Recent Posts

Pages: [1] 2 3 ... 10
1
Automated system identifies dense tissue, a risk factor for breast cancer, in mammograms
16 October 2018, 4:09 pm

Researchers from MIT and Massachusetts General Hospital have developed an automated model that assesses dense breast tissue in mammograms — which is an independent risk factor for breast cancer — as reliably as expert radiologists.

This marks the first time a deep-learning model of its kind has successfully been used in a clinic on real patients, according to the researchers. With broad implementation, the researchers hope the model can help bring greater reliability to breast density assessments across the nation.

It’s estimated that more than 40 percent of U.S. women have dense breast tissue, which alone increases the risk of breast cancer. Moreover, dense tissue can mask cancers on the mammogram, making screening more difficult. As a result, 30 U.S. states mandate that women must be notified if their mammograms indicate they have dense breasts.

But breast density assessments rely on subjective human assessment. Due to many factors, results vary — sometimes dramatically — across radiologists. The MIT and MGH researchers trained a deep-learning model on tens of thousands of high-quality digital mammograms to learn to distinguish different types of breast tissue, from fatty to extremely dense, based on expert assessments. Given a new mammogram, the model can then identify a density measurement that closely aligns with expert opinion.

“Breast density is an independent risk factor that drives how we communicate with women about their cancer risk. Our motivation was to create an accurate and consistent tool, that can be shared and used across health care systems,” says Adam Yala, a PhD student in MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) and second author on a paper describing the model that was published today in Radiology.

The other co-authors are first author Constance Lehman, professor of radiology at Harvard Medical School and the director of breast imaging at the MGH; and senior author Regina Barzilay, the Delta Electronics Professor at CSAIL and the Department of Electrical Engineering and Computer Science at MIT and a member of the Koch Institute for Integrative Cancer Research at MIT.

Mapping density

The model is built on a convolutional neural network (CNN), which is also used for computer vision tasks. The researchers trained and tested their model on a dataset of more than 58,000 randomly selected mammograms from more than 39,000 women screened between 2009 and 2011. For training, they used around 41,000 mammograms and, for testing, about 8,600 mammograms.

Each mammogram in the dataset has a standard Breast Imaging Reporting and Data System (BI-RADS) breast density rating in four categories: fatty, scattered (scattered density), heterogeneous (mostly dense), and dense. In both training and testing mammograms, about 40 percent were assessed as heterogeneous and dense.

During the training process, the model is given random mammograms to analyze. It learns to map the mammogram with expert radiologist density ratings. Dense breasts, for instance, contain glandular and fibrous connective tissue, which appear as compact networks of thick white lines and solid white patches. Fatty tissue networks appear much thinner, with gray area throughout. In testing, the model observes new mammograms and predicts the most likely density category.

Matching assessments

The model was implemented at the breast imaging division at MGH. In a traditional workflow, when a mammogram is taken, it’s sent to a workstation for a radiologist to assess. The researchers’ model is installed in a separate machine that intercepts the scans before it reaches the radiologist, and assigns each mammogram a density rating. When radiologists pull up a scan at their workstations, they’ll see the model’s assigned rating, which they then accept or reject.

“It takes less than a second per image … [and it can be] easily and cheaply scaled throughout hospitals.” Yala says.

On over 10,000 mammograms at MGH from January to May of this year, the model achieved 94 percent agreement among the hospital’s radiologists in a binary test — determining whether breasts were either heterogeneous and dense, or fatty and scattered. Across all four BI-RADS categories, it matched radiologists’ assessments at 90 percent. “MGH is a top breast imaging center with high inter-radiologist agreement, and this high quality dataset enabled us to develop a strong model,” Yala says.

In general testing using the original dataset, the model matched the original human expert interpretations at 77 percent across four BI-RADS categories and, in binary tests, matched the interpretations at 87 percent.

In comparison with traditional prediction models, the researchers used a metric called a kappa score, where 1 indicates that predictions agree every time, and anything lower indicates fewer instances of agreements. Kappa scores for commercially available automatic density-assessment models score a maximum of about 0.6. In the clinical application, the researchers’ model scored 0.85 kappa score and, in testing, scored a 0.67. This means the model makes better predictions than traditional models.

In an additional experiment, the researchers tested the model’s agreement with consensus from five MGH radiologists from 500 random test mammograms. The radiologists assigned breast density to the mammograms without knowledge of the original assessment, or their peers’ or the model’s assessments. In this experiment, the model achieved a kappa score of 0.78 with the radiologist consensus.

Next, the researchers aim to scale the model into other hospitals. “Building on this translational experience, we will explore how to transition machine-learning algorithms developed at MIT into clinic benefiting millions of patients,” Barzilay says. “This is a charter of the new center at MIT — the Abdul Latif Jameel Clinic for Machine Learning in Health at MIT — that was recently launched. And we are excited about new opportunities opened up by this center.”

Source: MIT News - CSAIL - Robotics - Computer Science and Artificial Intelligence Laboratory (CSAIL) - Robots - Artificial intelligence

Reprinted with permission of MIT News : MIT News homepage



Use the link at the top of the story to get to the original article.
2
XKCD Comic / XKCD Comic : Tectonics Game
« Last post by Tyler on Today at 12:00:10 pm »
Tectonics Game
19 October 2018, 5:00 am

They're limiting the playtesters to type A3 V stars, so the games will all end before the Sun consumes the Earth.

Source: xkcd.com

3
AI News / Re: In need of a psychological coach?
« Last post by ruebot on Today at 09:16:20 am »
Quote
(The inability to spell "practicing counselors" correctly in your manifesto does not inspire confidence.)
Actually their spelling is correct and yours is not. practice ought to be written with an S when it's used as a verb, and only Americans write counsellor with a single L. Perhaps what you need is a spelling coach.

Quote
counselor
noun

    a person who counsels; adviser.
    a faculty member who advises students on personal and academic problems, career choices, and the like.
    an assistant at a children's camp, often a high-school or college student, who supervises a group of children or directs a particular activity, as nature study or a sport.

https://www.dictionary.com/browse/counselor?s=t

Quote
practicing

verb (used with object), prac·ticed, prac·tic·ing.

    to perform or do habitually or usually: to practice a strict regimen.
    to follow or observe habitually or customarily: to practice one's religion.
    to exercise or pursue as a profession, art, or occupation: to practice law.
    to perform or do repeatedly in order to acquire skill or proficiency: to practice the violin.
    to train or drill (a person, animal, etc.) in something in order to give proficiency.

https://www.dictionary.com/browse/practicing?s=t
4
AI News / Re: In need of a psychological coach?
« Last post by Don Patrick on Today at 08:09:05 am »
Quote
(The inability to spell "practicing counselors" correctly in your manifesto does not inspire confidence.)
Actually their spelling is correct and yours is not. practice ought to be written with an S when it's used as a verb, and only Americans write counsellor with a single L. Perhaps what you need is a spelling coach.
5
General AI Discussion / Re: ETHICS
« Last post by Art on Today at 04:13:50 am »
Yes, by all means, Do as we (humans) say, not as we do!  O0
6
General Chat / Re: Cats are evolving...
« Last post by Hopefully Something on Today at 03:29:22 am »
Oh dear... This is worse than the singularity.
7
General AI Discussion / Re: Gaining more insight and access to ANNs
« Last post by AgentSmith on Today at 02:59:24 am »
ANNs are tangled. There's no point in looking there, really.

Looking at the weights of an ANN is pointless, very right. And nobody does so, because in general no human is able to predict the outcome of an ANN (and yes, this includes the "experts"). What you look at is the result your ANN produces after you let it run. Quite simple and intuitive.
8
General Chat / Cats are evolving...
« Last post by korrelan on Today at 02:06:01 am »
Ask any neuroscientist and they will tell you only certain animals are considered self aware, dolphins, primates, elephants, etc... Cats are not considered part of this exclusive group.



Mimo obviously didn't get the memo.

https://en.m.wikipedia.org/wiki/Mirror_test

 :)
9
General AI Discussion / Re: Gaining more insight and access to ANNs
« Last post by LOCKSUIT on October 19, 2018, 11:25:25 pm »
OH YEAH I MADE THIS THREAD

ANNs are tangled. There's no point in looking there, really. But do look there. It's where you begin your smarty pants. Now you on top.
10
General Chat / Re: We can't prove or disprove some sentences
« Last post by LOCKSUIT on October 19, 2018, 02:26:33 pm »
"There are sentences in this theory that can't be proved nor disproved." No, I as a human myself can prove/disprove this and understand it. This is a sentence that says the sentences in it (the theory) can't be proven nor disproven, so let's take this sentence, prove/disprove it, and, we disprove it, because there is not sentences in this sentence that can't be proven nor disproven, otherwise if there was then we would prove this sentence. Also, there is sentences in this theory that may be disproven by certain readers, or proven, and I realized this after thinking "There are sentences in this theory that you can't be mad at.". But there are sentences in this theory that can be acknowledged, feared, hated, loved, bought, and so on! Ok so it basically says you can't disproven nor prove this sentence, and I think: what is there to prove/disprove?, this i guess > "you can't prove nor disprove this sentence", I'm sure we can reach an agreement. Ok I did it, the sentence is proven true if I CAN'T prove it nnoorr disprove it, else false disprove. Clearly there is nothing that can be proven, nor that can be disproven. So there are sentences in this theory, that can't be proven? True. Nor disproven? True. So True. But now wait it just said itself cannot be proven and I just agreed with the line that itself cannot be proven yet I just proved itself true that it is true.....yes i did, I proven this line. Then again I might reframe to FALSE because it said itself cannot be proven.....but then it said itself couldn't be disproven false either and if i do say so now then it is false......so false?.....................it true if can't be true or false.........if i can say true or false which i did then it is false OMG DID IT

"There are sentences in this theory that can't be proved nor disproved."
FALSE AFAIK

The main concept of my discovery here is that there is no object in it that can be proven or disproven, so it, is, true..........yet it says itself cant be true nor false so because i have indeed said one of them (true), then the sentence is, false.

To go further, i guess it, is an object, and it says itself as an object can't be proven nor disproven (unless we focus on the "sentences IN this theory"), well, i can mark it true or false, hence it'd be false.

SO: If i can mark it true positive or false negative then it is false. And I can. So false. Also, in a sense yes true there is sentences in itself/itself can't be really proven or disproven so yes True but then being true makes it false anyhow.

However if you look at your grading/marking of it as a separate thing, then True, it can't be proven nor disoroven, so it's True and positive. Then you could make a clone of this and make this one False for being that you say it true yet it say it cannot be true hence it false no good, that way you get the both of best worlds.

Now you end up with in ur brain:
"False i can mark it +/-."
"True it is right."
"False cus it said it couldn't be right nor wrong."
"And same for if you mean it or the sentences in it."
"Or whatever the hell your brain thinks."

Who wants cake!?
Pages: [1] 2 3 ... 10

Users Online

66 Guests, 0 Users

Most Online Today: 116. Most Online Ever: 208 (August 27, 2008, 09:36:30 am)

Articles