Article
Radiology, Nuclear Medicine & Medical Imaging
Ranjeet Ranjan Jha, Sudhir K. Pathak, Vishwesh Nath, Walter Schneider, B. V. Rathish Kumar, Arnav Bhavsar, Aditya Nigam
Summary: Diffusion MRI (dMRI) is widely used for studying the brain structure, particularly the white matter region. The high angular resolution diffusion imaging (HARDI) technique is favored by researchers for its accurate estimation of fiber orientation. However, accurately estimating the intravoxel structure is challenging with the current single-shell HARDI. To address this issue, we propose a novel generative adversarial network, VRfRNet, for reconstructing multi-shell multi-tissue fiber orientation distribution function from single-shell HARDI volumes.
MAGNETIC RESONANCE IMAGING
(2022)
Article
Engineering, Electrical & Electronic
Remi Carloni Gertosio, Jerome Bobin
Summary: With the development of multichannel imagers, blind source separation (BSS) algorithms have become widely used in astrophysics to analyze multispectral images. However, applying BSS algorithms to data from large radio interferometers poses challenges due to incomplete and deteriorated data, as well as non-coplanar effects. To address these challenges, a joint non-coplanar deconvolution and BSS algorithm called wGMCA is introduced, which shows robustness and advantages compared to classical methods.
DIGITAL SIGNAL PROCESSING
(2023)
Article
Acoustics
Yuki Mitsufuji, Norihiro Takamune, Shoichi Koyama, Hiroshi Saruwatari
Summary: This paper addresses the conversion issue between object format and HOA format in VR systems, proposes blind source separation in the spherical harmonic domain, and enhances existing methods by estimating model parameters of non-negative tensor factorization for near-field sources.
IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING
(2021)
Article
Engineering, Electrical & Electronic
Huijuan Wu, Yimeng Liu, Yunlin Tu, Yuwen Sun, Dengke Gan, Yuanfeng Song, Yunjiang Rao
Summary: In this paper, a blind multi-source separation method based on fast independent component analysis (FastICA) is proposed for fiber-optical distributed acoustic sensor (DAS). The method utilizes the independency and non-Gaussianity of different sources to solve the challenge of detecting and identifying unpredictable vibration sources when they are superimposed at the same fiber receiving point. The method includes discussions on multi-source mixing mechanisms and separability, linear simultaneous mixing mode assumption, estimation of source number, preprocessing of denoising and anti-mixing, separation with FastICA by maximizing negative entropy, and evaluation of the method's feasibility through simulations and real field tests.
JOURNAL OF LIGHTWAVE TECHNOLOGY
(2022)
Article
Acoustics
Mathieu Fontaine, Kouhei Sekiguchi, Aditya Arie Nugraha, Yoshiaki Bando, Kazuyoshi Yoshii
Summary: This paper introduces a heavy-tailed extension method for blind source separation called GSM-FastMNMF, based on a wider class of heavy-tailed distributions. It outperforms existing methods in speech enhancement and separation.
IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING
(2022)
Article
Engineering, Electrical & Electronic
Qadri Mayyala, Karim Abed-Meraim, Azzedine Zerguine, Abdulmajid Lawal
Summary: A novel class of fast Multi-Modulus algorithms for Blind Source Separation and deconvolution are proposed in this work. These algorithms minimize the Multi-Modulus criterion through a fast fixed-point optimization rule and belong to the fixed step-size gradient descent family. They converge even faster with the proposed algebraic variable step-size and do not require any user-defined parameters.
IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Tiantian He, Yew-Soon Ong, Pengwei Hu
Summary: The paper introduces a novel network clustering framework called MSPANC, which utilizes multi-source vertex features to uncover clusters in network data, outperforming most previous methods.
Article
Biophysics
Julia Sauer, Merle Streppel, Niklas M. Carbon, Eike Petersen, Philipp Rostalski
Summary: This study proposes a blind source separation method to address the crosstalk and cardiac activity issues in respiratory sEMG signals of ventilated patients. The method consists of a two-step procedure using wavelet domain and nonnegative matrix factorization algorithms. Experimental results on clinical datasets and simulated data show significant improvement across various conditions. The proposed method can help clinicians interpret respiratory sEMG signals and distinguish between inspiratory effort and other muscle activities.
PHYSIOLOGICAL MEASUREMENT
(2022)
Article
Computer Science, Artificial Intelligence
Christophe Kervazo, Nicolas Gillis, Nicolas Dobigeon
Summary: In this work, we introduce two algorithms to tackle blind source separation focusing on the linear-quadratic (LQ) model, with the first algorithm SNPALQ explicitly modeling product terms to improve separation quality, and the second algorithm BF used as a postprocessing step to discard mixed samples. Both algorithms are shown to be relevant in realistic numerical experiments, with SNPALQ capable of recovering ground truth factors even in the presence of noise, while BF demonstrates robustness under easier-to-check conditions than SNPALQ.
SIAM JOURNAL ON IMAGING SCIENCES
(2021)
Article
Radiology, Nuclear Medicine & Medical Imaging
Ranjeet Ranjan Jha, Gaurav Jaswal, Arnav Bhavsar, Aditya Nigam
Summary: Single or multi-shell high angular resolution diffusion imaging is an important technique for studying brain white matter fibers. The existing single-shell technique has limitations in estimating the intravoxel structure at the desired resolution. In this study, a deep learning architecture is proposed to reconstruct diffusion MRI volumes for different b-values using single-shell acquisitions, allowing for higher resolution and accurate fiber tracts. The proposed framework incorporates contextual information within each slice and across the slices to optimize network learning, achieving promising results in the validation.
MAGNETIC RESONANCE IMAGING
(2022)
Article
Engineering, Electrical & Electronic
Baoze Ma, Tianqi Zhang, Zeliang An, Chen Yi
Summary: This paper proposes a new method for addressing permutation ambiguity issue of convolutive blind source separation in the frequency domain. The technique utilizes nonnegative matrix factorization (NMF) and canonical correlation analysis (CCA) to effectively separate and align the signals of convolutive mixtures.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS
(2022)
Article
Engineering, Electrical & Electronic
Yifan Wang, Yibing Li, Qian Sun, Yingsong Li
Summary: This paper proposes a blind source separation algorithm for frequency-hopping signals, which consists of three stages for parameter estimation, mixing matrix estimation, and signal estimation. Experimental results demonstrate the superior performance of the algorithm.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS
(2023)
Article
Nuclear Science & Technology
Hanan Arahmane, El-Mehdi Hamzaoui, Yann Ben Maissa, Rajaa Cherkaoui El Moursli
Summary: This study focuses on achieving fast and accurate neutron-gamma discrimination in a mixed radiation field without prior knowledge of the signals or experimental setup. By utilizing nonnegative tensor factorization (NTF) and support vector machine (SVM), a novel method was proposed to extract and discriminate the components from mixed signals, resulting in a high true-positive rate for neutron-gamma classification. The approach demonstrated a superior discrimination quality compared to the charge comparison method, with a figure of merit of 2.20.
NUCLEAR SCIENCE AND TECHNIQUES
(2021)
Article
Geochemistry & Geophysics
Si Chen, Yue Yuan, Sixiang Wang, Huanhuan Yang, Lingzhi Zhu, Shuning Zhang, Huichang Zhao
Summary: A novel multi-electromagnetic jamming countermeasure for airborne SAR based on maximum SNR blind source separation is proposed in this article. By identifying the real target echo signal and using corresponding imaging methods, high-resolution images of the interested target area are achieved.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Engineering, Electrical & Electronic
Remi Carloni Gertosio, Jerome Bobin, Fabio Acero
Summary: Blind source separation (BSS) algorithms are unsupervised methods that allow for physically meaningful data decompositions. This article proposes a semi-supervised source separation approach that combines a projected alternating least-square algorithm with a learning-based regularization scheme. By constraining the mixing matrix using generative models, the proposed method, sGMCA, achieves improved accuracy and physically interpretable solutions in challenging scenarios.
Article
Neurosciences
Mathieu Vandenbulcke, Laura Van de Vliet, Jiaze Sun, Yun-An Huang, Maarten J. A. Van den Bossche, Stefan Sunaert, Ron Peeters, Qi Zhu, Wim Vanduffel, Beatrice de Gelder, Francois-Laurent De Winter, Jan Van den Stock
Summary: This study investigates the neural integrity of hyperspecialized and domain-general cortical social brain areas in behavioral variant frontotemporal dementia (bvFTD) by using structural and functional magnetic resonance imaging. The results reveal compromised structure and function in hyperspecialized social areas in bvFTD.
Article
Oncology
Michiel B. de Ruiter, Paul F. C. Groot, Sabine Deprez, Pim Pullens, Stefan Sunaert, Dirk de Ruysscher, Sanne B. Schagen, Jose Belderbos
Summary: The study found that HA-PCI can reduce hippocampal atrophy, but both types of radiotherapy are associated with considerable brain injury. There was no significant association between hippocampal atrophy and memory decline.
Article
Clinical Neurology
Jiaze Sun, Francois-Laurent De Winter, Fiona Kumfor, Daphne Stam, Kristof Vansteelandt, Ron Peeters, Stefan Sunaert, Rik Vandenberghe, Mathieu Vandenbulcke, Jan Van den Stock
Summary: This study explores the presence of neural functional compensation in the manifest stage of neurodegenerative diseases. The results suggest that compensatory processes can still occur in clinically manifest neurodegeneration, and these processes may operate along nodes in intrinsically connected networks. The findings highlight the potential of using multidimensional neural markers as novel biomarkers for diagnosis and therapy.
JOURNAL OF NEUROLOGY
(2023)
Article
Clinical Neurology
Laura Michiels, Liselot Thijs, Nathalie Mertens, Stefan Sunaert, Mathieu Vandenbulcke, Guy Bormans, Geert Verheyden, Michel Koole, Koen Van Laere, Robin Lemmens
Summary: The risk of Alzheimer's disease increases after stroke, and this may not be solely due to traditional vascular risk factors. Tau proteins released from neuronal death may contribute to the formation of neurofibrillary tangles (NFT) after ischemia. The study used F-18-MK-6240 PET to explore the distribution of NFT after ischemic stroke in vivo.
Review
Gastroenterology & Hepatology
Pieter Sinonquel, Severine Vermeire, Frederik Maes, Raf Bisschops
Summary: This article reviews the latest developments in advanced GI endoscopy, focusing on screening, diagnosis, and surveillance of common upper and lower GI pathology. The importance of artificial intelligence in the field is emphasized, and the potential future impact of the literature is assessed.
GE PORTUGUESE JOURNAL OF GASTROENTEROLOGY
(2023)
Article
Pharmacology & Pharmacy
Wouter Botermans, Michel Koole, Koen Van Laere, Jonathan R. Savidge, John A. Kemp, Stefan Sunaert, Maeve M. Duffy, Steven Ramael, Andrea M. Cesura, Kevin D'Ostilio, Denis Gossen, Torsten M. Madsen, Thomas Lodeweyckx, Jan de Hoon
Summary: SDI-118 is a small molecule that enhances cognitive function by modulating the activity of SV2A in the brain. The first-in-human study showed that SDI-118 was well tolerated and safe in healthy male subjects. These findings support further clinical exploration of SDI-118 in patients with cognitive disorders.
FRONTIERS IN PHARMACOLOGY
(2023)
Article
Biology
Steven Jillings, Ekaterina Pechenkova, Elena Tomilovskaya, Ilya Rukavishnikov, Ben Jeurissen, Angelique Van Ombergen, Inna Nosikova, Alena Rumshiskaya, Liudmila Litvinova, Jitka Annen, Chloe De Laet, Catho Schoenmaekers, Jan Sijbers, Victor Petrovichev, Stefan Sunaert, Paul M. Parizel, Valentin Sinitsyn, Peter zu Eulenburg, Steven Laureys, Athena Demertzi, Floris L. Wuyts
Summary: This study used functional magnetic resonance imaging (fMRI) to investigate the impact of prolonged microgravity on the human brain. The results showed changes in brain functional connectivity, providing insights into the adaptations and modifications occurring during spaceflight and upon return to Earth.
COMMUNICATIONS BIOLOGY
(2023)
Article
Oncology
Heleen Bollen, Siri Willems, Marilyn Wegge, Frederik Maes, Sandra Nuyts
Summary: An automated gross tumor volume (GTV) delineation approach based on a 3D convolutional neural network (CNN) was developed for head and neck cancer (HNC) radiation therapy planning. The multi-modality CNNs showed better performance compared to the single-modality CNN, and were proven to be more efficient and consistent than manual delineation in a clinical setting, leading to increased efficiency and reduced interobserver variability.
RADIOTHERAPY AND ONCOLOGY
(2023)
Article
Psychology, Clinical
Michelle Melis, Jeroen Blommaert, Ahmed Radwan, Ann Smeets, Katleen van der Gucht, Sabine Deprez, Stefan Sunaert
Summary: This study found that mindfulness-based interventions may be associated with short-term structural changes in the brain related to cognitive impairment in breast cancer survivors, but no long-term effects were found.
Correction
Psychology, Clinical
Michelle Melis, Jeroen Blommaert, Ahmed Radwan, Ann Smeets, Katleen Van der Gucht, Sabine Deprez, Stefan Sunaert
Article
Neuroimaging
Elise Turk, Marion I. van den Heuvel, Charlotte Sleurs, Thibo Billiet, Anne Uyttebroeck, Stefan Sunaert, Maarten Mennes, Bea R. H. van den Bergh
Summary: This study investigates the impact of maternal anxiety during pregnancy on offspring's brain through the connectome. The findings suggest a long-term negative impact of high maternal anxiety on the functional connectivity between the medial prefrontal cortex and the inferior frontal gyrus in adult offspring.
BRAIN IMAGING AND BEHAVIOR
(2023)
Meeting Abstract
Clinical Neurology
B. De Wel, L. Iterbeke, L. Huysmans, R. Peeters, V. Goosens, S. Ghysels, K. Byloos, G. Putzeys, N. Dubuisson, P. van den Bergh, V. Van Parijs, G. Remiche, F. Maes, P. Dupont, K. Claeys
NEUROMUSCULAR DISORDERS
(2023)
Article
Biophysics
Mladen Rakic, Federico Turco, Guodong Weng, Frederik Maes, Diana M. Sima, Johannes Slotboom
Summary: With the rise of novel 3D magnetic resonance spectroscopy imaging (MRSI) acquisition protocols in clinical practice, the proposed automated preprocessing pipeline based on deep-learning classifiers filters out bad-quality spectra and identifies contaminated but salvageable spectra prior to metabolite quantification. The AI model, consisting of convolutional autoencoders and multilayer perceptron networks, achieves high F1 scores for different classes of spectral quality. The model not only minimizes user interaction, but also reduces computation time and displays clinically relevant information.
NMR IN BIOMEDICINE
(2023)
Article
Engineering, Biomedical
L. Vandewinckele, T. Reynders, C. Weltens, F. Maes, W. Crijns
Summary: This study presents a method to directly predict volumetric modulated arc therapy (VMAT) multi-leaf collimator (MLC) apertures and monitor units (MUs) from patient anatomy using deep learning, and fine-tunes the predicted plan using an optimizer. In testing, this method achieved clinically acceptable plans while reducing planning time by more than half.
PHYSICS IN MEDICINE AND BIOLOGY
(2023)
Meeting Abstract
Oncology
D. Callens, L. Vandewinckele, P. Berkovic, F. Maes, M. Lambrecht, W. Crijns
RADIOTHERAPY AND ONCOLOGY
(2023)
Article
Neurosciences
Jose Sanchez-Bornot, Roberto C. Sotero, J. A. Scott Kelso, Ozguer Simsek, Damien Coyle
Summary: This study proposes a multi-penalized state-space model for analyzing unobserved dynamics, using a data-driven regularization method. Novel algorithms are developed to solve the model, and a cross-validation method is introduced to evaluate regularization parameters. The effectiveness of this method is validated through simulations and real data analysis, enabling a more accurate exploration of cognitive brain functions.