Article
Neurosciences
Yueqi Qiu, Haoran Bai, Hao Chen, Yue Zhao, Hai Luo, Ziyue Wu, Zhiyong Zhang
Summary: This work optimized the imaging protocol for low-field SWI and proposed methods to improve the signal-to-noise ratio (SNR) performance. A comparison of SWI at different field strengths demonstrated its capability in identifying tissue differences.
FRONTIERS IN NEUROSCIENCE
(2022)
Article
Neurosciences
Korbinian Eckstein, Beata Bachrata, Gilbert Hangel, Georg Widhalm, Christian Enzinger, Markus Barth, Siegfried Trattnig, Simon Daniel Robinson
Summary: The study optimized the processing of Susceptibility Weighted Imaging to propose the CLEAR-SWI method, which not only improves contrast and image homogeneity, but also reduces artefacts and signal dropouts. Experimental results in healthy volunteers and brain tumor patients show that CLEAR-SWI has better contrast-to-noise and homogeneity, as well as higher image quality compared to other methods.
Article
Clinical Neurology
Rupsa Bhattacharjee, Mamta Gupta, Tanu Singh, Shalini Sharma, Gaurav Khanna, Suhail P. Parvaze, Rana Patir, Sandeep Vaishya, Sunita Ahlawat, Anup Singh, Rakesh Kumar Gupta
Summary: This study aimed to retrospectively evaluate the potential of qualitative and quantitative multiparametric features assessed on T-2, post-contrast T-1, DWI, DCE-MRI, and SWI in differentiating PCNSL vs GB. The results showed that SWI features provided the best diagnostic performance in differentiating PCNSL and GB.
Article
Veterinary Sciences
Nadja Wolfer, Adriano Wang-Leandro, Katrin M. Beckmann, Henning Richter, Matthias Dennler
Summary: SWI is more effective than T2*WI in detecting ASV in dogs, with the ability to differentiate between hemorrhagic and calcified lesions. Frontal sinus conformation can impact image interpretation, with brachycephalic dogs showing less severe artifacts in both sequences.
FRONTIERS IN VETERINARY SCIENCE
(2021)
Review
Radiology, Nuclear Medicine & Medical Imaging
Tanja Platt, Mark E. Ladd, Daniel Paech
Summary: Ultrahigh magnetic fields, referred to as UHFs with B-0 >= 7T, offer higher signal-to-noise and contrast-to-noise ratios. While they provide advantages in resolving structures and visualizing physiological/pathophysiological effects, challenges such as inhomogeneities and higher energy deposition in the human body exist. This review discusses the advantages, challenges, and promising clinical applications of UHF.
INVESTIGATIVE RADIOLOGY
(2021)
Article
Neurosciences
Balint S. Kornyei, Viktor Szabo, Gabor Perlaki, Bendeguz Balogh, Dorottya K. Szabo Steigerwald, Szilvia A. Nagy, Luca Toth, Andras Buki, Tamas Doczi, Peter Bogner, Attila Schwarcz, Arnold Toth
Summary: The study aimed to validate the phenomenon of TMBs becoming temporarily less detectable in SWI after traumatic brain injury. Results showed that TMBs may become less visible in SWI MRI during the subacute period of 24-72 hours post-injury.
FRONTIERS IN NEUROSCIENCE
(2021)
Review
Medicine, Research & Experimental
Enrico Capobianco, Marco Dominietto
Summary: This review discusses the potential applications of using multimodal imaging, radiomic data processing, and brain atlases in GBM studies, as well as the development of inference tools that can be generalized to other cancers. The focus is on building radiomic models from multimodal imaging data and translating suitably processed information into more accurate patient stratifications and evaluations of treatment efficacy using machine learning and other computational tools.
JOURNAL OF TRANSLATIONAL MEDICINE
(2023)
Article
Clinical Neurology
Kevin L. Tay, Stewart R. Leason, Laughlin C. Dawes, Sophia L. Thomas, Claudia M. Hillenbrand
Summary: This study investigated the validity of qualitative phase assessment on the cranial or caudal margins for distinguishing haemorrhage and calcification in lesions with ambiguous phase on SWI. Results showed that assessment at the cranial or caudal margins achieved a sensitivity of 100% for both haemorrhage and calcification, which was significantly superior to other methods such as dominant phase or in-plane margin assessment.
CLINICAL NEURORADIOLOGY
(2022)
Review
Medicine, General & Internal
Han-wen Zhang, Yuan-qing Zhang, Xiao-lei Liu, Yong-qian Mo, Yi Lei, Fan Lin, Yu-ning Feng
Summary: By studying 3 cases of Lhermitte-Duclos disease (LDD) and using fMRI and conventional MRI techniques, we found that brain tumor surgery is a good treatment for LDD. Magnetic resonance spectroscopy and susceptibility-weighted imaging combined with conventional MRI can be used to better diagnose LDD.
Review
Neurosciences
Giuseppe Barisano, Kirsten M. Lynch, Francesca Sibilia, Haoyu Lan, Nien-Chu Shih, Farshid Sepehrband, Jeiran Choupan
Summary: This article provides an overview of current neuroimaging methods for studying perivascular spaces (PVS) in humans using brain MRI. It highlights the increasing role of PVS in cerebrospinal/interstitial fluid circulation and waste product clearance, as well as their associations with neurological diseases. The article explores novel strategies and techniques for quantitatively analyzing the structure and function of PVS in humans, and provides guidance on acquisition protocols and analysis techniques. It also reviews neuroimaging studies on PVS across the lifespan and in neurological disorders.
Article
Clinical Neurology
Maarten Lequin, Floris Groenendaal, Jeroen Dudink, Paul Govaert
Summary: Diagnosing kernicterus in the acute phase is challenging, as the conventional T1w sequence may show high signal intensity due to early myelination. However, the SWI sequence appears to be more sensitive and reliable for detecting damage in the globus pallidum area.
Article
Radiology, Nuclear Medicine & Medical Imaging
Ashmita De, Hongfu Sun, Derek J. Emery, Kenneth S. Butcher, Alan H. Wilman
Summary: The purpose of this study was to optimize quantitative susceptibility-weighted imaging for strong susceptibility sources like hemorrhage and compare it to standard susceptibility-weighted imaging and quantitative susceptibility mapping. The results showed that quantitative susceptibility-weighted imaging minimized blooming effects and phase wrap artifacts observed in susceptibility-weighted imaging. However, it requires an altered upper threshold for best hemorrhage depiction.
MAGNETIC RESONANCE IMAGING
(2022)
Article
Computer Science, Interdisciplinary Applications
Datta Singh Goolaub, Jiawei Xu, Eric M. Schrauben, Davide Marini, John C. Kingdom, John G. Sled, Mike Seed, Christopher K. Macgowan
Summary: The paper presents a method for volumetric imaging of fetal flow using MRI. The method overcomes the challenges of small vascular structures and unpredictable motion through the use of orthogonal multislice stacks and motion correction. The results demonstrate the feasibility of the method in both adult and late gestation fetus, providing insights into complex flow pathways in human fetal circulation.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2022)
Article
Computer Science, Artificial Intelligence
Chanseok Lee, Gookho Song, Hyeonggeon Kim, Jong Chul Ye, Mooseok Jang
Summary: Reconstructing holographic images is difficult due to the ill posed inverse mapping problem. However, Lee and colleagues propose a deep learning method that incorporates a physical model and can handle physical perturbations in holographic image reconstruction, making it more reliable and applicable to a wider range of imaging problems.
NATURE MACHINE INTELLIGENCE
(2023)
Article
Clinical Neurology
Marjolaine Uginet, Gautier Breville, Jeremy Hofmeister, Paolo Machi, Patrice H. Lalive, Andrea Rosi, Aikaterini Fitsiori, Maria Isabel Vargas, Frederic Assal, Gilles Allali, Karl-Olof Lovblad
Summary: In patients with COVID-19 encephalopathy, vascular changes in the basilar and vertebral arteries, suggestive of endotheliitis, were found in a high prevalence, indicating a potential inflammatory mechanism underlying the condition.
CLINICAL NEURORADIOLOGY
(2022)
Review
Engineering, Biomedical
S. M. Seyedpour, S. Nafisi, M. Nabati, D. M. Pierce, J. R. Reichenbach, T. Ricken
Summary: MRI-based mathematical and computational modeling studies can enhance our understanding of cartilage mechanics and diseases, as well as optimize artificial cartilage production. These studies demonstrate the potential for an engineering-level approach to analyze the effects of cartilage diseases on material properties and function.
JOURNAL OF THE MECHANICAL BEHAVIOR OF BIOMEDICAL MATERIALS
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Christian Kames, Jonathan Doucette, Christoph Birkl, Alexander Rauscher
Summary: The study demonstrated the feasibility of recovering SWI-filtered phase using deep learning for QSM computation, showing that the recovered phase can effectively compute susceptibility maps with comparable accuracy to standard QSM processing. The network trained with all 13 processing methods exhibited good generalization performance and matched the reconstruction accuracy of networks trained on a single filter.
MAGNETIC RESONANCE IN MEDICINE
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Daniel Guellmar, Nina Jacobsen, Andreas Deistung, Dagmar Timmann, Stefan Ropele, Jurgen R. Reichenbach
Summary: The application of deep neural networks for segmentation in medical imaging has gained significant interest in recent years. This study systematically investigates the sensitivity of deep learning in medical image segmentation to variations in input data. The study also explores the effectiveness of different data augmentation strategies, providing a useful tool for selecting appropriate parameters for augmentation.
ZEITSCHRIFT FUR MEDIZINISCHE PHYSIK
(2022)
Article
Multidisciplinary Sciences
Mariya V. Cherkasova, Jessie F. Fu, Michael Jarrett, Poljanka Johnson, Shawna Abel, Roger Tam, Alexander Rauscher, Vesna Sossi, Shannon Kolind, David K. B. Li, A. Dessa Sadovnick, Lindsay Machan, J. Marc Girard, Francois Emond, Reza Vosoughi, Anthony Traboulsee, A. Jon Stoessl
Summary: Despite limited therapeutic effects, certain multiple sclerosis patients demonstrated a transient improvement in health-related quality of life, indicating a placebo response. The study found that placebo responders had higher lesion activity and exhibited a different cortical architecture compared to non-responders.
SCIENTIFIC REPORTS
(2022)
Article
Clinical Neurology
Hyunwoo Lee, Vanessa Wiggermann, Alexander Rauscher, Christian Kames, Mirza Faisal Beg, Karteek Popuri, Roger Tam, Kevin Lam, Claudia Jacova, Elham Shahinfard, Vesna Sossi, Jacqueline A. Pettersen, Ging-Yuek Robin Hsiung
Summary: This study explored the structural magnetic resonance imaging (MRI) abnormalities of mixed dementia (MixD) and found imaging characteristics specific to MixD, including higher burden of white-matter signal abnormalities (WMSA) on T1-weighted MRI, frontal lobar preponderance of WMSA, higher fractional anisotropy values within normal-appear white matter tissues, and lower R2* values within the T2-FLAIR WMSA areas.
CANADIAN JOURNAL OF NEUROLOGICAL SCIENCES
(2023)
Article
Neurosciences
Meng Li, Lena Vera Danyeli, Lejla Colic, Gerd Wagner, Stefan Smesny, Tara Chand, Xin Di, Bharat B. Biswal, Jorn Kaufmann, Juergen R. Reichenbach, Oliver Speck, Martin Walter, Zuemruet Duygu Sen
Summary: Reproducible resting-state functional connectivity patterns and their alterations have significant implications in neuropsychiatric research. This study utilizes multimodal imaging and magnetic resonance spectroscopy to investigate the correlation between regional neurotransmitter levels and rsFC strength, providing insights into the modulation of interaction between brain regions at a macroscopic level.
HUMAN BRAIN MAPPING
(2022)
Article
Clinical Neurology
Jeryn Chang, Thomas B. Shaw, Cory J. Holdom, Pamela A. McCombe, Robert D. Henderson, Jurgen Fripp, Markus Barth, Christine C. Guo, Shyuan T. Ngo, Frederik J. Steyn
Summary: This study found that lower hypothalamic volume is associated with lower and higher BMI in ALS patients, unlike AD patients and controls. Hypothalamic volume is not related to loss of appetite or hypermetabolism. In ALS patients, lower hypothalamic volume with lower BMI is associated with weight loss and earlier death.
EUROPEAN JOURNAL OF NEUROLOGY
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Aisling Fothergill, Christoph Birkl, Christian Kames, Wayne Su, Alexander Weber, Alexander Rauscher
Summary: This study investigated the impact of face mask wearing on cerebral blood flow (CBF) and cerebral venous oxygen saturation measured with MRI. The results showed that wearing a face mask had minimal effects on blood flow and oxygenation, with only CBF significantly increasing when wearing a 3-ply mask.
JOURNAL OF MAGNETIC RESONANCE IMAGING
(2023)
Article
Clinical Neurology
Vanessa Wiggermann, Verena Endmayr, Enedino Hernandez-Torres, Romana Hoeftberger, Gregor Kasprian, Simon Hametner, Alexander Rauscher
Summary: Magnetic resonance imaging (MRI) can provide important insights into multiple sclerosis (MS) by detecting focal or diffuse myelin damage or remyelination. This study used three myelin-sensitive MRI scans and histopathological measurements to evaluate the different stages of MS pathology, including chronic demyelinated and remyelinated lesions. The results showed that inactive lesions in chronic MS cases had increased myelin densities, indicating low-level remyelination.
Article
Radiology, Nuclear Medicine & Medical Imaging
Christian Kames, Jonathan Doucette, Alexander Rauscher
Summary: This study proposes a multi-echo field-to-source forward model that can reconstruct tissue magnetic susceptibility without fitting multi-echo phase data. The model achieves state-of-the-art results on the QSM Challenge 2.0 datasets. Additionally, a correction step is introduced to partially recover underestimated low-frequency susceptibility distributions.
MAGNETIC RESONANCE IN MEDICINE
(2023)
Article
Computer Science, Artificial Intelligence
Francesco Cognolato, Kieran O'Brien, Jin Jin, Simon Robinson, Frederik B. Laun, Markus Barth, Steffen Bollmann
Summary: Deep learning based Quantitative Susceptibility Mapping (QSM) has shown great potential, but current methods lack data consistency and often result in error propagation. We developed a new framework, NeXtQSM, that solves the QSM problem jointly and overcomes these limitations.
MEDICAL IMAGE ANALYSIS
(2023)
Article
Cell Biology
Sidra Gull, Christian Gaser, Karl-Heinz Herrmann, Anja Urbach, Marcus Boehme, Samia Afzal, Juergen R. Reichenbach, Otto W. Witte, Silvio Schmidt
Summary: By using MRI and DBM, we examined the structural changes in the brains of male RccHan:WIST rats. The study found that the overall brain volume increased with age, but there were also divergent local morphologic alterations. The visual, auditory, and somatosensory cortical areas showed shrinkage, while the higher-order brain areas such as the ectorhinal, entorhinal, retrosplenial, and cingulate cortical regions were preserved and grew with age.
Article
Radiology, Nuclear Medicine & Medical Imaging
Chia-Yin Wu, Jin Jin, Carl Dixon, Donald Maillet, Markus Barth, Martijn A. Cloos
Summary: Replacing standard excitation pulses with parallel transmit pulses can improve the efficiency of velocity selective labeling.
MAGNETIC RESONANCE IN MEDICINE
(2023)
Article
Neurosciences
Beata Bachrata, Steffen Bollmann, Jin Jin, Monique Tourell, Assunta Dal -Bianco, Siegfried Trattnig, Markus Barth, Stefan Ropele, Christian Enzinger, Simon Daniel Robinson
Summary: Quantitative Susceptibility Mapping (QSM) has the potential to provide additional insights into neurological diseases. We propose an ultra-fast acquisition method based on three orthogonal 2D simultaneous multislice EPI scans, which can generate high-resolution data in a short time. This method can be used to acquire QSM without additional imaging time.
Meeting Abstract
Neurosciences
Jacob Stubbs, Andrea Jones, Kristina Gicas, Alexandra Vertinsky, Manraj Heran, Wayne Su, Breanna Nelson, Donna Lang, Thalia Field, Allen Thornton, Alasdair Barr, Olga Leonova, William MacEwan, Alexander Rauscher, William Honer, William Panenka
BIOLOGICAL PSYCHIATRY
(2022)
Article
Computer Science, Artificial Intelligence
Hong Liu, Dong Wei, Donghuan Lu, Xiaoying Tang, Liansheng Wang, Yefeng Zheng
Summary: This study proposes a framework based on hybrid 2D-3D convolutional neural networks for obtaining continuous 3D retinal layer surfaces from OCT volumes. The framework works well with both full and sparse annotations and utilizes alignment displacement vectors and layer segmentation to align the B-scans and segment the layers. Experimental results show that the framework outperforms state-of-the-art 2D deep learning methods in terms of layer segmentation accuracy and cross-B-scan 3D continuity.
MEDICAL IMAGE ANALYSIS
(2024)
Article
Computer Science, Artificial Intelligence
Simon Oxenford, Ana Sofia Rios, Barbara Hollunder, Clemens Neudorfer, Alexandre Boutet, Gavin J. B. Elias, Jurgen Germann, Aaron Loh, Wissam Deeb, Bryan Salvato, Leonardo Almeida, Kelly D. Foote, Robert Amaral, Paul B. Rosenberg, David F. Tang-Wai, David A. Wolk, Anna D. Burke, Marwan N. Sabbagh, Stephen Salloway, M. Mallar Chakravarty, Gwenn S. Smith, Constantine G. Lyketsos, Michael S. Okun, William S., Zoltan Mari, Francisco A. Ponce, Andres Lozano, Wolf-Julian Neumann, Bassam Al-Fatly, Andreas Horn
Summary: Spatial normalization is a method to map subject brain images to an average template brain, allowing comparison of brain imaging results. We introduce a novel tool called WarpDrive, which enables manual refinements of image alignment after automated registration. The tool improves accuracy of data representation and aids in understanding patient outcomes.
MEDICAL IMAGE ANALYSIS
(2024)
Article
Computer Science, Artificial Intelligence
Ricards Marcinkevics, Patricia Reis Wolfertstetter, Ugne Klimiene, Kieran Chin-Cheong, Alyssia Paschke, Julia Zerres, Markus Denzinger, David Niederberger, Sven Wellmann, Ece Ozkan, Christian Knorr, Julia E. Vogt
Summary: This study presents interpretable machine learning models for predicting the diagnosis, management, and severity of suspected appendicitis using ultrasound images. The proposed models utilize concept bottleneck models (CBM) that facilitate interpretation and intervention by clinicians, without compromising performance or requiring time-consuming image annotation.
MEDICAL IMAGE ANALYSIS
(2024)
Article
Computer Science, Artificial Intelligence
Jian-Qing Zheng, Ziyang Wang, Baoru Huang, Ngee Han Lim, Bartlomiej W. Papiez
Summary: This article introduces a new method for medical image registration, which utilizes a separable motion backbone and a residual aligner module to better handle the discontinuous motion of multiple neighboring objects. The proposed method achieves excellent registration results on abdominal CT scans and lung CT scans.
MEDICAL IMAGE ANALYSIS
(2024)
Article
Computer Science, Artificial Intelligence
Xiangqiong Wu, Guanghua Tan, Hongxia Luo, Zhilun Chen, Bin Pu, Shengli Li, Kenli Li
Summary: This study develops a user-friendly framework for the automated diagnosis of thyroid nodules in ultrasound videos, simulating the diagnostic workflow of radiologists. By interpreting image characteristics and modeling temporal contextual information, the efficiency and generalizability of the diagnosis can be improved.
MEDICAL IMAGE ANALYSIS
(2024)
Article
Computer Science, Artificial Intelligence
Riddhish Bhalodia, Shireen Elhabian, Jadie Adams, Wenzheng Tao, Ladislav Kavan, Ross Whitaker
Summary: This paper introduces DeepSSM, a deep learning-based framework for image-to-shape modeling. By learning the functional mapping from images to low-dimensional shape descriptors, DeepSSM can directly infer statistical representation of anatomy from 3D images. Compared to traditional methods, DeepSSM eliminates the need for heavy manual preprocessing and segmentation, and significantly improves computational time.
MEDICAL IMAGE ANALYSIS
(2024)
Article
Computer Science, Artificial Intelligence
Florentin Liebmann, Marco von Atzigen, Dominik Stutz, Julian Wolf, Lukas Zingg, Daniel Suter, Nicola A. Cavalcanti, Laura Leoty, Hooman Esfandiari, Jess G. Snedeker, Martin R. Oswald, Marc Pollefeys, Mazda Farshad, Philipp Furnstahl
Summary: This study presents a marker-less approach for automatic registration and real-time navigation of lumbar spinal fusion surgery using a deep neural network, avoiding radiation exposure and surgical errors. The method was validated on an ex-vivo surgery and a public dataset.
MEDICAL IMAGE ANALYSIS
(2024)
Article
Computer Science, Artificial Intelligence
Piyush Tiwary, Kinjawl Bhattacharyya, A. P. Prathosh
Summary: Domain shift refers to the change of distributional characteristics between training and testing datasets, leading to performance drop. For medical image tasks, domain shift can be caused by changes in imaging modalities, devices, and staining mechanisms. Existing approaches based on generative models suffer from training difficulties and lack of diversity. In this paper, the authors propose the use of energy-based models (EBMs) for unpaired image-to-image translation in medical images. The proposed method, called Cycle Consistent Twin EBMs (CCT-EBM), employs a pair of EBMs in the latent space of an Auto-Encoder to ensure translation symmetry and coupling between domains.
MEDICAL IMAGE ANALYSIS
(2024)
Article
Computer Science, Artificial Intelligence
Yutong Xie, Jianpeng Zhang, Lingqiao Liu, Hu Wang, Yiwen Ye, Johan Verjans, Yong Xia
Summary: This paper proposes a hybrid pre-training paradigm that combines self-supervised learning and supervised learning to improve the representation quality for medical image segmentation tasks. It introduces a reference task in self-supervised learning and optimizes the model using a gradient matching method. The experimental results demonstrate the effectiveness of this approach on multiple medical image segmentation benchmarks.
MEDICAL IMAGE ANALYSIS
(2024)
Article
Computer Science, Artificial Intelligence
Youyi Song, Jing Zou, Kup-Sze Choi, Baiying Lei, Jing Qin
Summary: Cell classification is crucial for intelligent cervical cancer screening, but the variation in cells' appearance and shape poses challenges. A new learning algorithm, worse-case boosting, is proposed to improve classification accuracy for under-represented data. Experimental results demonstrate the effectiveness of this algorithm in two publicly available datasets, achieving a 4% improvement in accuracy.
MEDICAL IMAGE ANALYSIS
(2024)
Article
Computer Science, Artificial Intelligence
Sangjoon Park, Eun Sun Lee, Kyung Sook Shin, Jeong Eun Lee, Jong Chul Ye
Summary: The increasing demand for AI systems to monitor human errors and abnormalities in healthcare presents challenges. This study presents a model called Medical X-VL, which is tailored for the medical domain and outperformed current state-of-the-art models in two medical image datasets. The model enables various zero-shot tasks for monitoring AI in the medical domain.
MEDICAL IMAGE ANALYSIS
(2024)
Article
Computer Science, Artificial Intelligence
Anna Klimovskaia Susmelj, Berkan Lafci, Firat Ozdemir, Neda Davoudi, Xose Luis Dean-Ben, Fernando Perez-Cruz, Daniel Razansky
Summary: Optoacoustic imaging is a technique that uses optical excitation and ultrasound detection for biological tissue imaging. The quality of the images depends on the extent of tomographic coverage provided by the ultrasound detector arrays. However, full coverage is not always possible due to experimental constraints. The proposed signal domain adaptation network aims to reduce limited-view artifacts in the images.
MEDICAL IMAGE ANALYSIS
(2024)
Article
Computer Science, Artificial Intelligence
Srijay Deshpande, Muhammad Dawood, Fayyaz Minhas, Nasir Rajpoot
Summary: In this work, a novel framework called SynCLay is proposed for automated synthesis of histology images based on user-defined cellular layouts. The framework can generate realistic and high-quality histology images with different cellular arrangements, which is helpful for studying the role of cells in the tumor microenvironment. The framework integrates a nuclear segmentation and classification model to refine nuclear structures and generate nuclear masks. Evaluation using quantitative metrics and feedback from pathologists shows that the synthetic images generated by SynCLay have high realism scores and can accurately differentiate between benign and malignant tumors.
MEDICAL IMAGE ANALYSIS
(2024)
Article
Computer Science, Artificial Intelligence
Ahmed H. Shahin, An Zhao, Alexander C. Whitehead, Daniel C. Alexander, Joseph Jacob, David Barber
Summary: Survival analysis is a valuable tool in healthcare for predicting the time to specific events. This paper introduces CenTime, a novel approach that directly estimates the time to event. The method performs well with censored data and can be easily integrated with deep learning models. Compared to standard methods, CenTime offers superior performance in predicting event time while maintaining comparable ranking performance.
MEDICAL IMAGE ANALYSIS
(2024)
Article
Computer Science, Artificial Intelligence
Bingyuan Liu, Jose Dolz, Adrian Galdran, Riadh Kobbi, Ismail Ben Ayed
Summary: Most segmentation losses, such as CE and Dice, are variants of the Cross-Entropy or Dice losses. This work provides a theoretical analysis that shows a deeper connection between CE and Dice than previously thought. From a constrained-optimization perspective, both CE and Dice decompose into similar ground-truth matching terms and region-size penalty terms. The analysis uncovers hidden region-size biases: Dice has an intrinsic bias towards extremely imbalanced solutions, while CE implicitly encourages the ground-truth region proportions. Based on this analysis, a principled and simple solution is proposed to explicitly control the region-size bias.
MEDICAL IMAGE ANALYSIS
(2024)
