PA Images / Kristan Lieb/Chicago Tribune/TNS/Sipa USA

One in eight women will be diagnosed with breast cancer throughout their lives. In an effort to help with quicker detection, researchers have trained a deep-learning algorithm to spot breast cancer in screening scans as accurately or better than a radiologist.

While still at an early stage, the research could eventually help reduce incorrect results in the US and help alleviate the shortage of radiologists in the UK. As early detection is key to treatment, women over the age of 50 are tested in the US and UK even if they don't show signs of the disease. False negatives, when cancer is present but not spotted, can prove deadly, while false positives can be distressing.

Google-owned DeepMind has already worked with NHS organisations to develop AI to read eye scans and spot neck cancer. Over the past two years, researchers from Cancer Research UK Imperial College, Northwestern University, Royal Surrey County Hospital, and Google Health have used a deep-learning system developed by DeepMind on two different datasets of breast scans, one from the US and one from the UK, suggesting the AI could help read mammograms accurately.

"This is another step along the way of trying to answer some of the questions that will be critical for us to actually deploying this in the real world," says Dominic King, director and UK lead of Google Health. "This is another step closer to trying to deploy this type of technology safely and effectively."

The system was first trained on de-identified mammograms from 76,000 British women, using Cancer Research UK's OPTIMAM dataset, as well as 15,000 scans from the US. Once trained, the algorithm was tested on 25,000 scans in the UK, and a further 3,000 in the US. Four images from each mammogram were pulled into a neural network, spitting out scores for three different models between zero and one, with the latter a high risk of cancer.

The AI's conclusions were then compared against real-life results in the future and what radiologists said at the time, says Christopher Kelly, clinician scientist at Google Health, and co-author of the research, which has been published in the journal Nature. "Our ground truth was based on biopsy results and follow-ups, so if they had a normal screen two or three years later," he explains.

The research says the AI model could predict breast cancer with the same level of accuracy as a single expert radiologist. Compared to human experts, the system saw a reduction in false positives by 5.7 per cent in the US and 1.2 per cent in the UK, and in false negatives of 9.4 per cent in the US and 2.7 per cent in the UK.

However, those results don't necessarily reflect how such scans are read in real life. In the US, breast scans are normally checked by a single radiologist, while in the NHS mammograms are checked by a minimum of two radiologists. If those two "readers" disagree on the result, the scan is checked by a third and potentially even a fourth.

The study claims the DeepMind algorithm performs better than a single radiologist, and is "non inferior" versus two."The model performs better than an individual radiologist in both the UK and the US," Kelly says. "In the UK we have this double reading system, where two radiologists or maybe three or four look at each scan we're statistically the same as that, but not better than that."

However, the Royal College of Radiologists says its workforce modelling research suggests the UK is short of at least 1,104 radiologists; there are currently 542 expert breast radiologists in the UK, but eight per cent of hospital posts for such roles are unfilled.

If the role of the second reader could be partially replaced by AI, that could alleviate some of the staff shortages, notes King indeed, he says radiologists asked Google Health to look into AI for screening scans for just this reason. "We had a group of senior breast radiologists in the UK contact us three or four years ago to say that this was an area they felt was amenable to artificial intelligence but also it was critical to start thinking about how technology could start supporting the sustainability of the service, because currently there can be very lengthy delays," he says.

To test that idea, the researchers ran a side project, simulating how the algorithm could work with a human radiologist. The AI and the human radiologists agreed 88 per cent of the time, meaning only 12 per cent of scans would then have to be read by another radiologist. However, the reader study was run with a more limited data set and only six radiologists, all of whom were US trained and only two of whom had fellowship-level training in breast imaging. "We'd like to test this with further work, but as a kind of simulation, it was quite exciting to see this as a suggestion towards a potential system on the future," says Kelly.

Regardless of the success of such research, radiologists can't be fully replaced by AI but they could be assisted, stresses Caroline Rubin, vice president for Clinical Radiology at the Royal College of Radiologists, who was not involved in the research. Like the rest of the health service, breast imaging and UK radiology more widely is under-staffed and desperate for help," she says. "AI programmes will not solve the human staffing crisis as radiologists and imaging teams do far more than just look at scans but they will undoubtedly help by acting as a second pair of eyes and a safety net."

Alongside the accuracy checks and the reader simulation, the researchers also examined whether the system could be generalised, meaning the system could be trained on a single data set and used everywhere. To test this, they ran the algorithm trained on UK data on the American scans. The results were not as good as the system trained on US data a 3.5 per cent reduction in false positives, versus 5.7 per cent using local data but they were still positive, suggesting some generalisation may be possible, but training on a localised data set remains preferable.

While the results are positive, the researchers stress the work remains in the early stages. The Google researchers say they would like to see more research done not on retrospective, historical data, but with current patients. "Prospective studies are the only way you find out how these things perform in the real world," Kelly says, in particular how clinicians would interact with the system. Plans for such a project is in the works. "That's a different programme of research that we're now excited to be exploring," King says.

The RCP's Rubin agrees, calling for rigorous testing and careful regulation. "The next step for promising [breast screening AI] products is for them to be used in clinical trials, evaluated in practice and used on patients screened in real-time, a process that will need to be overseen by the UK public health agencies that have overall responsibility for the breast screening programmes," she says.

Thanks in part to the Royal Free debacle, concerns around data privacy have stalked Google's latest foray into medical research. The researchers stressed that the mammograms were de-identified, adding that the algorithm looked only at the scans and no other patient information. The UK data was sourced from a set collected specifically for research by Cancer Research UK.

Another concern with AI development is the spectre of bias, and the paper says that checks of bias found none. That suggests the algorithm should work equally well on any scan though it's actually better at spotting invasive cancers, that's positive as humans find them more troublesome regardless of the specific details of the individual. To check that, Kelly says the team looked at metadata associated with each image to ensure the AI wasn't "underperforming" on minority subgroups, he says, adding that more in depth analysis can and should be done to ensure no bias.

Andrew Holding, senior research associate at Cancer Research UK's Cambridge Institute, who was not involved with the research, says the best way to avoid such bias is training with a diverse data set. "In this study data from the US and the UK was used in recognition of these challenges, nonetheless, were still a long way from representing the full diversity of people who present in the clinic," he says.

"A clinician would rapidly adapt to something as simple as skin pigmentation by drawing on their wider life experiences, but an AI having never seen it would diagnose in an unpredictable manner. Similar problems could occur because one hospital uses a slightly older piece of equipment to take the mammogram leading, and that might lead to different patient outcomes. These problems aren't unsolvable, but they do present a huge challenge."

One bias to be considered in future research is the manufacturer of the scanning machines. The study coincidentally used scans that were predominantly produced via Hologic machines, and future work should ensure the algorithm works as well with other scanners.

There is one concern raised by the paper, says Holding: the code used by the algorithm has not been released. "The code used for training the models has a large number of dependencies on internal tooling, infrastructure and hardware, and its release is therefore not feasible," the paper notes, saying it's described in enough detail in the supplementary materials to be replicated with non-proprietary libraries.

"The work presents a fantastic effort, but its a shame that the authors have decided only to include instructions on how the AI was built and not provide the source code," Holding says, pointing to a campaign for reproducible science. "Including source code is vital for increasing the impact of the research. It allows other scientists to build on the work rather than having to start again from scratch and provides for a better understanding of how the results were obtained. It also helps the researchers who do the work. By working reproducibly, you avoid costly mistakes and help your own research group build on your results."

Beyond research itself, Holding argues researchers owe it to patients whose data they use to release such information freely. If patients are generous enough to consent to their data to be used by companies like Google for research purposes, ideally the results and methods generated from the data should be available to them for free," he says. "The research simply isnt possible without that consent.

Google Health's Kelly admits that black-box algorithms are less useful in clinical settings, as it's helpful to physicians to see their workings. This particular system is made up of three different models, which are combined into a single score. One of those, the local model, is perhaps the most important to clinicians as it highlights areas of concern. "When you look at workflows, localisation is actually really important to a radiologist they look at the images and draw on the area that is suspicious," says Kelly.

"When that goes to a consensus process, they compare all these diagrams." That means that while the "global model", which includes the local and the other data, has the most accurate cancer predictions, it may not be as useful to human radiologists. "Although it might perform the best, it's a black box." That said, Holding adds, we don't always know how radiologists or physicians make their decisions, either. "While it's in my nature to be sceptical of 'black-box' software, there is the counterpoint that we don't really know how each individual clinician pulls together years of experience to make a decision about a patient."

Work remains on this particular project as well as in the wider field of AI to read medical images, but such progress not only could cut costs, but improve care, Holding says. "These studies provide the path to a second digital pair of eyes that are never tired and see every patient," he says. "By then using these AIs to catch potential mistakes, we can avoid concerns of putting all our faith in software, and still apply the technology to give better patient outcome. And that is really exciting.

SUVs are worse for the planet than anyone realised

Science says we should work shorter hours in winter

The illegal trade of Siberian mammoth tusks revealed

I ditched Google for DuckDuckGo. Here's why you should too

How to use psychology to get people to answer your emails

Get The Email from WIRED, your no-nonsense briefing on all the biggest stories in technology, business and science. In your inbox every weekday at 12pm sharp.

by entering your email address, you agree to our privacy policy

Thank You. You have successfully subscribed to our newsletter. You will hear from us shortly.

Sorry, you have entered an invalid email. Please refresh and try again.

More:

DeepMind's new AI can spot breast cancer just as well as your doctor - Wired.co.uk

In the Vale. Background image, Parish Church, St Nicholas, Vale of Glamorgan/ Picture by John Lord (CC BY-SA 2.0)

Jon Gower

The eponymous vale of this evocative historical novel is the Vale of Glamorgan though at times it might just as easily be the Vale of Tears as children die prematurely, hunger stalks the land and the shadowy clouds of the Napoleonic Wars lengthen.

The central characters are George, a cash-strapped curate and his wife Sarah, who conveniently marries him after becoming pregnant with the child of a local landowner, Richard Aubrey. Hiding his paternity Aubrey pulls some strings and George is gifted the ancient church at Llantrithyd to look after, and thus a scant but secure living, as this is still a period when the church was granted tithes from its parishioners. They tilled the land and the churchmen prayed for them in turn, and took their scant money.

With the sure authority of Anthony Trollope writing about the clergy of Barchester, Adams details the life of the parish and a clergymans responsibilities therein. He is especially good on the patterns, rigours and terminology of the agricultural life. So we find George weeding a barley crop, His companion in labour was one of a pair of oxen from the hovel, which needed the flick of a long, thin whip on its flank as a reminder to lift one broad hoof after another with the same measured pace the share would spill out fresh furrows through the day.

On a warm day in August, the communal act of haymaking sees landowners and tenants alike working in tandem, having fallen into line and into the rhythm of cutting, the scythe close to the ground, arms straight, the body twisting left to right, slicing the grass neatly, adding stroke by stroke to the swathe lengthening across the field.

All this detail, all these rhythms of the farming year are delivered with quiet authority. Here is a novelist who knows how to employ his research without show or flourish, allowing his evocations of London life, or autumnal shooting parties out to bag hares and partridges, or the rituals of rural funerals to settle fully in the readers mind.

Defences

Sam Adams is a long-established poet as well as a novelist so naturally, the book is shot through with passages of quiet lyricism:

Nearing Penlline, the lane climbed steeply and passed between tall trees and hedges so high and weighted by foliage they bowed inwards, hiding the sky. In the damp warmth of May, blossoms on the hedge banks, primrose and dandelion, bluebell and celandine, strained up towards chinks of light in the canopy.

One of the unexpected subjects of the book is that of inoculation, as Jenners ideas for ensuring children should not be affected by the disfigurements and threat of death by smallpox are trialled by George and Sarah on their child Bloom, with tragic consequences. Life and death, plantings and harvests, blossoming and decay: these are the patternings of the novel.

As the story proceeds so too do Britains preparations to deal with Napoleons commandeering of Europe, and a threat to British shores. Richard Aubrey becomes the commandant of the Glamorgan Militia, which finds itself on the south coast of England, digging underground defences to ward off French invasion across the channel.

For a brief while, George joins him as a curate, where the learns the men are a different breed. When he asks them where they come from, which parishes and who are their vicars one of them replies No vicars, said the Dowlais man, chapel us. And, as work began, the deep voice from the gloom rose again in a hymn tune, in a moment taken up it seemed by the whole body of the men, and the tune, the strong Welsh words, followed the chaplain

Meanwhile, back in Wales, the effect of the war is felt deeply as bread prices rise and food shortages spread and many already poor people find themselves on starvation rations. A rebellious spirit spreads among them and George, in his role as magistrate, has to stand up to them and in particular to one of the ringleaders, his surly, obnoxious gardener who turns out to have the blackest heart of all the books characters.

The Vale offers an absorbing insight into life in both grand houses and near-derelict hovels, setting the lives of working folk in a small corner of Wales in the context of sweeping events such as wars in Spain or Napoleons attempt to annexe Austria.

Sam Adams has won his spurs as a writer about history with previous works such as the novel Prichards Nose and Where the Stream Ran Red, a memoir of life in Gilfach Goch, the mining village where he grew up. Here, once again he proves himself to be an artist who is happy to work with a broad and busy canvas, filling it with finely detailed landscapes and cityscapes and characters who linger a long time in the mind.

Sam Adams himself has been writing since the 1960s and this latest novel shows absolutely no signs whatsoever of his tale-tellers gifts diminishing. In fact, quite the opposite.

In the Vale by Sam Adams costs 9.99 and can be bought here.

See original here:

Review: In the Vale is full of characters who linger long in the mind - Nation.Cymru

Bigger, better, more and now.

This is the society we live in today. Christmas Trees up before Halloween, shopping, parties, spending, eating, drinking, rushing, worrying all in the name of holidays.

The word holiday comes from the word haligdaeg (holy day) and refers to a day of rest & relaxation, without work. And we work so hard for that day without work.

And then, in the blink of an eye, its here. Christmas morning, the kids rush down the stairs, begging to open gifts. We desperately pour another cup of coffee, and act surprised by what Santa left under the tree. We watch as they tear open gift after gift. Through tired eyes, we smile from the inside out seeing the joy on their faces. And then, in an instant, its over. Were left with an empty tree, a floor covered in paper, and decorations demanding to be taken down. And despite all the smiles, we cant help but feel a little deflated.

But what if there was more? What if the end was only the beginning? What if weve been doing it wrong this whole time?

Now before you start saying, More??! No. Im done. I cant take any more, Alicia! Just hear me out. Id like to propose we start a new tradition in 2020. Bigger, better, longer. I call it Big Rest.

And heres how it works:

1. Question everything:

You know that obnoxious Christmas song about the 12 Days of Christmas that makes no sense? Did you know that the 12 days dont end on Christmas? The 12 days begin on Christmas day and last through January 6th (3 Kings Day). So what does that mean?

But I dont celebrate Christmas, Alicia.

Thats OK. Christmas came from the Yule Festival, a Nordic tradition celebrated in Scandinavia and other parts of northern Europe. It begins on Winter Solstice (roughly December 21 or 22) and lasts for. 12 days! Later, Christians adapted these traditions to the celebration of Christs birth.

The point is, this isnt about traditions or religion. Its about digging a little deeper into why we do the things we do. And just because weve always done things a certain way, doesnt mean we cant change. After all, dont we love bigger, better, faster and more?

2. Lean In:

Now that our holiday boundaries have expanded, lets take a deep breath, and lean into the season longer. We, in the Northern Hemisphere, are in the middle of the winter season. It neither begins or ends with the holiday.

The winter solstice time is no longer celebrated as it once was, with the understanding that this is a period of descent and rest like nature and the animal kingdom around us, this time of hibernation is so necessary for our tired limbs, our burdened minds people are left feeling that winter is hard, because for those of us without burning fires and big festive families, it can be lonely and isolating. But winter is kind. She points us in her quiet, soft way towards our inner self, towards this annual time of peace and reflection, embracing the darkness and forgiving, accepting and embracing the past year and then, just around the corner the new year will begin again, and like a seed planted deep in the earth, we will all rise with renewed energy once again.

-Bridget Anna McNeil

3. Rest:

In the month of December, we rush and eat and spend and consume, and suddenly, its all over in a day. Then we hit the ground running on January 1st and resume our busy lives. Pay off the credit card, lose weight, make more money, hit the quota.

What if instead of ending and beginning the year in a rush, we ended and began with rest? What if we embraced winter, from December through February, as a time of peace and renewal? What if instead of reacting to the previous year, we took time to prepare for the next one? To recharge, to listen, to read, to explore, to plan.

Everything in nature is cyclical phases of the moon, the seasons, even our own bodies. So why, then, do we constantly push and rush and drive without ceasing? The moon isnt full all the time. The earth isnt spring year-round (unless you live in San Diego). The sun doesnt shine 24 hours a day. No, the sun sets, and we rest. In fact, we should spend 30% of our day in rest (approximately 8 hours). And yet, we view it as a burden to productivity instead of the fuel that enables us to create.

When we sleep, our bodies perform very vital functions to recharge, replenish, and heal itself. And all of this without our conscious mind to plan, execute, manage and analyze when we rest, our bodies recharge with no effort from us.

So, if youre left feeling drained at the end of the year, like an overdrawn bank account, I challenge you to prioritize Big Rest in 2020. Let go of unceasing effort, lean in, and trust the power and process of rest. Bigger, better, more and now.

To help illustrate the power of resting a little longer, Ill leave you with this old, but relevant, 80s commercial featuring the kid from A Christmas Story.

Happy Holidays! May they be full of rest and last a little longer.

Read this article:

Re: Your Account Is Overdrawn - Thrive Global

A simple device, just a heat and movement sensor attached to digital camera, has revolutionised the way that conservationists learn about animals in the wild.

Camera traps are a very simple solution to the task of working out when, where and how wildlife interacts with its environment. Monitoring populations without damaging habitats, these relatively simple devices have provided some astonishing finds including revealing species previously hidden in the untouched depths of the forest. Elusive new creatures arent their only speciality, however, as in 2015, similar devices helped reveal that the critically endangered Javan rhinoceros was breeding and significantly adding to its tiny population.

After identifying a likely area for a sighting, usually with the help of local guides, traps are placed at animal height on trees and posts and left to wait until wildlife walks by. A memory card stores the images for conservationists to fetch with these cameras often being deployed for months at a time and collecting hundreds of thousands of images with invaluable information about biodiversity.

And herein lies the problem. Whilst the traps have allowed millions of photographs worth of data to be amassed, the time it takes to weed out empty frames and label the species that cross their lens is astronomical. The delay in processing data can mean that vital information about biodiversity could be out of date before it can be used.

Watch More | Baby dolphin trapped in debris rescued from drowning

Tanya Birch of Google Earth Outreach and Jorge Ahumada from Conservation International explain in a blog post that, while a highly trained human being can label somewhere between 300 and 1000 of these images in an hour, new AI technology could analyze 3.6 million. Despite the challenging task of identifying species, Wildlife Insights is pretty accurate as well. Looking at data from the 614 species it has currently been trained on, the chance that it will guess correctly is over 80%. Perhaps, more importantly, the AI system is very good at removing camera trap images that feature no animals at all which can be up to 80% of the images taken.

This rapid computer analysis would make camera trap data available in near real-time removing the delay in processing that can make performing even basic research an incredibly slow process. Speed is vital in conservation to allow scientists and officials to react to ecological changes and rapid environmental disruption. Making this wealth of information rapidly available will help to support those attempting to save critically endangered species and their habitats.

Google has highlighted the story of scientists in Columbia who, after the demobilization of the FARC in 2017, have been gradually rediscovering parts of the Amazon that they could not enter during the conflict in the region. The short film shows how field biologist Anglica Diaz-Pulido of the Instituto Humbolt is using camera traps to monitor umbrella species like the Jaguar in the forest that skirts the unique Rainbow River region.

The challenges we biodiversity researchers currently face are focused on carrying out research quickly and ensuring it gets to decision-makers in time, Diaz-Pulido explains in the short film. Threats to forests in this region are looming and intervention needs data to be processed rapidly so that recommendations can be passed on to the Minister of Environment. Leaving camera traps for just one month, however, can generate nearly 150,000 images which would take the Instituto Humbolts researchers up to a year to sort through.

Machine learning helps to speed up the processing of this massive amount of data so that the scientists on the raw edge of this impending mass extinction can use the most recent biodiversity figures to inform policymaking. This speedy analysis could help save species on the brink of extinction by getting the news of their decline out there before the images before irrelevant.

Read More | I visited an ecoresort helping to protect Brazil's biodiversity

The launch of Wildlife Insights not only revolutionises the use of camera trap data by scientists but also allows open access to 4.5 million photos that date back to 1990. Ahumada hopes that the platform will help close the gap between the revolution that technology has brought about it allowing this data to be captured and our inability to use it. "AI can dramatically increase the speed at which camera traps can be processed and analyzed," he told euronews Living, "this will allow people to quickly assess wildlife populations from camera trap data and focus on actions that can help save and manage these species."

By making the data accessible Ahmuda hopes that everyone from managers of anti-poaching programmes to governments to children will use the wealth of images to identify wildlife in their area and grow the database. We want citizen scientists, teachers and children to use the platform, he says, These are the future generations who will benefit from wildlife conservation.

Using AI to process camera trap technology is by no means a new idea. Recently DeepMind detailed their work with Tanzanias Serengeti National Park which will see the technology used to identify behaviours impacted by encroaching human activity in the area. As with many other projects, previously processing the millions of images collected by the team relied on volunteers using Zooniverse, a site where people help citizen scientists by identifying and counting species.

Link:

AI from Google is helping identify animals deep in the rainforest - Euronews