检索结果-内蒙古大学图书馆

Deep learning in medical image registration: magic or mirage? 24

学校读者我要写书评

暂无评论

Deep learning in medical image registration: magic or mirage...

Proceedings of the 38th International Conference on Neural information Processing Systems

作者： Rohit Jena Deeksha Sethi Pratik Chaudhari James C. Gee Computer and Information Science and Penn Image Computing and Science Laboratory Computer and Information Science Computer and Information Science and Electrical and Systems Engineering Computer and Information Science and Radiology and Penn Image Computing and Science Laboratory

ISBN: (纸本)9798331314385

关键词：

Deep Learning in Medical image Registration: Magic or Mirage?

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Jena, Rohit Sethi, Deeksha Chaudhari, Pratik Gee, James C. Computer and Information Science Electrical and Systems Engineering Radiology Penn Image Computing and Science Laboratory United States

Classical optimization and learning-based methods are the two reigning paradigms in deformable image registration. While optimization-based methods boast generalizability across modalities and robust performance, learning-based methods promise peak performance, incorporating weak supervision and amortized optimization. However, the exact conditions for either paradigm to perform well over the other are shrouded and not explicitly outlined in the existing literature. In this paper, we make an explicit correspondence between the mutual information of the distribution of per-pixel intensity and labels, and the performance of classical registration methods. This strong correlation hints to the fact that architectural designs in learning-based methods is unlikely to affect this correlation, and therefore, the performance of learning-based methods. This hypothesis is thoroughly validated with state-of-the-art classical and learning-based methods. However, learning-based methods with weak supervision can perform high-fidelity intensity and label registration, which is not possible with classical methods. Next, we show that this high-fidelity feature learning does not translate to invariance to domain shift, and learning-based methods are sensitive to such changes in the data distribution. Finally, we propose a general recipe to choose the best paradigm for a given registration problem, based on these observations. © 2024, CC BY.

关键词： Contrastive Learning

FireANTs: Adaptive Riemannian Optimization for Multi-Scale Diffeomorphic Matching

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Jena, Rohit Chaudhari, Pratik Gee, James C. Computer and Information Science University of Pennsylvania United States Electrical and Systems Engineering University of Pennsylvania United States Radiology Perelman School of Medicine University of Pennsylvania United States Penn Image Computing and Science Laboratory University of Pennsylvania United States

The paper proposes FireANTs, the first multi-scale Adaptive Riemannian Optimization algorithm for dense diffeomorphic image matching. One of the most critical and understudied aspects of diffeomorphic image matching algorithms are its highly ill-conditioned nature. We quantitatively capture the extent of ill-conditioning in a typical MRI matching task, motivating the need for an adaptive optimization algorithm for diffeomorphic matching. To this end, FireANTs generalizes the concept of momentum and adaptive estimates of the Hessian to mitigate this ill-conditioning in the non-Euclidean space of diffeomorphisms. Unlike common non-Euclidean manifolds, we also formalize considerations for multi-scale optimization of diffeomorphisms. Our rigorous mathematical results and operational contributions lead to a state-of-the-art dense matching algorithm that can be applied to generic image data with remarkable accuracy and robustness. We demonstrate consistent improvements in image matching performance across a spectrum of community-standard medical and biological correspondence matching challenges spanning a wide variety of image modalities, anatomies, resolutions, acquisition protocols, and preprocessing pipelines. This improvement is supplemented by from 300× up to 3200× speedup over existing state-of-the-art algorithms. For the first time, we perform diffeomorphic matching of sub-micron mouse cortex volumes at native resolution. Our fast implementation also enables hyperparameter studies that were intractable with existing correspondence matching algorithms. © 2024, CC BY.

关键词： Neuroimaging

Compartment-specific estimation of T2 and T2* with diffusion-PEPTIDE MRI

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Gong, Ting Fair, Merlin J. Setsompop, Kawin Zhang, Hui Centre for Medical Image Computing Department of Computer Science University College London London United Kingdom Radiological Sciences Laboratory Department of Radiology Stanford University StanfordCA United States Department of Electrical Engineering Stanford University StanfordCA United States

We present a microstructure imaging technique for estimating compartment-specific T2 and T2* simultaneously in the human brain. Microstructure imaging with diffusion MRI (dMRI) has enabled the modelling of intra-neurite and extra-neurite diffusion signals separately allowing for the estimation of compartment-specific tissue properties. These compartment-specific properties have been widely used in clinical studies. However, conventional dMRI cannot disentangle differences in relaxations between tissue compartments, causing biased estimates of diffusion measures which also change with TE. To solve the problem, combined relaxometry-diffusion imaging methods have been developed in recent years, providing compartmental T2-diffusion or T2*-diffusion imaging respectively, but not T2 and T2* together. As they provide complementary information, a technique that can estimate both jointly with diffusion is appealing to neuroimaging studies. The aim of this work is to develop a method to map compartmental T2-T2*-diffusion simultaneously. Using an advanced MRI acquisition called diffusion-PEPTIDE, a novel microstructure model is proposed and a multi-step fitting method is developed to estimate parameters of interest. We demonstrate for the first time that compartmental T2, T2* can be estimated simultaneously from in vivo data. we further show the accuracy and precision of parameter estimation with simulation. Copyright © 2024, The Authors. All rights reserved.

关键词： Tissue

Optimizing the Size of Peritumoral Region for Assessing Non-Small Cell Lung Cancer Heterogeneity Using Radiomics 12th

学校读者我要写书评

暂无评论

Optimizing the Size of Peritumoral Region for Assessing No...

12th International Conference on Health information science, HIS 2023

作者： Zhang, Xingping Zhang, Guijuan Qiu, Xingting Yin, Jiao Tan, Wenjun Yin, Xiaoxia Yang, Hong Wang, Kun Zhang, Yanchun Cyberspace Institute of Advanced Technology Guangzhou University Guangzhou510006 China School of Computer Science and Technology Zhejiang Normal University Jinhua321000 China Department of Respiratory and Critical Care First Affiliated Hospital of Gannan Medical University Ganzhou341000 China Institute for Sustainable Industries and Liveable Cities Victoria University Melbourne3011 Australia Department of New Networks Peng Cheng Laboratory Shenzhen518000 China Department of Radiology First Affiliated Hospital of Gannan Medical University Ganzhou341000 China Key Laboratory of Intelligent Computing in Medical Image Ministry of Education Northeastern University Shenyang110189 China

ISBN: (纸本)9789819971077

Objectives: Radiomics has a novel value in accurately and noninvasively characterizing non-small cell lung cancer (NSCLC), but the role of peritumoral features has not been discussed in depth. This work aims to systematically assess the additional value of peritumoral features by exploring the impact of peritumoral region size. Materials and methods: A total of 370 NSCLC patients who underwent preoperative contrast-enhanced CT scans between October 2017 and September 2021 were retrospectively analyzed. The study was carefully designed with a radiomics pipeline to predict lymphovascular invasion, pleural invasion, and T4 staging. To assess the impact of peritumoral features, tumor regions of interest (ROIs) annotated by two medical experts were automatically expanded to produce peritumoral ROIs of different regional sizes, with edge thicknesses of 1 mm, 3 mm, 5 mm, and 7 mm. In a custom pipeline, prediction models were constructed using peritumoral features with different margin thicknesses and intratumoral features of the primary tumor. Results: Radiomics features combining intratumoral and peritumoral regions were created based on the best features of each ROI. Models incorporating peritumoral features yielded varying degrees of improvement in AUCs compared to models using only intratumoral features. The choice of peritumoral size may impact the degree of improvement in radiomics analysis. Conclusions: The integration of peritumoral features has shown potential for improving the predictive value of radiomics. However, selecting an appropriate peritumoral region size is constrained by various factors such as clinical issues, imaging modalities, and ROI annotations. Therefore, future radiomics studies should consider these factors and optimize peritumoral features to cater to specific applications. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2023.

关键词： Tumors

Mutually-Constrained Cross-Sectional and Longitudinal Non-Negative Matrix Factorization: Application to Modeling Brain Aging Trajectories

学校读者我要写书评

暂无评论

Mutually-Constrained Cross-Sectional and Longitudinal Non-Ne...

IEEE International Symposium on Biomedical Imaging

作者： Ioanna Skampardoni Junhao Wen Erus Guray Haochang Shou Konstantina Nikita Christos Davatzikos Centre for Biomedical Image Computing and Analytics University of Pennsylvania Philadelphia PA USA School of Electrical and Computer Engineering National Technical University of Athens Athens Greece Laboratory of AI and Biomedical Science (LABS) Stevens Neuroimaging and Informatics Institute Keck School of Medicine of USC University of Southern California Los Angeles CA USA Department of Biostatistics Epidemiology & Informatics Penn Statistics in Imaging and Visualization Center University of Pennsylvania Philadelphia PA USA

ISBN: (数字)9798350313338

ISBN: (纸本)9798350313345

Brain aging is a multifaceted and highly heterogeneous process accompanied by several pathologies. Here, we propose a method for dissecting the heterogeneity of neuropathologic processes occurring with aging using machine learning and leveraging information from cross-sectional and longitudinal data. Specifically, we hypothesize that the heterogeneity observed in brain aging can be captured by a set of patterns consistent with longitudinal trajectories of brain change, the latter directly capturing evolving neuropathologic processes on an individual basis. Applying the method to structural magnetic resonance imaging data from the BLSA study, we derived five distinct, reproducible, and clinically informative components of neuroanatomical brain change, highlighting the method’s potential as a tool for precision medicine.

关键词： Pathology Magnetic resonance imaging Precision medicine Sociology Aging Hazards Trajectory

USFM: A Universal Ultrasound Foundation Model Generalized to Tasks and Organs towards Label Efficient image Analysis

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Jiao, Jing Zhou, Jin Li, Xiaokang Xia, Menghua Huang, Yi Huang, Lihong Wang, Na Zhang, Xiaofan Zhou, Shichong Wang, Yuanyuan Guo, Yi Department of Electronic Engineering School of Information Science and Technology Fudan University Shanghai China Fudan University Shanghai Cancer Center Shanghai China Department of Radiology and Biomedical Imaging Yale School of Medicine New HavenCT United States SenseTime Research Shanghai China Shanghai Artificial Intelligence Laboratory Shanghai China Key Laboratory of Medical Imaging Computing and Computer Assisted Intervention of Shanghai Shanghai China

Inadequate generality across different organs and tasks constrains the application of ultrasound (US) image analysis methods in smart healthcare. Building a universal US foundation model holds the potential to address these issues. Nevertheless, the development of such foundational models encounters intrinsic challenges in US analysis, i.e., insufficient databases, low quality, and ineffective features. In this paper, we present a universal US foundation model, named USFM, generalized to diverse tasks and organs towards label efficient US image analysis. First, a large-scale Multi-organ, Multi-center, and Multi-device US database was built, comprehensively containing over two million US images. Organ-balanced sampling was employed for unbiased learning. Then, USFM is self-supervised pre-trained on the sufficient US database. To extract the effective features from low-quality US images, we proposed a spatial-frequency dual masked image modeling method. A productive spatial noise addition-recovery approach was designed to learn meaningful US information robustly, while a novel frequency band-stop masking learning approach was also employed to extract complex, implicit grayscale distribution and textural variations. Extensive experiments were conducted on the various tasks of segmentation, classification, and image enhancement from diverse organs and diseases. Comparisons with representative US image analysis models illustrate the universality and effectiveness of USFM. The label efficiency experiments suggest the USFM obtains robust performance with only 20% annotation, laying the groundwork for the rapid development of US models in clinical practices. Copyright © 2023, The Authors. All rights reserved.

关键词： Efficiency

Treatment Outcome Prediction for Intracerebral Hemorrhage via Generative Prognostic Model with Imaging and Tabular Data

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Ma, Wenao Chen, Cheng Abrigo, Jill Mak, Calvin Hoi-Kwan Gong, Yuqi Chan, Nga Yan Han, Chu Liu, Zaiyi Dou, Qi Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong Center for Advanced Medical Computing and Analysis Harvard Medical School Boston United States Department of Imaging and Interventional Radiology The Chinese University of Hong Kong Hong Kong Department of Neurosurgery Queen Elizabeth Hospital Hong Kong Southern Medical University Guangzhou China Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application Guangzhou China

Intracerebral hemorrhage (ICH) is the second most common and deadliest form of stroke. Despite medical advances, predicting treatment outcomes for ICH remains a challenge. This paper proposes a novel prognostic model that utilizes both imaging and tabular data to predict treatment outcome for ICH. Our model is trained on observational data collected from non-randomized controlled trials, providing reliable predictions of treatment success. Specifically, we propose to employ a variational autoencoder model to generate a low-dimensional prognostic score, which can effectively address the selection bias resulting from the non-randomized controlled trials. Importantly, we develop a variational distributions combination module that combines the information from imaging data, non-imaging clinical data, and treatment assignment to accurately generate the prognostic score. We conducted extensive experiments on a real-world clinical dataset of intracerebral hemorrhage. Our proposed method demonstrates a substantial improvement in treatment outcome prediction compared to existing state-of-the-art approaches. Code is available at https://***/med-air/TOP-GPM. Copyright © 2023, The Authors. All rights reserved.

关键词： Forecasting

Aging and Alzheimer’s Disease Have Dissociable Effects on Local and Regional Medial Temporal Lobe Connectivity

学校读者我要写书评

暂无评论

Alzheimer's & Dementia 2023年第S10期19卷

作者： Stanislau Hrybouski Sandhitsu R. Das Long Xie Laura EM Wisse Melissa Kelley Jacqueline Lane Monica Sherin Michael DiCalogero Ilya M. Nasrallah John A. Detre Paul A. Yushkevich David A. Wolk University of Pennsylvania Philadelphia PA USA Penn Image Computing and Science Laboratory (PICSL) University of Pennsylvania Philadelphia PA USA Lund University Lund Sweden Department of Radiology University of Pennsylvania Philadelphia PA USA

Background The extent to which pathological processes in aging and Alzheimer’s disease (AD) relate to functional disruption of the medial temporal lobe (MTL)-dependent brain networks is poorly understood. To address this knowledge gap, we examined functional connectivity (FC) alterations between anterior and posterior regions of the MTL and in MTL-associated functional communities – the Anterior-Temporal (AT) and Posterior-Medial (PM) networks – in normal agers, individuals with preclinical AD, and patients with Mild Cognitive Impairment or mild dementia due to AD. Method In this cross-sectional study, we analyzed data from 179 individuals from the Aging Brain Cohort study of the penn ADRC. Detailed information about participants is provided in Table 1. For intra-MTL FC comparisons, the MTL subregions were segmented using the automated segmentation of hippocampal subfields-T1 (ASHS-T1) pipeline (Fig. 1a). When modeling the MTL’s interactions with the rest of the cortex, we employed four MTL ROIs (left/right × anterior/posterior) derived from an ex vivo atlas of tau accumulation in the MTL. Our functional datasets were preprocessed using a customized fMRIprep pipeline. Sparse network estimations and modularity-based consensus clustering were used to reconstruct the AT and PM network systems (Fig. 1b). Age effect analyses and group comparisons along the AD continuum were performed using the General Linear Model within the network-based statistical framework. Result The preclinical stage of AD was characterized by increased FC between the perirhinal cortex and other regions of the MTL, as well as between the anterior MTL and its direct neighbors in the AT network (Fig. 1c-d). This effect was not present in symptomatic AD. Instead, symptomatic patients displayed reduced hippocampal and intra-PM connectivity. For normal aging, our results led to three main conclusions (for visuals, see Fig. 2). First, intra-network connectivity of both the AT and PM networks decreases wi

关键词：