we have modeled the spectra for the smallest pillar and have extracted its material parameters. The magnetization and exchange stiffness constant of the CoFeB free layer are thereby found to be substantially reduced compared to the corresponding thin film values. Moreover, we could infer that the pinning of the magnetization at the lateral boundaries must be weak. Finally, the interlayer dipolar coupling between the free layer and the synthetic antiferromagnet causes mode anticrossings with gap openings up to 2 GHz. At low fields and in the larger pillars, there is clear evidence for strong non-uniformities of the layer magnetizations. In particular, at zero field the lowest mode is not the fundamental mode, but a mode most likely localized near the layer edges. △ Less

Submitted 8 January, 2010; v1 submitted 22 July, 2009; originally announced July 2009.

Comments: 16 pages, 4 figures, (re)submitted to PRB

Auto-oscillation threshold and line narrowing in MgO-based spin-torque oscillators

Authors: S. Cornelissen, L. Bianchini, G. Hrkac, M. Op de Beeck, L. Lagae, Joo-Von Kim, T. Devolder, P. Crozat, C. Chappert, T. Schrefl

Abstract: We present an experimental study of the power spectrum of current-driven magnetization oscillations in MgO tunnel junctions under low bias. We find the existence of narrow spectral lines, down to 8 MHz in width at a frequency of 10.7 GHz, for small applied fields with clear evidence of an auto-oscillation threshold. Micromagnetics simulations indicate that the excited mode corresponds to an edge… ▽ More We present an experimental study of the power spectrum of current-driven magnetization oscillations in MgO tunnel junctions under low bias. We find the existence of narrow spectral lines, down to 8 MHz in width at a frequency of 10.7 GHz, for small applied fields with clear evidence of an auto-oscillation threshold. Micromagnetics simulations indicate that the excited mode corresponds to an edge mode of the synthetic antiferromagnet. △ Less

Submitted 16 September, 2009; v1 submitted 7 October, 2008; originally announced October 2008.

Journal ref: Europhys. Lett. 87, 57001 (2009)

Auto-oscillation threshold, narrow spectral lines, and line jitter in spin-torque oscillators based on MgO magnetic tunnel junctions

Authors: T. Devolder, S. Cornelissen, L. Bianchini, Joo-Von Kim, P. Crozat, C. Chappert, M. Op de Beeck, L. Lagae

Abstract: We demonstrate spin torque induced auto-oscillation in MgO-based magnetic tunnel junctions. At the generation threshold, we observe a strong line narrowing down to 6 MHz at 300K and a dramatic increase in oscillator power, yielding spectrally pure oscillations free of flicker noise. Setting the synthetic antiferromagnet into autooscillation requires the same current polarity as the one needed to… ▽ More We demonstrate spin torque induced auto-oscillation in MgO-based magnetic tunnel junctions. At the generation threshold, we observe a strong line narrowing down to 6 MHz at 300K and a dramatic increase in oscillator power, yielding spectrally pure oscillations free of flicker noise. Setting the synthetic antiferromagnet into autooscillation requires the same current polarity as the one needed to switch the free layer magnetization. The induced auto-oscillations are observed even at zero applied field, which is believed to be the acoustic mode of the synthetic antiferromagnet. While the phase coherence of the auto-oscillation is of the order of microseconds, the power autocorrelation time is of the order of milliseconds and can be strongly influenced by the free layer dynamics. △ Less

Submitted 6 October, 2008; originally announced October 2008.