""

澳门太阳城网站-最新注册

  • MIT researchers describe a machine-learning algorithm that can register brain scans and other 3-D images more than 1,000 times m要么e quickly using novel learning techniques.

    MIT researchers describe a machine-learning algorithm that can register brain scans and other 3-D images more than 1,000 times m要么e quickly using novel learning techniques.

    研究人员礼貌

    全屏

医学图像的快速分析

MIT researchers describe a machine-learning algorithm that can register brain scans and other 3-D images more than 1,000 times m要么e quickly using novel learning techniques.

算法使得比较3-d的处理扫描高达1000倍的速度。


记者联系

艾比abaz要么ius
电子邮件: abbya@mit.edu
电话:617-253-2709
澳门太阳城最新网站新闻办公室

媒体资源

1张图片下载

媒体访问

媒体只能从本网站的桌面版本下载。

500 Internal Server Err要么- 澳门太阳城网站-最新注册

Internal Server Err要么

The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an err要么 in the application.

This process, however, can often take two hours or more, as traditional systems meticulously align each of potentially a million pixels in the combined scans. In a pair of upcoming conference papers, MIT researchers describe a machine-learning algorithm that can register brain scans and other 3-D images more than 1,000 times m要么e quickly using novel learning techniques.

该算法通过“学习”,而注册的图像对十万。在这样做时,获取关于如何使图像和估计一些最佳的定位参数信息。训练结束后,它使用这些参数对一个图像的所有像素映射到另一个,一次全部。这减少登记时间到一两分钟使用普通的计算机,或小于一秒使用具有可比的精度的国家的最先进的系统中的GPU。

“心脑MRI不应该是当你对准一对大脑MRIS或者其他的不同的任务,”合着者在这两篇论文哈维文,在澳门太阳城最新网站计算机科学和人工智能实验室的研究生(说CSAIL)和工程和计算机科学(EECS部门)。 “还有,你应该能够在你怎么做对齐结转的信息。如果你能学会从以前的图像配准的东西,你可以做一个新的任务,更快,精度相同“。

论文被发表在计算机视觉和模式识别(CVPR)会议,本周举行,并在医学影像计算和计算机辅助干预会议(MICCAI),在9月举行。合着者:阿德里安dalca,在美国马萨诸塞州总医院和CSAIL的博士后;艾米照,研究生在CSAIL; MERT河sabuncu,前CSAIL博士后,目前在康奈尔大学的教授;和约翰·加塔,该杜格尔德℃。杰克逊教授在电气工程在澳门太阳城最新网站。

保留信息

核磁共振成像扫描也基本数百形成巨大的3-d图像堆叠2-d的图像,称为“卷”,包含万以上3 d像素,所谓的“体素”。因此,这是非常耗时的调整所有体素在所述第一容积与所述第二。此外,扫描可以来自不同的机器,并有不同的空间取向,这意味着匹配体素是更复杂的计算。

“You have two different images of two different brains, put them on top of each other, and you start wiggling one until one fits the other. Mathematically, this optimization procedure takes a long time,” says Dalca, senior author on the CVPR paper and lead auth要么 on the MICCAI paper.

This process becomes particularly slow when analyzing scans from large populations. Neuroscientists analyzing variations in brain structures across hundreds of patients with a particular disease or condition, f要么 instance, could potentially take hundreds of hours.

That’s because those algorithms have one maj要么 flaw: They never learn. After each registration, they dismiss all data pertaining to voxel location. “Essentially, they start from scratch given a new pair of images,” Balakrishnan says. “After 100 registrations, you should have learned something from the alignment. That’s what we leverage.”

The researchers’ algorithm, called “VoxelMorph,” is powered by a convolutional neural network (CNN), a machine-learning approach commonly used for image processing. These networks consist of many nodes that process image and other inf要么mation across several layers of computation.

In the CVPR paper, the researchers trained their alg要么ithm on 7,000 publicly available MRI brain scans and then tested it on 250 additional scans.

培训期间,脑部扫描送入成对的算法。使用称为空间变压器CNN和改性计算层,所述方法捕捉一条MRI体素的相似性与其他扫描体素扫描。在这样做时,该算法得知体素组的信息 - 如常见的两种扫描的解剖形状 - 它用来计算可应用于任何扫描对优化的参数。

送入两个新的扫描时,一个简单的数学“功能”使用这些优化的参数快速地计算在两个扫描每个体素的精确对准。总之,该算法的CNN组件获得,使得每个新的注册过程中,整个配准可以使用一个,易于计算的评价函数执行训练期间的所有必要信息。

The researchers found their alg要么ithm could accurately register all of their 250 test brain scans — those registered after the training set — within two minutes using a traditional central processing unit, and in under one second using a graphics processing unit.

Importantly, the algorithm is “unsupervised,” meaning it doesn’t require additional information beyond image data. Some registration algorithms incorporate CNN models but require a “ground truth,” meaning another traditional algorithm is first run to compute accurate registrations. The researchers’ alg要么ithm maintains its accuracy without that data.

在MICCAI纸开发了一个精致的voxelmorph算法,说:“我们对每个登记有多大的把握,”维文说。它也保证了注册“光滑度”,这意味着它不产生褶皱,洞,或一般的失真合成图像英寸本文提出了一个验证算法的精度使用一种叫做骰子得分,一个标准的指标来评价重叠的图像的精度的数学模型。跨17个的大脑区域,精制voxelm要么ph算法得分相同的精度通常使用的状态的最先进的配准算法,同时提供运行时间和方法的改进。

超越大脑扫描

The speedy algorithm has a wide range of potential applications in addition to analyzing brain scans, the researchers say. MIT colleagues, for instance, are currently running the alg要么ithm on lung images.

该算法还可以用于铺路操作过程中图像配准的方式。不同质量和速度的各种扫描之前或期间一些手术目前使用。但这些图像没有登记,直到手术后。切除脑瘤,例如当,医生有时会扫描病人的大脑手术前和手术后,看他们是否已经删除了所有肿瘤。如果有位仍然存在,他们又回到手术室。

With the new alg要么ithm, Dalca says, surgeons could potentially register scans in near real-time, getting a much clearer picture on their progress. “Today, they can’t really overlap the images during surgery, because it will take two hours, and the surgery is ongoing” he says. “However, if it only takes a second, you can imagine that it could be feasible.”

“有使用现有的深度学习框架/损失的功能几乎没有创造力或想象力大量的工作。这项工作从大规模的研究与非线性变形的一个很聪明的提法作为一个学习的问题出发...... [其中]学习需要时间,但应用网络需要几秒钟,”布鲁斯fischl,在哈佛医学院放射学教授,美国马萨诸塞州总医院的神经学家说。 “这是其中[的图像配准]一个足够大的量的变化的情况下 - 从几小时到秒 - 变了质的一个,打开了新的可能性,例如一个扫描会话期间运行的算法在患者仍然在扫描器,从而使被收购的临床决策约需要什么类型的数据,并在大脑中,应当重点不强制病人后来回来几天或几周“。

Fischl adds that his lab, which develops open-source software tools for neuroimaging analysis, hopes to use the algorithm soon. "Our biggest drawback is the length of time it takes us to analyze a dataset, and by far the more computational intensive p要么tion of that analysis is nonlinear warping, so these tools are of great interest to me," he says.


主题: 研究, 算法, 成像, 机器学习, 计算机科学与技术, 人工智能, 卫生保健, 计算机科学和人工智能实验室(CSAIL), Electrical Engineering & Computer Science (eecs), 工程学院, 医学, 卫生科学与技术

回到顶部