The Arctic is continuing to warm twice as fast as the rest of the planet; air temperatures every year from 2014 to today have been warmer than all previous records since 1900.
The reasons for this trend include less snow and ice to reflect sunlight, warmer oceans releasing heat into the atmosphere later into the fall, and increasing winter cloudiness, according to the US National Oceanic and Atmospheric Administration (NOAA).
NOAA released its 13th annual Arctic Report Card, synthesizing the peer-reviewed work of 81 scientists from 12 countries, on 11 December 2018 at the American Geophysical Union (AGU) Fall Meeting in Washington, DC. The report sounded warnings about the current effects of Arctic warming in such diverse fields as caribou populations and poisoning of shellfish.
Emily Osborne, who heads NOAA’s Arctic research programme, told reporters at AGU that the higher air and water temperatures are pushing the Arctic into uncharted territory. The average temperature this past year was 1.7 °C above the long-term average, she said, and an unusually sluggish and wavy high-altitude jet stream coincided with a heat wave at the North Pole and a swarm of severe winter storms in the eastern US.
The 12 lowest extents of summer Arctic sea ice have occurred in the past 12 years, according to Donald Perovich of Dartmouth College, US, lead author of the report’s sea ice chapter. The 2018 winter extent is the second lowest on record and the minimum summer extent the sixth lowest.
“Sea ice continues to be younger, thinner, and less stable,” said Perovich, adding that “ice over four years of age makes up only 1% of the Arctic ice pack. The oldest ice has declined by 95% in the past 33 years”. The younger ice reflects back less solar radiation, contributing to further warming.
In the Bering Sea, ice reached an all-time record low for virtually the entire winter, Perovich told reporters. During two weeks in February 2018, usually the time of greatest ice growth, the Bering Sea lost around 215,000 square km of ice.
Perovich and his colleagues attribute the loss to persistent southerly circulation that brought in warm air and surface water temperatures that pushed sea ice northward. The significance, he said, was that Bering Sea ice supports one of America’s most valuable commercial fisheries and is home to a variety of seals, birds, and fish.
Howard Epstein of the University of Virginia, US, reported on land effects of warming, which can lead to either greening or browning of the tundra landscape. The trend is taking its toll on herds of caribou and wild reindeer, important species across the Arctic, he said, and their numbers are declining sharply.
Similarly, reported Karen Frey of Clark University, US, toxic algal blooms are discussed for the first time in this year’s report. Toxins are now found in Arctic birds, mammals, and fish, posing food safety issues for coastal communities where these species form a major part of the diet.
The 2018 Report Card also covers plastic pollution for the first time, especially microplastics, which are more highly concentrated in the Arctic than in other ocean basins. “We are at the beginning of understanding how this problem is affecting the Arctic,” Frey said.
Each year since 2009, Physics World has been awarding one excellent popular-science book with the title of Book of the Year, not to mention creating a shortlist of nine other top titles from all the books we reviewed that year. We also love talking about physics books, and ever since our first such podcast in 2011, we get together each December to discuss our shortlist and reveal our winner. As is becoming a tradition, this chat was hosted by our regular podcast presenter and producer Andrew Glester, in his garden shed, where he can often be found musing about “science fiction, science fact and everything in-between” for his own podcast the Cosmic Shed.
As this year’s winner is the 10th to bag our Book of the Year, we decided to catch up with some previous winners to see what they are working on today; to chat about how their books have aged; and hear what they would do differently today. Tune in to the podcast to hear 2009 winnerGraham Farmelo talk about Paul Dirac and his family; find out what 2015 winnerAmanda Gefter is working on today; hear what 2010 winnerAnil Ananthaswamy has to say about travel and science-writing; and find out more about hippies and physics from 2012 winnerDavid Kaiser.
Of course we also discuss the various exciting books on the 2018 shortlist, and reveal our 10th winner of the Physics World Book of the Year, so tune in to the podcast to hear from a host of interesting writers and scientists.
We hope that everyone will find something to appreciate on this list, and hopefully we have given you a few ideas for some excellent holiday presents.
Shortlist for Physics World Book of the Year 2018 (in no particular order):
Two new approaches to creating 3D images using X-rays could improve screening for diseases and the study of very fast processes, as well as enable analysis of material properties and provide structural information of opaque objects with unprecedented detail, an international research team has discovered.
One approach may reduce the X-ray doses needed for some types of medical imaging, such as breast cancer screening. The other approach may allow 3D imaging of delicate biological samples or the study of very fast processes to speed development of more durable materials.
Led by Andrew Kingston of the Australian National University along with a team at the European Synchrotron Radiation Facility (ESRF) in France, the researchers demonstrated for the first time that the imaging approach known as ghost imaging can be used to obtain 3D X-ray images of the interior of objects opaque to visible light (Optica 10.1364/OPTICA.5.001516).
“Because of the potential for significantly lower doses of X-rays with 3D ghost imaging, this approach could revolutionize medical imaging by making X-ray screening for early signs of disease much cheaper, more readily available, and able to be undertaken much more often,” said senior author David Paganin, from Monash University in Australia, in a statement. “This would greatly improve early detection of diseases, including cancers.”
Horizontal (A) and vertical (B) 2D slices through the 3D X-ray ghost tomography (GT) reconstructed volume; corresponding 2D slices (C, D) through conventional 3D tomography reconstructed volume obtained from the same experiments. Semi-transparent rendering (E), horizontal (F) and vertical (G) cutaway images of the GT volume. (Courtesy: A Kingston et al Optica 10.1364/OPTICA.5.001516).
Ghost imaging correlates two X-ray beams that individually do not carry any meaningful information about the object. One beam encodes a random pattern that acts as a reference and never directly probes the sample, while the other beam passes through the sample.
The researchers created random X-ray patterns by shining a bright beam of X-ray light through metal foam. They took a 2D image of this random beam and then passed a very weak copy of it through the sample. A large-area, single-pixel detector captured X-rays that passed through the sample. The process was repeated for multiple illuminating patterns and sample-object orientations to construct a 3D tomographic image of the object’s internal structure.
As a proof-of-concept experiment, the group performed ghost X-ray tomography on an aluminium cylinder with a diameter of 5.6 mm and containing two holes smaller than 2.0 mm in diameter. The researchers were able to produce 3D images with 1.4 million voxels.
X-ray ghost imaging is a new field that needs to be further explored and developed, the study authors noted. “With more development, we envision ghost X-ray tomography as a route to cheaper and, therefore, much more readily available 3D X-ray imaging machines for medical imaging, industrial imaging, security screening, and surveillance,” Kingston stated.
High-brilliance X-ray sources
In the other paper, a team from the Paul Scherrer Institute in Switzerland, led by Marco Stampanoni, and researchers from the Deutsches Elektronen-Synchrotron (DESY) in Germany and the ESRF, acquired 3D images using high-brilliance X-ray sources. Their new approach uses a single exposure to obtain 3D information from X-rays 100 billion times brighter than a hospital X-ray source. The rays can only be produced at specialized synchrotron facilities (Optica 10.1364/OPTICA.5.001521).
The technique can make the measurements necessary to create a 3D image before destruction of the sample, so it could be useful for studying the mechanics of delicate biological samples, such as examining the internal 3D structure of intact viruses or proteins.
The new single-shot approach uses a crystal to split one incoming X-ray beam into nine beams that simultaneously illuminate the sample. Using detectors oriented to record information from each beam allows researchers to acquire at once nine different 2D projections of a sample object before it is destroyed by the intense X-ray probe beams.
The researchers used the approach to image a moth, which demonstrated the potential for studying insect mechanics with 3D microscale resolution at speeds ranging from microseconds to femtoseconds. They also showed they could achieve nanoscale resolution by imaging a gold nanostructure.
The researchers plan to use their single-shot multiprojection imaging technique to better understand insect biomechanics, which could inspire new engineering setups. They also want to study new, lighter materials that might lower fuel consumption for vehicles, and plan to examine the fast processes that occur when space debris hits satellites, which could aid development of protective materials.
“Stunning” high-resolution images of 20 nearby protoplanetary discs have given astronomers a wealth of new information about how planets form around stars. Taken by the ALMA radio-telescope array in Chile, the images reveal that gas-giant planets ranging in size from Neptune to Saturn can form much faster than previously thought. A comprehensive study of the images also suggests that gas giants can be created much further from their host stars than had been expected. The observations also provide important clues about how Earth-like rocky planets are created.
Over the past three decades, astronomers have catalogued nearly 4000 planets orbiting stars other than the Sun. These extra-solar planets – or exoplanets – include exotic objects such as “hot Jupiters” and “super Earths” and astronomers now know that exoplanetary systems exist in myriad forms – with some being very different to our familiar solar system.
While astronomers have yet to observe an exoplanet that is truly Earth-like, studying how exoplanetary systems form and evolve could provide important clues about how the Earth itself came into being. Armed with this knowledge, astronomers could work-out how many Earth-like exoplanets should be lurking nearby in the Milky Way.
Sticky dust
Systems of planets form from protoplanetary discs of gas and dust that surround young stars. The conventional model says that this process occurs slowly over many millions of years and in a hierarchical manner – with dust particles colliding and sticking together to create larger objects that then collide and stick together.
The ALMA survey was done by the DSHARP collaboration and was led by Sean Andrews at the Harvard-Smithsonian Center Astrophysics; Andrea Isella of Rice University; Laura Pérez of the University of Chile; and Cornelis Dullemond of Heidelberg University. Light from the host stars interacts with dust, causing it to glow with radio emissions. The team made systematic comparisons of the structures of 20 discs and were able to discern features as small as several astronomical units (1 AU is the distance between the Earth and Sun) in some of the discs.
Unseen planets
The team found that some structures such as concentric gaps and narrow rings are present in nearly all of the discs. Spiral and arc-like features were also spotted in some of the discs. They believe that some small-scale features are created by unseen gas-giant planets, and their existence in relatively young protoplanetary discs – some just one million years old – suggests that gas giants form much faster than previously thought.
“The most compelling interpretation of these highly diverse, small-scale features is that there are unseen planets interacting with the disc material,” explains Andrews.
Some of these features are very distant from the host stars, with the furthest being more than 100 AU away – about three times the distance between Neptune and the Sun. This suggests that some gas giants form in very large orbits, which was not expected before astronomers began observing protoplanetary discs.
The astronomers believe that protoplanetary disc structures could hold the answer to reconciling an apparent flaw in the hierarchical model of planet formation. In a smooth disc, once an object reaches about 1 km in diameter it is prone to being sucked into the star. This is not necessarily the case once the disc has become structured, and this could explain how rocky planets like Earth could form relatively close to stars. Planetary formation should also proceed more rapidly within a ring – where dust density is high — thus accelerating the planet-formation process.
Hans Castorp’s X-ray is like a moment out of Frankenstein.
The protagonist in Thomas Mann’s 1924 novel The Magic Mountain is visiting his cousin Joachim, who is suffering from tuberculosis at a sanitarium in the Alps. The story is set shortly before the First World War and it’s only a few years after Wilhelm Röntgen’s discovery of X-rays. Within a few weeks of his arrival, Hans himself gets sick and goes together with Joachim for the procedure. They descend a set of stairs to a dark room. Hans watches as an assistant positions Joachim, instructs him to hold his breath, and then flips a switch.
“For two seconds,” writes Mann, “the dreadful forces necessary to penetrate matter were let loose” as jets of flame fuelled by thousands of volts erupt all over the machine. “Discharges exploded like gunshots. The gauges sizzled with blue light. Long sparks cracked along the wall.”
What a difference a century makes. When I recently had an X-ray. I sat in an ordinary chair in an ordinary, well-lit room. A polite technician covered my lap with a heavy blanket and vanished behind a window. Harmless red lines crossed my body to aim the machine. I heard a soft buzz. The technician reappeared almost instantly and directed me to another office. There my pictures were displayed on a large, flat wall-mounted screen. The doctor and I discussed them as pleasantly as if we were talking about vacation photos.
Mann based his description on close knowledge of X-ray procedures at the time. His wife Katia had been a student of Röntgen and had had one; Mann visited a sanitarium and may have had one too. But in the century since, not only has the procedure of getting X-rayed changed, so has its meaning.
Reality matters
The X-ray scenes in Mann’s novel embody ambivalent feelings about truth. The issue comes up early in the book, when Castorp reminiscences about his grandfather, who had raised him after his own parents died. Castorp realizes that his memories of his grandfather had “lapses and eccentricities” that fail to capture the man. “His essential reality was quite different, much more handsome and authentic than his everyday appearance”. The truth, Castorp thinks, is embodied in a life-sized, magnificent portrait on canvas of “his authentic and real grandfather”, compared with which his household memories are but “a temporary, imperfectly adapted improvisation”.
Years later, in the sanitarium, Castorp takes X-ray images as a new and deeper kind of truth. But while the painting of his grandfather is of his exterior and directly accessible to perception, the X-ray picture is more removed, showing something beyond the visible, and has been created by technicians with special equipment who harnessed nature’s powers. Its truth has to be interpreted by experts, and is mysterious and perhaps dangerous. Surveying the equipment in the sanitarium’s X-ray room, Castorp thinks: “You couldn’t tell if you were in a photographer’s studio, a darkroom, or an inventor’s workshop and sorcerer’s laboratory.”
Watching Joachim’s interior become transparent, Castorp is stunned by the sight of a green glob pulsing in the middle of the fluorescent plate: “Good God, it was the heart.” When Castorp contemplates his own X ray, he is alarmed by being able to see through his flesh to the one bodily part that will survive his death: his skeleton. “For the first time in his life he understood that he would die”. The doctor tells him, “Spooky, isn’t it? There’s no mistaking the whiff of spookiness”.
Castorp tries to ignore, yet cannot avoid, that spookiness. At one point he loftily tells a friend that X-rays provide “real diagnostic certainty” and “positive knowledge”. Suspicious, the friend remarks that X-rays can deceive – whereupon Castorp accuses him, with some bravado, of not believing in science. But Castorp’s attraction to X-rays is more than cerebral; it’s partly mystical and at times erotic. He asks the woman he loves to give him hers as she leaves for a trip so that he can contemplate the framework of her body, “delicately surrounded by the soft, ghostlike forms of her flesh”.
Through its protagonist, Mann’s novel makes clear that X-rays don’t solve, but only complicate, the problem of truth. Nor, for that matter, do any of today’s scientific tools, such as genetic analysis or neuroscience, that objectify features of human life. By showing us that we don’t have to choose between the portrait and the X ray – between art and science – and that we can live comfortably with them together, the novel shows us a deeper truth about human life that cannot be experimentally verified but only artistically revealed.
Entropy effects
Novelists sometimes create fascinating plot twists with the aid of props – such as computers, vaccines and weapons – derived from various branches of science and physics in particular. A few others use science to help reach deeper truths about humans and their relation to the world. Mann is not the only one. James Joyce wove ideas about entropy throughout Portrait of the Artist as a Young Man (1916) and Ulysses (1922), in which entropy is a central metaphor connecting the universe, personal development, and individual creativity. All of these, in Joyce’s works, exhibit thermodynamic system-like behaviours of energy cooling or disintegrating.
Entropy also fascinated the American novelist Thomas Pynchon, who appealed to it directly in his short story “Entropy” (1960) and in his novel The Crying of Lot 49 (1966), using it to connect ideas about the degeneration of weather, information, and ultimately civilization itself. Unlike Mann, both Joyce and Pynchon use a scientific idea, not to affirm a deeper truth, but to undermine truth and expose its fragile and fraught foundation.
But the well-read readers of Physics World are surely aware of more examples than these. Send them to me and I’ll write about them in a future column.
By Liam Quinn from Canada – Adélie Penguin, CC BY-SA 2.0, https://commons.wikimedia.org/w/index.php?curid=24445015
In 2014 pink guano deposits that showed up in Landsat images revealed that some 1.5 million Adélie penguins live on the Danger Islands off the Antarctic Peninsula in a previously unknown colony. Now scientists have used the precise shade of pink to investigate what the penguins eat. Adélie penguins in the west of Antarctica tend to eat more krill whilst those in the east prefer fish, as Casey Youngflesh of the University of Connecticut detailed at a press conference at the AGU Fall Meeting in Washington DC.
It’s possible that these differences are because there are fewer Antarctic silverfish near the Antarctic Peninsula, perhaps because of changes in sea ice. Youngflesh is continuing investigations on this front.
Youngflesh collected guano samples and analysed them in “a small unventilated room in the belly of a ship”, as he told journalists. “It’s quite smelly.” Chemical analysis of each sample revealed what the birds had eaten, enabling Youngflesh to link the colour of guano in satellite images to penguin diet via statistics.
Images from the satellite archive indicates that the penguins haven’t changed their diet over time.
You can find out more about Antarctica’s penguins, and also become a “penguin detective” at penguinmap.com, a portal created to provide policymakers with easily accessible information.
I remember being particularly terrified after watching The Day After, which was a fictional account of nuclear war that aired 35 years ago in the US on 20 November 1983. The two-hour film follows several characters in and around Kansas City as they lived through a large-scale nuclear attack on the US.
The Bulletin of the Atomic Scientists has special report by contributing editor Dawn Stover about the impact of The Day After on American politicians and the public. Many people commented that they had nightmares after watching the terrifying scenes of destruction (I certainly did) and it is claimed by some that the film played a role in encouraging US politicians to pursue nuclear arms treaties with the Soviet Union.
Stover’s article is called “Rising nuclear tensions: echoes of 1983” and includes a clip of the attack sequence in The Day After, which shocked viewers across the US with how fast it could have happened.
The Bulletin has also published a separate interview with the veteran American TV journalist Ted Koppel, who moderated a panel discussion that was broadcast after the film. Panellists included Henry Kissinger, Robert McNamara, Carl Sagan and Elie Wiesel and you can watch it above.
Relations between the US and the Soviets had been particularly chilly in 1983, when Ronald Reagan was in his third year as president. According to Stover, just two weeks before The Day After was broadcast the Soviets had suspected that a NATO military exercise was in fact a ruse in preparation for war. At about the same time, the US began to deploy Pershing II missiles to West Germany – nuclear weapons intended for use in a battlefield.
The Bulletin of the Atomic Scientists is famous for its countdown clock to Armageddon and when The Day After appeared in 1983, the clock was set at four minutes to midnight. Today, it is at two minutes to midnight.
A new 3D nanofabrication technique called Implosion Fabrication could be used to create a wide variety of nano- and microstructures not previously possible. The technique, which can print 3D objects of nearly any shape by patterning a polymer scaffold with a laser and then shrinking the structure to a thousandth of its original volume, might be used to make novel optical metamaterials and electronics devices.
Shrinking hydrogel scaffold
Most existing nanofabrication techniques are limited in what they can produce. Direct laser writing methods, for example, can produce 2D patterns but not 3D ones, which need to be built up a layer at a time – a process that is difficult and slow. Lithography, one of the oldest nanofabrication techniques, can again only print 2D layers on patterned surfaces.
Researchers led by Edward Boyden of the Massachusetts Institute of Technology have now put forward a new technique in which they use a laser to create patterns of reactive chemical groups inside a hydrogel scaffold. They then deposit material (which can be anything from quantum dots, a piece of DNA or gold nanoparticles) into the reactive groups in 3D. Finally, they dehydrate the gel (using an acid), which implodes the scaffold and the printed object it contains to a thousandth its original volume and form a nanoscale structure.
The researchers showed that their ImpFab technique works by printing highly conducting 3D nanostructures in silver that keep their complex array shape after shrinkage.
Almost any shape is possible
“As mentioned, most lithography techniques in use today are limited to 2D and while direct laser writing can fabricate 3D objects, these must be connected and self-supporting and must be made out of specific polymers suitable for the laser writing process,” explain team members Daniel Oran and Sam Rodriques. “In contrast, our technique has no limitation on the geometries it can fabricate – and the structures don’t need to be self-supporting or connected. We can also directly pattern more functionally useful materials like metals or semiconductor nanocrystals.”
Commonly available equipment could help democratize ImpFab
“Traditional lithography also requires massive fabrication facilities and extraordinarily expensive equipment along with costly and often toxic materials, they add. “Our process makes use of very cheap biocompatible materials and more commonly available equipment (like the two-photon microscopes we employed in this work). A simple process like ours could help democratize the field.”
“Ultimately we believe that this technique will revolutionize the way people think about 3D nanofabrication,” Boyden tells Physics World. “Ever since the invention of photolithography, nanofabrication has been dominated by planar processes in which you pattern a surface and then deposit materials onto the surface. We have extended this concept into the third dimension: we pattern a scaffold and then deposit materials volumetrically onto the scaffold. In the short run, this technology will allow us to make structures of interest for photonics research (for example, specialized lenses to study the fundamental properties of light), but in the longer term, we see it being used to create new kinds of optics, electronics and metamaterials.”
The researchers, reporting their work in Science 10.1126/science.aau5119, say they are now looking to apply their approach to patterning different kinds of materials and finding collaborators who might be interested in trying it out. “We have a long history of technology sharing and this topic will be no different,” says Boyden.
Dose distributions for a head-and-neck cancer patient. Left column: predicted plan; centre: automatic generated plan; right: voxel-by-voxel difference maps. (Courtesy: Med. Phys. 10.1002/mp.13271)
A multi-institutional team in Shanghai has developed an automated treatment planning system for intensity-modulated radiation therapy (IMRT) based on 3D dose prediction and dose distribution-based optimization. The researchers demonstrated that the deep-learning-based method shows promise for generating highly individualized radiotherapy plans optimized for precision and quality (Med. Phys. 10.1002/mp.13271).
Treatment planning software packages that use knowledge-based techniques to estimate dose–volume histogram (DVH) objectives are commercially available. But according to the authors, such methods do not provide reasonable estimates of patient-specific achievable dose distributions. They point out that because a large number of dose distributions satisfy the same DVH objectives, this may result in plans with acceptable DVH objectives but unacceptable dose distributions in some areas of clinical concern.
Lead author Jiawei Fan from Fudan University Shanghai Cancer Center and co-authors have presented a deep-learning method for 3D dose prediction for head-and-neck cancer treatments. The system provides estimates of DVH objectives, as well as giving voxel-level feedback to planners showing where the dose distribution could be improved.
To predict the achievable dose matrix for a given patient anatomy, the researchers developed a residual network-based framework trained to correlate voxel geometry to voxel dose. The input data for the network were acquired from individual trans-axial CT slices of the patient’s anatomy; the output was the dose matrix for each slice.
The model also included data from two additional images: an image of the planning target volume (PTV) describing its region, shape and size; and an image describing the shape, size and relative position of 12 organs-at-risk (OARs). These OARs included the brainstem, spinal cord, right and left parotid glands, right and left temporal lobes, right and left lenses, right and left optic nerves, chiasm and oral cavity.
Each training sample included one CT image, two delineation images and one dose distribution image. The researchers utilized data from 270 patients who underwent IMRT for head-and-neck cancers. They subdivided the dataset into a training set of 195 patients, a validation set of 25 patients and a testing set of 50 patients.
After training and testing their deep-learning model, the researchers compared the predicted dose distributions and DVH curves with results calculated by the treatment planning system. The results demonstrated that the deep-learning method could predict clinically acceptable dose distributions. Tests on a range of different prescription patterns with a large number of plans revealed that the automated plans were comparable to those that had been manually generated by dosimetrists.
The authors note that one distinctive feature of their method is its ability to obtain acceptable predictions for prescription doses that were not included in the training set. The model can also be used on patients with several PTVs that have different prescription doses.
The researchers are currently investigating the applicability of voxel-based dose prediction for other cancer cases. They are also working to improve the predicted results for small-volume OARs by using a training set with consistent clinical priorities and plan qualities to train the deep-learning network.
“The proposed automated treatment planning strategy opens up the possibility of a voxel-by-voxel cost optimization system and eliminates the need to convert a desired dose distribution to DVH values,” the authors conclude. “In the future, it would be a dramatic improvement if a physician’s clinical intent could be directly reflected in the target dose for each voxel through a treatment planning interface designed to make use of the predicted 3D dose distribution.”
My mum left a Doctor Who book on my bed when I was about six, and casually mentioned that it was probably a bit too old for me. She’d worked out my psychology – I read it just to prove I could. I was hooked on the characters, but most of all on the possibilities that travelling in space and time brought. At the same time, I was busy idolizing my older brother and sister, who were both pretty good at school, particularly my brother who focused on maths and science, so I copied what they did.
It wasn’t until I got to A levels (age 16–18) that the power of physics really kicked in for me. I had a chaotic but inspirational teacher. I had to do a lot of reading of my own to cover the syllabus, but the passion with which he described his PhD was something else. I went to the University of Manchester to study physics and ended up staying there to do a PhD in particle physics.
Did you ever consider a permanent academic career after completing your PhD?
Only very briefly. I’d figured out by then that while I was decent enough when it came to learning, I didn’t particularly enjoy the day-to-day coding and detector maintenance that was part of being a researcher in particle physics. I preferred learning and applying the concepts that others had discovered, rather than doing the primary research itself.
How did your interest in healthcare policy develop?
I went into management consulting with McKinsey & Co after my PhD, and after three years there I took a secondment to the Prime Minister’s Delivery Unit in the Cabinet Office. They gave me the title of “senior problem solver”, which sounded very grand. It was 2003 and Tony Blair was trying to reform health and justice in the same way that he’d done in education during his first term. After a spell looking at prolific offending, I was asked to look at waiting times in the NHS and that was it. I was fascinated by how healthcare – one of the most basic needs in society – was being delivered and how that could be improved. The excitement of presenting ideas to the prime minister and his team was intoxicating.
After three years in the Cabinet Office, I joined a couple of friends from McKinsey to set up and run our own consultancy – 2020 Delivery – focusing on solving problems for NHS organizations and government. I kept the links with people in central government, though, and when four years later I heard that a role was going at 10 Downing Street, I put my application straight in. I landed the government health adviser job, working jointly for David Cameron and Nick Clegg.
What was it like working at 10 Downing Street?
It’s hard to describe the feeling of going through that iconic black door each morning. There’s a strong sense of history, of responsibility and, most of all, of the need to deliver fast. 24-hour news cycles and five-year parliaments force the discipline of providing advice quickly and accurately. I worked as hard as I’ve ever done, and rarely switched off, even at weekends. I don’t think I’d have done the job if I’d known how tough it would be, but it was one of the most exciting times in my professional life and I enjoyed (almost) every minute of it.
What was the main motivation for you, when it came to this demanding role?
There’s always a buzz in being part of a team, and all the more so when the results of the team are very much in the public eye. And every so often it’s possible to do something that really improves people’s lives – like doubling the research money for tackling dementia, or ensuring that all hospitals and GPs can easily get feedback on how they are doing from their patients. I used to be fairly cynical about politicians, but my time working with them changed that – almost without exception, they want to make the world a better place and put in long hours to do so.
How did you become interested in improving healthcare services using AI technologies?
I joined the Care Quality Commission after leaving Number 10, helping design how hospitals, care homes and GPs should be regulated, and how their quality can be compared. GP practices fascinate me – there’s very little outcome data, and what is discussed between a patient and their GP is intensely private. How could we understand which practices were performing well, and why was it that so many surgeries had such long waiting times? I wanted to find better and more transparent ways to deliver primary care, and I moved to Babylon Health because it brings together machine learning, remote consultations and clinicians, to change the way people can see a doctor or get advice.
What does your current role as director of NHS services at Babylon Health involve?
I look after the NHS side of Babylon – my team and I are a bit like a snow plough for the rest of the company, in that we’re there to create a clear path for our technology and clinical services teams to deliver. We’re trying to give people choice about how they access healthcare.
What are some of the projects you are working on at the moment?
Just over a year ago we launched the UK’s only digital-first NHS GP practice. More than 35,000 people in London have already switched their NHS practice to “Babylon GP at hand”, as it’s called. They get appointments with NHS GPs, usually within two hours, 24/7 – all free like any other NHS practice. The difference is that in-person consultations are the exception, not the rule. People can check their symptoms and build their “digital twin” using our AI, with GP appointments on the smartphone. There are Babylon clinics around London, but more than 85% of the time, a smartphone appointment is all that’s needed.
How has your physics background been helpful in your work, if at all?
It’s hugely helpful, but in slightly less obvious ways than applying field theory or doing redshift calculations. It’s given me the confidence to break down problems into a structure and work through the analyses needed to find solutions. I’ve done a lot of quantitative work in my career, and the ability to manipulate numbers and apply them to real-world issues is a big positive.
More subtly, I’ve found that people change the way they discuss things with me once they know I was a particle physicist. The scientists I work with are willing to explain their work in more detail, and the medics tend to respect the work that goes into a quantitative PhD. It’s a sort of badge that says “go on, try me – I might understand”.
Any advice for today’s students?
I’ve got plenty of health-policy advice, but that may not be so interesting. Maya Angelou once said that people will forget what you said, people will forget what you did, but people will never forget how you made them feel. I’ve loved that from the moment I heard it – physics is a hardcore, conceptually tough subject, but it’s still a very human endeavour. And humans need to feel good to do their best.
One other thing – it’s best not to wear luminous trousers or sport a mullet haircut. I’m generally happy with where I’ve got to, but the less said about my student fashion-sense the better.
