paper_authors: Charlie Budd, Jianrong Qiu, Oscar MacCormac, Martin Huber, Christopher Mower, Mirek Janatka, Théo Trotouin, Jonathan Shapey, Mads S. Bergholt, Tom Vercauteren
results: 研究发现，使用深度学习自适应关注法可以提高激光成像系统的焦点精度，并且在比较 tradicional 焦点调整策略时表现出 statistically significan advantage。此外，两名 neurosurgeon 在不知情的用户测试中，对不同的焦点调整策略进行比较，并评价了我们的新方法，发现其最为满意。Abstract
Hyperspectral imaging (HSI) captures a greater level of spectral detail than traditional optical imaging, making it a potentially valuable intraoperative tool when precise tissue differentiation is essential. Hardware limitations of current optical systems used for handheld real-time video HSI result in a limited focal depth, thereby posing usability issues for integration of the technology into the operating room. This work integrates a focus-tunable liquid lens into a video HSI exoscope, and proposes novel video autofocusing methods based on deep reinforcement learning. A first-of-its-kind robotic focal-time scan was performed to create a realistic and reproducible testing dataset. We benchmarked our proposed autofocus algorithm against traditional policies, and found our novel approach to perform significantly ($p<0.05$) better than traditional techniques ($0.070\pm.098$ mean absolute focal error compared to $0.146\pm.148$). In addition, we performed a blinded usability trial by having two neurosurgeons compare the system with different autofocus policies, and found our novel approach to be the most favourable, making our system a desirable addition for intraoperative HSI.
摘要
This work integrates a focus-tunable liquid lens into a video HSI exoscope and proposes novel video autofocusing methods based on deep reinforcement learning. A first-of-its-kind robotic focal-time scan was performed to create a realistic and reproducible testing dataset. The proposed autofocus algorithm was benchmarked against traditional policies, and the results showed that the novel approach performed significantly better ($p<0.05$) than traditional techniques ($0.070\pm.098$ mean absolute focal error compared to $0.146\pm.148$).In addition, a blinded usability trial was conducted by having two neurosurgeons compare the system with different autofocus policies, and the results showed that the novel approach was the most favourable, making the system a desirable addition for intraoperative HSI.
For: The paper is focused on the concept of computational image formation (CIF) and its applications in imaging through adverse weather conditions.* Methods: The paper introduces the idea of using an approximate mapping $\mathcal{H}_{\theta}$ to simulate the forward degradation process in imaging, and discusses the attributes of a CIF simulator, including accuracy, speed, well-posedness, and differentiability.* Results: The paper provides a detailed case study on imaging through atmospheric turbulence using CIF, and discusses other examples of CIF applications. The paper also shares thoughts on the future direction and recommendations for the community.Abstract
At the pinnacle of computational imaging is the co-optimization of camera and algorithm. This, however, is not the only form of computational imaging. In problems such as imaging through adverse weather, the bigger challenge is how to accurately simulate the forward degradation process so that we can synthesize data to train reconstruction models and/or integrating the forward model as part of the reconstruction algorithm. This article introduces the concept of computational image formation (CIF). Compared to the standard inverse problems where the goal is to recover the latent image $\mathbf{x}$ from the observation $\mathbf{y} = \mathcal{G}(\mathbf{x})$, CIF shifts the focus to designing an approximate mapping $\mathcal{H}_{\theta}$ such that $\mathcal{H}_{\theta} \approx \mathcal{G}$ while giving a better image reconstruction result. The word ``computational'' highlights the fact that the image formation is now replaced by a numerical simulator. While matching nature remains an important goal, CIF pays even greater attention on strategically choosing an $\mathcal{H}_{\theta}$ so that the reconstruction performance is maximized. The goal of this article is to conceptualize the idea of CIF by elaborating on its meaning and implications. The first part of the article is a discussion on the four attributes of a CIF simulator: accurate enough to mimic $\mathcal{G}$, fast enough to be integrated as part of the reconstruction, providing a well-posed inverse problem when plugged into the reconstruction, and differentiable in the backpropagation sense. The second part of the article is a detailed case study based on imaging through atmospheric turbulence. The third part of the article is a collection of other examples that fall into the category of CIF. Finally, thoughts about the future direction and recommendations to the community are shared.
摘要
在计算成像领域的尽头是合理化相机和算法的共同优化。然而，这并不是唯一的计算成像方式。在如影像受恶征天气等问题中，更大的挑战是如何准确地模拟前向干扰过程，以便可以合成数据来训练重建模型和/或将前向模型纳入重建算法中。这篇文章介绍了计算成像形成（CIF）的概念。与标准的逆问题where the goal is to recover the latent image $\mathbf{x}$ from the observation $\mathbf{y} = \mathcal{G}(\mathbf{x})$不同，CIF将注意力集中在设计一个approximate mapping $\mathcal{H}_{\theta}$，使其 approximate $\mathcal{G}$，同时提供更好的图像重建结果。“计算”一词 highlights the fact that the image formation is now replaced by a numerical simulator。而Matching nature remains an important goal，CIF pays even greater attention on strategically choosing an $\mathcal{H}_{\theta}$ so that the reconstruction performance is maximized。本文的目标是把CIF的概念进行详细说明和分析，包括其意义和影响。文中的第一部分是讨论CIF simulator的四个特征：准确地模拟 $\mathcal{G}$，快速 enough to be integrated as part of the reconstruction，提供一个准确的逆问题when plugged into the reconstruction，以及在反射卷积中可微分。第二部分是基于大气扩散的详细案例研究。第三部分是收集其他符合CIF的例子。最后，文章结束了未来方向和对社区的建议。
Cascaded multitask U-Net using topological loss for vessel segmentation and centerline extraction
results: 这paper的实验结果表明，使用U-Net计算血管skeleton可以提高分割结果的准确性，并且可以提供更加准确的血管中心线。Abstract
Vessel segmentation and centerline extraction are two crucial preliminary tasks for many computer-aided diagnosis tools dealing with vascular diseases. Recently, deep-learning based methods have been widely applied to these tasks. However, classic deep-learning approaches struggle to capture the complex geometry and specific topology of vascular networks, which is of the utmost importance in most applications. To overcome these limitations, the clDice loss, a topological loss that focuses on the vessel centerlines, has been recently proposed. This loss requires computing, with a proposed soft-skeleton algorithm, the skeletons of both the ground truth and the predicted segmentation. However, the soft-skeleton algorithm provides suboptimal results on 3D images, which makes the clDice hardly suitable on 3D images. In this paper, we propose to replace the soft-skeleton algorithm by a U-Net which computes the vascular skeleton directly from the segmentation. We show that our method provides more accurate skeletons than the soft-skeleton algorithm. We then build upon this network a cascaded U-Net trained with the clDice loss to embed topological constraints during the segmentation. The resulting model is able to predict both the vessel segmentation and centerlines with a more accurate topology.
摘要
船 Segmentation 和中心线抽取是许多计算机支持诊断工具处理血管疾病的两项重要前置任务。现在，深度学习基于方法广泛应用于这两项任务。然而， classical deep-learning 方法很难捕捉血管网络的复杂 геометри和特有的 topologic，这在大多数应用中是非常重要的。为了解决这些限制，recently proposed clDice loss 强调血管中心线的 topological loss。这个损失函数需要计算，使用我们提议的 soft-skeleton 算法，真实的血管skeleton 和预测 segmentation 的skeleton。然而，soft-skeleton 算法在 3D 图像上提供了不佳的结果，使 clDice hardly suitable for 3D images。在本文中，我们提议将 soft-skeleton 算法 replaced by a U-Net，从 segmentation 直接计算血管skeleton。我们示出了我们的方法可以提供更加准确的 skeletons。然后，我们在这个网络上建立了一个 cascaded U-Net，通过 clDice 损失函数 embedding topological constraints During the segmentation。结果是一个能够预测血管 segmentation 和中心线的更加准确 topology。
CortexMorph: fast cortical thickness estimation via diffeomorphic registration using VoxelMorph
methods: 该方法使用了不监督的深度学习来直接预测 deformation field，以便用 DiReCT 方法计算 cortical thickness。
results: 研究表明，通过结合 CortexMorph 和深度学习基于的 segmentation 模型，可以在秒钟内从 T1 束缚图像中计算区域化 cortical thickness，同时保持检测 cortical atrophy 的能力。Abstract
The thickness of the cortical band is linked to various neurological and psychiatric conditions, and is often estimated through surface-based methods such as Freesurfer in MRI studies. The DiReCT method, which calculates cortical thickness using a diffeomorphic deformation of the gray-white matter interface towards the pial surface, offers an alternative to surface-based methods. Recent studies using a synthetic cortical thickness phantom have demonstrated that the combination of DiReCT and deep-learning-based segmentation is more sensitive to subvoxel cortical thinning than Freesurfer. While anatomical segmentation of a T1-weighted image now takes seconds, existing implementations of DiReCT rely on iterative image registration methods which can take up to an hour per volume. On the other hand, learning-based deformable image registration methods like VoxelMorph have been shown to be faster than classical methods while improving registration accuracy. This paper proposes CortexMorph, a new method that employs unsupervised deep learning to directly regress the deformation field needed for DiReCT. By combining CortexMorph with a deep-learning-based segmentation model, it is possible to estimate region-wise thickness in seconds from a T1-weighted image, while maintaining the ability to detect cortical atrophy. We validate this claim on the OASIS-3 dataset and the synthetic cortical thickness phantom of Rusak et al.
摘要
cortical 带的厚度与多种神经内科和心理科学疾病相关，通常通过表面基本方法如Freesurfer在MRI研究中估算。DiReCT方法，它通过Diffomorphic deformation of gray-white matter interface towards the pial surface来计算 cortical thickness，为表面基本方法提供了一种alternative。最近的研究使用了Synthetic cortical thickness phantom表明，combined DiReCT and deep-learning-based segmentation more sensitive to subvoxel cortical thinning than Freesurfer。 whereas anatomical segmentation of a T1-weighted image now takes only seconds, existing implementations of DiReCT rely on iterative image registration methods that can take up to an hour per volume. On the other hand, learning-based deformable image registration methods like VoxelMorph have been shown to be faster than classical methods while improving registration accuracy. This paper proposes CortexMorph, a new method that employs unsupervised deep learning to directly regress the deformation field needed for DiReCT. By combining CortexMorph with a deep-learning-based segmentation model, it is possible to estimate region-wise thickness in seconds from a T1-weighted image, while maintaining the ability to detect cortical atrophy. We validate this claim on the OASIS-3 dataset and the synthetic cortical thickness phantom of Rusak et al.
FedAutoMRI: Federated Neural Architecture Search for MR Image Reconstruction
paper_authors: Ruoyou Wu, Cheng Li, Juan Zou, Shanshan Wang
for: 用于MR图像重建
methods: 使用分布式协同学习方法和不同数据分布稳定化方法
results: 实现了较好的表现，使用轻量级模型并且比经典联合学习方法具有更少的参数数量Abstract
Centralized training methods have shown promising results in MR image reconstruction, but privacy concerns arise when gathering data from multiple institutions. Federated learning, a distributed collaborative training scheme, can utilize multi-center data without the need to transfer data between institutions. However, existing federated learning MR image reconstruction methods rely on manually designed models which have extensive parameters and suffer from performance degradation when facing heterogeneous data distributions. To this end, this paper proposes a novel FederAted neUral archiTecture search approach fOr MR Image reconstruction (FedAutoMRI). The proposed method utilizes differentiable architecture search to automatically find the optimal network architecture. In addition, an exponential moving average method is introduced to improve the robustness of the client model to address the data heterogeneity issue. To the best of our knowledge, this is the first work to use federated neural architecture search for MR image reconstruction. Experimental results demonstrate that our proposed FedAutoMRI can achieve promising performances while utilizing a lightweight model with only a small number of model parameters compared to the classical federated learning methods.
摘要
中央化训练方法在MR图像重建中表现出了扎实的成果，但是收集数据从多家机构时，隐私问题就会出现。基于分布式合作训练的联邦学习（Federated Learning）可以利用多中心数据无需将数据传输到机构之间。然而，现有的联邦学习MR图像重建方法通常采用手动设计的模型，这些模型具有较多的参数，并且面临着数据不均衡问题时会导致性能下降。为此，这篇论文提出了一种新的FederAted neUral archiTecture search Approach（FedAutoMRI）。提议的方法使用可微分的建筑搜索自动找到最佳网络架构。此外，我们还引入了指数移动平均方法，以提高客户端模型的数据不均衡问题的Robustness。到目前为止，这是我们知道的第一篇使用联邦神经建筑搜索MR图像重建的论文。实验结果表明，我们的提议的FedAutoMRI可以实现扎实的性能，同时使用轻量级的模型，只有少量的模型参数，与传统的联邦学习方法相比，具有显著的优势。
UWAT-GAN: Fundus Fluorescein Angiography Synthesis via Ultra-wide-angle Transformation Multi-scale GAN
results: 实验结果表明，UWAT-GAN比现有方法有更高的图像质量和更好的抽象能力。code可以在GitHub上找到：https://github.com/Tinysqua/UWAT-GAN。Abstract
Fundus photography is an essential examination for clinical and differential diagnosis of fundus diseases. Recently, Ultra-Wide-angle Fundus (UWF) techniques, UWF Fluorescein Angiography (UWF-FA) and UWF Scanning Laser Ophthalmoscopy (UWF-SLO) have been gradually put into use. However, Fluorescein Angiography (FA) and UWF-FA require injecting sodium fluorescein which may have detrimental influences. To avoid negative impacts, cross-modality medical image generation algorithms have been proposed. Nevertheless, current methods in fundus imaging could not produce high-resolution images and are unable to capture tiny vascular lesion areas. This paper proposes a novel conditional generative adversarial network (UWAT-GAN) to synthesize UWF-FA from UWF-SLO. Using multi-scale generators and a fusion module patch to better extract global and local information, our model can generate high-resolution images. Moreover, an attention transmit module is proposed to help the decoder learn effectively. Besides, a supervised approach is used to train the network using multiple new weighted losses on different scales of data. Experiments on an in-house UWF image dataset demonstrate the superiority of the UWAT-GAN over the state-of-the-art methods. The source code is available at: https://github.com/Tinysqua/UWAT-GAN.
摘要
血液照片是诊断和区分疾病的基本检查方法。最近，ultra-wide-anglefundus（UWF）技术，UWFfluoresceinangiography（UWF-FA）和UWF扫描镜观察（UWF-SLO）逐渐普及。但是，fluoresceinangiography（FA）和UWF-FA需要注射Na fluorescein，可能有不良影响。为了避免这些影响，多modal医学影像生成算法已经被提出。然而，目前的基于基准图像的检查方法无法生成高分辨率图像，也无法捕捉微小血管损伤区域。本文提出了一种新的冲激生成随机网络（UWAT-GAN），用于从UWF-SLO中生成UWF-FA。通过多级生成器和融合模块贴图，我们的模型可以生成高分辨率图像。此外，我们还提出了一种注意力传输模块，帮助解码器更好地学习。此外，我们采用了多种新的质量权重损失来训练网络。实验结果表明，UWAT-GAN比现有的方法更高效。代码可以在GitHub上找到：https://github.com/Tinysqua/UWAT-GAN。
Bone mineral density estimation from a plain X-ray image by learning decomposition into projections of bone-segmented computed tomography
paper_authors: Yi Gu, Yoshito Otake, Keisuke Uemura, Mazen Soufi, Masaki Takao, Hugues Talbot, Seiji Okada, Nobuhiko Sugano, Yoshinobu Sato
For: The paper aims to estimate bone mineral density (BMD) from a plain X-ray image for opportunistic screening of osteoporosis.* Methods: The proposed method uses a novel approach that learns decomposition into projections of bone-segmented quantitative computed tomography (QCT) for BMD estimation under limited datasets.* Results: The proposed method achieved high accuracy in BMD estimation, with Pearson correlation coefficients of 0.880 and 0.920 observed for DXA-measured BMD and QCT-measured BMD estimation tasks, respectively. The root mean square of the coefficient of variation values were 3.27 to 3.79% for four measurements with different poses.Abstract
Osteoporosis is a prevalent bone disease that causes fractures in fragile bones, leading to a decline in daily living activities. Dual-energy X-ray absorptiometry (DXA) and quantitative computed tomography (QCT) are highly accurate for diagnosing osteoporosis; however, these modalities require special equipment and scan protocols. To frequently monitor bone health, low-cost, low-dose, and ubiquitously available diagnostic methods are highly anticipated. In this study, we aim to perform bone mineral density (BMD) estimation from a plain X-ray image for opportunistic screening, which is potentially useful for early diagnosis. Existing methods have used multi-stage approaches consisting of extraction of the region of interest and simple regression to estimate BMD, which require a large amount of training data. Therefore, we propose an efficient method that learns decomposition into projections of bone-segmented QCT for BMD estimation under limited datasets. The proposed method achieved high accuracy in BMD estimation, where Pearson correlation coefficients of 0.880 and 0.920 were observed for DXA-measured BMD and QCT-measured BMD estimation tasks, respectively, and the root mean square of the coefficient of variation values were 3.27 to 3.79% for four measurements with different poses. Furthermore, we conducted extensive validation experiments, including multi-pose, uncalibrated-CT, and compression experiments toward actual application in routine clinical practice.
摘要
骨质疾病（osteoporosis）是一种非常普遍的骨疾病，可以导致脆弱骨骼的折损，从而导致日常生活活动下降。双能X射线吸收测定（DXA）和量子计算Tomography（QCT）是骨质疾病的诊断非常准确的方法，但这些方法需要特殊的设备和扫描协议。为了经常监测骨健康，低成本、低剂量、 universally available的诊断方法很需求。在这个研究中，我们想使用普通X射线图像来估算骨质密度（BMD），以便于早期诊断。现有的方法通常使用多个阶段的方法，包括提取区域兴趣和简单的回归来估算BMD，这些方法需要大量的训练数据。因此，我们提出了一种高效的方法，该方法可以通过分解为骨 segmentation QCT 的投影来估算BMD，并在有限的数据集下进行学习。我们的方法实现了高精度的BMD 估算，其中 DXA 测量的BMD 和 QCT 测量的BMD 估算任务中的归一化相关系数为0.880和0.920，分解系数的平均方差为3.27%-3.79%。此外，我们进行了广泛的验证实验，包括多个姿势、不同扫描方式、压缩实验等，以便在实际临床医学中应用。
MatSpectNet: Material Segmentation Network with Domain-Aware and Physically-Constrained Hyperspectral Reconstruction
results: 对于LMD数据集和OpenSurfaces数据集，MatSpectNet实验表明，与最近一篇论文相比，MatSpectNet可以提高平均像素准确率1.60%，提高物种均准确率3.42%。Abstract
Achieving accurate material segmentation for 3-channel RGB images is challenging due to the considerable variation in a material's appearance. Hyperspectral images, which are sets of spectral measurements sampled at multiple wavelengths, theoretically offer distinct information for material identification, as variations in intensity of electromagnetic radiation reflected by a surface depend on the material composition of a scene. However, existing hyperspectral datasets are impoverished regarding the number of images and material categories for the dense material segmentation task, and collecting and annotating hyperspectral images with a spectral camera is prohibitively expensive. To address this, we propose a new model, the MatSpectNet to segment materials with recovered hyperspectral images from RGB images. The network leverages the principles of colour perception in modern cameras to constrain the reconstructed hyperspectral images and employs the domain adaptation method to generalise the hyperspectral reconstruction capability from a spectral recovery dataset to material segmentation datasets. The reconstructed hyperspectral images are further filtered using learned response curves and enhanced with human perception. The performance of MatSpectNet is evaluated on the LMD dataset as well as the OpenSurfaces dataset. Our experiments demonstrate that MatSpectNet attains a 1.60% increase in average pixel accuracy and a 3.42% improvement in mean class accuracy compared with the most recent publication. The project code is attached to the supplementary material and will be published on GitHub.
摘要
achieving accurate material segmentation for 3-channel RGB images is challenging due to the considerable variation in a material's appearance. Hyperspectral images, which are sets of spectral measurements sampled at multiple wavelengths, theoretically offer distinct information for material identification, as variations in intensity of electromagnetic radiation reflected by a surface depend on the material composition of a scene. However, existing hyperspectral datasets are impoverished regarding the number of images and material categories for the dense material segmentation task, and collecting and annotating hyperspectral images with a spectral camera is prohibitively expensive. To address this, we propose a new model, the MatSpectNet to segment materials with recovered hyperspectral images from RGB images. The network leverages the principles of color perception in modern cameras to constrain the reconstructed hyperspectral images and employs the domain adaptation method to generalize the hyperspectral reconstruction capability from a spectral recovery dataset to material segmentation datasets. The reconstructed hyperspectral images are further filtered using learned response curves and enhanced with human perception. The performance of MatSpectNet is evaluated on the LMD dataset as well as the OpenSurfaces dataset. Our experiments demonstrate that MatSpectNet attains a 1.60% increase in average pixel accuracy and a 3.42% improvement in mean class accuracy compared with the most recent publication. The project code is attached to the supplementary material and will be published on GitHub.Here's the translation in Traditional Chinese:achieving accurate material segmentation for 3-channel RGB images is challenging due to the considerable variation in a material's appearance. Hyperspectral images, which are sets of spectral measurements sampled at multiple wavelengths, theoretically offer distinct information for material identification, as variations in intensity of electromagnetic radiation reflected by a surface depend on the material composition of a scene. However, existing hyperspectral datasets are impoverished regarding the number of images and material categories for the dense material segmentation task, and collecting and annotating hyperspectral images with a spectral camera is prohibitively expensive. To address this, we propose a new model, the MatSpectNet to segment materials with recovered hyperspectral images from RGB images. The network leverages the principles of color perception in modern cameras to constrain the reconstructed hyperspectral images and employs the domain adaptation method to generalize the hyperspectral reconstruction capability from a spectral recovery dataset to material segmentation datasets. The reconstructed hyperspectral images are further filtered using learned response curves and enhanced with human perception. The performance of MatSpectNet is evaluated on the LMD dataset as well as the OpenSurfaces dataset. Our experiments demonstrate that MatSpectNet attains a 1.60% increase in average pixel accuracy and a 3.42% improvement in mean class accuracy compared with the most recent publication. The project code is attached to the supplementary material and will be published on GitHub.
BLISS: Interplanetary Exploration with Swarms of Low-Cost Spacecraft
results: 论文详细介绍了一种用于控制太阳帆的轨迹和低级 actuation控制，以及建议的载荷和计算机设计。 论文还 briefly Considered两个其他应用：从数十个金星家族彗星返回样本，以及遥感和拍摄遥行彗星。Abstract
Leveraging advancements in micro-scale technology, we propose a fleet of autonomous, low-cost, small solar sails for interplanetary exploration. The Berkeley Low-cost Interplanetary Solar Sail (BLISS) project aims to utilize small-scale technologies to create a fleet of tiny interplanetary femto-spacecraft for rapid, low-cost exploration of the inner solar system. This paper describes the hardware required to build a nearly 10 g spacecraft using a 1 m$^2$ solar sail steered by micro-electromechanical systems (MEMS) inchworm actuators. The trajectory control to a NEO, here 101955 Bennu, is detailed along with the low-level actuation control of the solar sail and the specifications of proposed onboard communication and computation. Two other applications are also shortly considered: sample return from dozens of Jupiter-family comets and interstellar comet rendezvous and imaging. The paper concludes by discussing the fundamental scaling limits and future directions for steerable autonomous miniature solar sails with onboard custom computers and sensors.
摘要
使用微型技术进行推动，我们提议一支自主、低成本、小型太阳帆船队进行 planetary exploration。 Berkeley Low-cost Interplanetary Solar Sail（BLISS）项目旨在利用小规模技术创造一支微型惯性空间飞船，用于快速、低成本地 explore 内太阳系。这篇文章描述了用于建立约10g空间飞船的硬件，包括1米²的太阳帆，由微型电子机械系统（MEMS）滚动器控制。文章还详细介绍了天体控制 trajectory 到 NEO 101955 Bennu，以及太阳帆的低级控制和船载通信和计算机的规格。此外，文章还 briefly 讨论了从数十个金星家族彗星返回样本和遥感柯梅丝 rendezvous 和拍摄。文章 conclude 了可控推进器的基本扩展限制和未来方向，包括自适应驱动器和船载特定计算机和感测器。
For: The paper aims to identify stable Electrophysiological Source Imaging (ESI) biomarkers associated with Gait Speed (GS) as a measure of functional decline in aging individuals.* Methods: The authors use a combination of flexible sparse/smooth/non-negative models (NN-SLASSO) and the Stable Sparse Classifier method to estimate ESI and select relevant features, including activation ESI (aESI) and connectivity ESI (cESI) features.* Results: The authors found that novel sparse aESI models outperformed traditional methods, and that combining aESI and cESI features improved the predictability of GS changes. The selected biomarkers were localized to orbitofrontal and temporal cortical regions.Here’s the Chinese translation of the three points:* For: 这篇论文目标是找到年轻人函散速度 (GS) 的稳定电physiological Source Imaging (ESI) 标记器。* Methods: 作者使用了一种组合 flexible sparse/smooth/non-negative models (NN-SLASSO) 和稳定粗粒分类方法来估算 ESI 和选择相关特征，包括 activation ESI (aESI) 和 connectivity ESI (cESI) 特征。* Results: 作者发现了一种新的简单 aESI 模型，并且将 aESI 和 cESI 特征组合可以提高 GS 变化的预测性。选择的标记器被localized到 orbitofrontal 和 temporal cortical region。Abstract
Objective: We seek stable Electrophysiological Source Imaging (ESI) biomarkers associated with Gait Speed (GS) as a measure of functional decline. Towards this end we determine the predictive value of ESI activation and connectivity patterns of resting-state EEG Theta rhythm on physical performance decline measured by a slowing GS in aging individuals. Methods: As potential biomarkers related to GS changes, we estimate ESI using flexible sparse/smooth/non-negative models (NN-SLASSO), from which activation ESI (aESI) and connectivity ESI (cESI) features are selected using the Stable Sparse Classifier method. Results and Conclusions: Novel sparse aESI models outperformed traditional methods such as the LORETA family. The models combining aESI and cESI features improved the predictability of GS changes. Selected biomarkers from activation/connectivity patterns were localized to orbitofrontal and temporal cortical regions. Significance: The proposed methodology contributes to understanding the activation and connectivity of ESI complex patterns related to GS, providing potential biomarker features for GS slowing. Given the known relationship between GS decline and cognitive impairment, this preliminary work suggests it might be applied to other, more complex measures of healthy and pathological aging. Importantly, it might allow an ESI-based evaluation of rehabilitation programs.
摘要
Methods: 作为可能的生物 marker相关于 GS 变化的方法，我们使用了 flexible sparse/smooth/non-negative 模型（NN-SLASSO），从中可以选择 activation ESI（aESI）和 connectivity ESI（cESI）特征，使用稳定的组合方法。Results and Conclusions: 我们发现了 Novel sparse aESI 模型，比 traditional LORETA 家族的方法更好。将 activation ESI 和 connectivity ESI 特征结合起来，可以提高 GS 变化的预测性。选择的生物 marker从 activation/connectivity 模式中的localized 到 orbitofrontal 和 temporal cortical region。Significance: 我们的方法可以帮助理解ESI复杂的模式之间的活动和连接相互关联，提供了可能的生物 marker特征，用于评估GS slowing。由于GS decline 和认知障碍之间的已知关系，这些早期的工作可能适用于其他更复杂的健康和疾病年龄。更重要的是，这种方法可能可以用于评估复健计划。
Treatment And Follow-Up Guidelines For Multiple Brain Metastases: A Systematic Review
results: 研究发现这种方法可以帮助医疗专业人员早期决定最佳治疗方案，并且可以提高病人的生活质量和神经保存。Abstract
Brain metastases are a complication of primary cancer, representing the most common type of brain tumor in adults. The management of multiple brain metastases represents a clinical challenge worldwide in finding the optimal treatment for patients considering various individual aspects. Managing multiple metastases with stereotactic radiosurgery (SRS) is being increasingly used because of quality of life and neurocognitive preservation, which do not present such good outcomes when dealt with whole brain radiation therapy (WBRT). After treatment, analyzing the progression of the disease still represents a clinical issue, since it is difficult to determine a standard schedule for image acquisition. A solution could be the applying artificial intelligence, namely predictive models to forecast the incidence of new metastases in post-treatment images. Although there aren't many works on this subject, this could potentially bennefit medical professionals in early decision of the best treatment approaches.
摘要
主要癌症的脑部 метаста团是一种常见的脑肿瘤，特别是在成人中。管理多个脑部 метаста团的治疗呈现出了全球的临床挑战，在考虑各个个体特点时寻找优化的治疗方案。使用STereotactic radiosurgery（SRS）来管理多个脑部 метаста团，因为它可以保持生活质量和神经功能，而不是整个脑部放射疗法（WBRT）所能提供的较差的结果。然而，在 после治疗时，评估疾病的进程仍然是临床问题，因为困难确定标准的图像获取时间间隔。一种可能的解决方案是通过应用人工智能，即预测模型，预测在后治疗图像中新形成的脑部 метаста团的发生率。虽然没有很多相关研究，但这可能可以帮助医疗专业人员在早期决定最佳治疗方案。
paper_authors: Xueshen Li, Zhenxing Dong, Hongshan Liu, Jennifer J. Kang-Mieler, Yuye Ling, Yu Gan
for: 提高镜像医学图像诊断和治疗的能力，特别是心血管和眼科领域。
methods: integrate three critical frequency-based modules (i.e., frequency transformation, frequency skip connection, and frequency alignment) and frequency-based loss function into a conditional generative adversarial network (cGAN)。
results: 在现有的 coronary OCT 数据集上进行了大规模的量化研究，证明了我们提出的框架在现有的深度学习框架之上具有超过性。此外，我们还在鱼眼层图像和兔脑层图像上应用了我们的框架，证明了它的 universality。Abstract
Optical coherence tomography (OCT) has stimulated a wide range of medical image-based diagnosis and treatment in fields such as cardiology and ophthalmology. Such applications can be further facilitated by deep learning-based super-resolution technology, which improves the capability of resolving morphological structures. However, existing deep learning-based method only focuses on spatial distribution and disregard frequency fidelity in image reconstruction, leading to a frequency bias. To overcome this limitation, we propose a frequency-aware super-resolution framework that integrates three critical frequency-based modules (i.e., frequency transformation, frequency skip connection, and frequency alignment) and frequency-based loss function into a conditional generative adversarial network (cGAN). We conducted a large-scale quantitative study from an existing coronary OCT dataset to demonstrate the superiority of our proposed framework over existing deep learning frameworks. In addition, we confirmed the generalizability of our framework by applying it to fish corneal images and rat retinal images, demonstrating its capability to super-resolve morphological details in eye imaging.
摘要
results: 实验结果表明，对于图像分割和净化任务，我们的Spiking-UNet可以与非神经元网络相比，并且超过现有的SNN方法。与未调整的Spiking-UNet相比，我们的Spiking-UNet可以降低推理时间约90%。Abstract
U-Net, known for its simple yet efficient architecture, is widely utilized for image processing tasks and is particularly suitable for deployment on neuromorphic chips. This paper introduces the novel concept of Spiking-UNet for image processing, which combines the power of Spiking Neural Networks (SNNs) with the U-Net architecture. To achieve an efficient Spiking-UNet, we face two primary challenges: ensuring high-fidelity information propagation through the network via spikes and formulating an effective training strategy. To address the issue of information loss, we introduce multi-threshold spiking neurons, which improve the efficiency of information transmission within the Spiking-UNet. For the training strategy, we adopt a conversion and fine-tuning pipeline that leverage pre-trained U-Net models. During the conversion process, significant variability in data distribution across different parts is observed when utilizing skip connections. Therefore, we propose a connection-wise normalization method to prevent inaccurate firing rates. Furthermore, we adopt a flow-based training method to fine-tune the converted models, reducing time steps while preserving performance. Experimental results show that, on image segmentation and denoising, our Spiking-UNet achieves comparable performance to its non-spiking counterpart, surpassing existing SNN methods. Compared with the converted Spiking-UNet without fine-tuning, our Spiking-UNet reduces inference time by approximately 90\%. This research broadens the application scope of SNNs in image processing and is expected to inspire further exploration in the field of neuromorphic engineering. The code for our Spiking-UNet implementation is available at https://github.com/SNNresearch/Spiking-UNet.
摘要
优等网络（U-Net），因其简单而高效的架构，广泛应用于图像处理任务，特别适合部署在神经omorphic芯片上。这篇论文提出了一种新的启发式神经网络（SNN）图像处理方法，其结合了U-Net架构和启发式神经网络（SNN）的优势。为确保高精度信息传递，我们面临两个主要挑战：保证信息传递的高精度和制定有效的训练策略。为Address the issue of information loss, we introduce multi-threshold spiking neurons, which improve the efficiency of information transmission within the Spiking-UNet. For the training strategy, we adopt a conversion and fine-tuning pipeline that leverage pre-trained U-Net models. During the conversion process, we observe significant variability in data distribution across different parts when utilizing skip connections. Therefore, we propose a connection-wise normalization method to prevent inaccurate firing rates. Furthermore, we adopt a flow-based training method to fine-tune the converted models, reducing time steps while preserving performance. Experimental results show that, on image segmentation and denoising, our Spiking-UNet achieves comparable performance to its non-spiking counterpart, surpassing existing SNN methods. Compared with the converted Spiking-UNet without fine-tuning, our Spiking-UNet reduces inference time by approximately 90\%. This research broadens the application scope of SNNs in image processing and is expected to inspire further exploration in the field of neuromorphic engineering. The code for our Spiking-UNet implementation is available at .