An AI tool that can analyse abnormalities in the shape and form of blood cells, and with greater accuracy and reliability than human experts, could change the way conditions such as leukaemia are diagnosed. Researchers have created a system called CytoDiffusion that uses generative AI – the same type of technology behind image generators such as DALL-E – to study the shape and structure of blood cells. Unlike many AI models, which are trained to simply recognise patterns, CytoDiffusion – developed by researchers at the University of Cambridge, University College London and Queen Mary University of London – could accurately identify a wide range of normal blood cell appearances and spot unusual or rare cells that may indicate disease. The results are reported in the journal Nature Machine Intelligence. Spotting subtle differences in blood cell size, shape and appearance is a cornerstone of diagnosing many blood disorders. But the task requires years of training, and even then, different doctors can disagree on difficult cases. “We’ve all got many different types of blood cells that have different properties and different roles within our body,” said Simon Deltadahl from Cambridge’s Department of Applied Mathematics and Theoretical Physics, the study’s first author. “White blood cells specialise in fighting infection, for example. Knowing what an unusual or diseased blood cell looks like under a microscope is an important part of diagnosing many diseases.” However, a typical blood ‘smear’ contains thousands of cells – far more than any human could analyse. “Humans can’t look at all the cells in a smear – it’s just not possible,” said Deltadahl. “Our model can automate that process, triage the routine cases, and highlight anything unusual for human review.” “The clinical challenge I faced as a junior haematology doctor was that after a day of work, I would have a lot of blood films to analyse,” said co-senior author Dr Suthesh Sivapalaratnam from Queen Mary University of London. “As I was analysing them in the late hours, I became convinced AI would do a better job than me.” To develop CytoDiffusion, the researchers trained the system on over half a million images of blood smears collected at Addenbrooke’s Hospital in Cambridge. The dataset – the largest of its kind – included both common blood cell types and rarer examples, as well as elements that can confuse automated systems. By modelling the full distribution of cell appearances rather than just learning to separate categories, the AI became more robust to differences between hospitals, microscopes and staining methods, and better able to recognise rare or abnormal cells. In tests, CytoDiffusion could detect abnormal cells linked to leukaemia with far greater sensitivity than existing systems. It also matched or surpassed current state-of-the-art models, even when given far fewer training examples, and quantified its own uncertainty. “When we tested its accuracy, the system was slightly better than humans,” said Deltadahl. “But where it really stood out was in knowing when it was uncertain. Our model would never say it was certain and then be wrong, but that is something that humans sometimes do.” “We evaluated our method against many of the challenges seen in real-world AI, such as never-before-seen images, images captured by different machines and the degree of uncertainty in the labels,” said co-senior author Professor Michael Roberts, also from Cambridge’s Department of Applied Mathematics and Theoretical Physics. “This framework gives a multi-faceted view of model performance, which we believe will be beneficial to researchers.” The team also showed that CytoDiffusion could generate synthetic blood cell images indistinguishable from real ones. In a ‘Turing test’ with ten experienced haematologists, the human experts were no better than chance at telling real from AI-generated images. “That really surprised me,” said Deltadahl. “These are people who stare at blood cells all day, and even they couldn’t tell.” As part of the project, the researchers are releasing what they say is the world’s largest publicly available dataset of peripheral blood smear images: more than half a million in total. “By making this resource open, we hope to empower researchers worldwide to build and test new AI models, democratise access to high-quality medical data, and ultimately contribute to better patient care,” said Deltadahl. While the results are promising, the researchers say that CytoDiffusion is not a replacement for trained clinicians. Instead, it is designed to support them by rapidly flagging abnormal cases for review and handling more routine ones automatically. “The true value of healthcare AI lies not in approximating human expertise at lower cost, but in enabling greater diagnostic, prognostic, and prescriptive power than either experts or simple statistical models can achieve,” said co-senior author Professor Parashkev Nachev from UCL. “Our work suggests that generative AI will be central to this mission, transforming not only the fidelity of clinical support systems but their insight into the limits of their own knowledge. This ‘metacognitive’ awareness – knowing what one does not know – is critical to clinical decision-making, and here we show machines may be better at it than we are.” The researchers say further work is needed to make the system faster and to test it across diverse patient populations to ensure fairness and accuracy. The research was supported in part by the Trinity Challenge, Wellcome, the British Heart Foundation, Cambridge University Hospitals NHS Foundation Trust, Barts Health NHS Trust, the NIHR Cambridge Biomedical Research Centre, NIHR UCLH Biomedical Research Centre, and NHS Blood and Transplant. The research was conducted by the Imaging working group of the BloodCounts! consortium, which aims to use AI to improve blood diagnostics globally. Simon Deltadahl is a Member of Lucy Cavendish College, Cambridge. Reference: Simon Deltadahl et al. ‘Deep generative classification of blood cell morphology.’ Nature Machine Intelligence (2025). DOI: 10.1038/s42256-025-01122-7 The text in this work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. Images, including our videos, are Copyright ©University of Cambridge and licensors/contributors as identified. All rights reserved. We make our image and video content available in a number of ways – on our main website under its Terms and conditions, and on a range of channels including social media that permit your use and sharing of our content under their respective Terms.

Complex digital images of tissue samples that can take an experienced pathologist up to 20 minutes to annotate could be analysed in just one minute using a new AI tool developed by researchers at the University of Cambridge. This could one day help doctors tailor treatments more effectively, moving to a more nuanced understanding of each patient’s cancer Zeyu Gao SMMILe, a machine learning algorithm, is able not only to correctly detect the presence of cancer cells on slides taken from biopsies and surgical sections, but it can predict where the tumour lesions are located and even the proportion of regions with different levels of aggressiveness. The tool could be used in the future to guide a patient’s treatment, as well as helping scientists better understand how cancer develops and identify new biological signatures to improve detection. Artificial intelligence (AI) tools offer incredible promise towards helping pathologists examine tissue samples from patients with suspected or confirmed cancer, producing ‘spatial maps’ that allow them to understand where the cancer cells are and how they are spreading. But training these tools has until now required a large number of high-quality, detailed reference slides annotated by trained pathologists. In research published today in Nature Cancer, scientists at the University of Cambridge have developed an AI tool that can be trained using slides that have been given simple, patient-level diagnostic labels, such as cancer type or grade. Importantly, these slides did not need to include detailed region-by-region annotations from pathologists, which are time-consuming to produce. Despite learning from such scant information, the algorithm – SMMILe (Superpatch-based Measurable Multiple Instance Learning) – was able to provide detailed information about each slide, including mapping the locations of tumour lesions, and estimating the proportions and spatial distribution of lesions with different subtypes and grades. Dr Zeyu Gao from the Early Cancer Institute at the University of Cambridge, who developed the algorithm, said: “Cancer isn’t always uniform. A single tumour can contain different subtypes, some that are more aggressive than others. Our model doesn’t just say ‘yes, there’s cancer’, it maps out these subtypes and their proportions within the tissue. This could one day help doctors tailor treatments more effectively, moving to a more nuanced understanding of each patient’s cancer.” The team tested the algorithm on eight datasets comprising 3,850 whole-slide images covering six cancer types: lung, kidney, ovarian, breast, stomach, and prostate cancer. When benchmarked against nine other state-of-the-art whole-slide image classification analysis AI tools, SMMILe’s performance matched – and in several cases exceeded – these tools at slide-level classification, while significantly outperforming them when it came to estimating the proportions and spatial distribution of lesions. Dr Mireia Crispin-Ortuzar, Co-Lead of the Cancer Research UK Cambridge Centre Integrated Cancer Medicine Virtual Institute and the study’s joint senior author, said: “What we’ve developed is akin to a ‘sonar’ for images that essentially allows us to see in the dark. Often, we have information about a tumour, but we don't know how it's distributed in the tissue. There are technologies that allow you to get this information, but they are very costly. “With our new AI method, we can accurately map the tumour samples – and the beauty is that it is trained on cheap, widely-available datasets that only contain bulk, non-spatial information.” Although SMMILe is currently focused on classifying tissue slides, the researchers plan to use the tool to predict biomarkers – biological signatures – that reveal how a tumour behaves at a molecular level. This will help further understanding of how cancers develop and spread as well as potentially opening the door to personalised treatment decisions for each patient, guided by both what the tumour looks like and what its biology reveals. Dr Gao added: “By allowing pathologists to make faster, more accurate diagnoses, we can make sure patients receive the best treatment even sooner, improving our chances of successfully treating their cancer. AI could have a huge impact on the lives of our patients.” The research was funded mainly by Cancer Research UK and GE HealthCare. Research Information Manager at Cancer Research UK, Dr Dani Skirrow, said: "We’re living in a golden age of cancer research, with new tools and technologies offering better, faster ways to diagnose cancer and personalise treatments. “This study suggests SMMILe could help doctors quickly get detailed information about a person’s cancer so that they can give each individual the best treatment option for them. Further studies are needed to check how well SMMILe works in the clinic, but these promising early-stage findings show how artificial intelligence tools have the potential to help people receive personalised care sooner.” The University of Cambridge and Addenbrooke's Charitable Trust (ACT) are fundraising for a new hospital that will transform how we diagnose and treat cancer. Cambridge Cancer Research Hospital, set to be built on the Cambridge Biomedical Campus, will bring together clinical excellence from Addenbrooke’s Hospital and world-leading researchers at the University of Cambridge. The research that takes place there promises to change the lives of cancer patients across the UK and beyond. Find out more here. Reference Gao, Z et al. SMMILE enables accurate spatial quantification in digital pathology using multiple instance learning. Nature Cancer; 19 Nov 2025; DOI: 10.1038/s43018-025-01060-8 The text in this work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. Images, including our videos, are Copyright ©University of Cambridge and licensors/contributors as identified. All rights reserved. We make our image and video content available in a number of ways – on our main website under its Terms and conditions, and on a range of channels including social media that permit your use and sharing of our content under their respective Terms.

‘The World Tree’, written by Professor Robert Macfarlane, will be performed for the first time today by the Helsinki Chamber Choir in Finland. I am interested as much in the longue durée of human-forest relations Robert Macfarlane The Sycamore Gap tree was an iconic 120-year-old sycamore tree growing at Hadrian's Wall in Northumberland, England. It was illegally felled in 2023, sparking international outrage. “It was in many ways an axis mundi, a ‘world tree’,” says Macfarlane, Professor of Literature and the Environmental Humanities at the Faculty of English and a Fellow of Emmanuel College. “Although it was a single tree growing where there should really be a forest, it nevertheless became a focus for many of our complex, passionate, contradictory feelings about trees, forests and the living world more broadly.” “I decided that, instead of just writing a series of contemporary requiems … I wanted to widen the whole frame of the work and take a much longer view of tree-human relations in England and beyond.” “In the libretto I pull the temporal lens right back to the re-emergence of trees and forests in northern Europe towards the end of the Pleistocene,” Macfarlane says. The World Tree’s first movement is called ‘Glacial Maximum’. “I am interested as much in the longue durée of human-forest relations as in the acute event of the Sycamore Gap Tree’s felling, and the outpouring of love, grief and anger which followed.” Macfarlane’s libretto has been set to music by the Finnish-Canadian composer Matthew Whittall and the premiere will be conducted by Nils Schweckendiek, himself an alumnus of Clare College, Cambridge. The final movement, ‘The Word for World Is Forest’, imagines a future forest flourishing on the uplands where that one tree once stood. Between these two movements come eight others with titles such as ‘Wildwood’, ‘Song of the Axe’ and ‘Pollen: a Polyphony’. Responding to the conviction of two men, in May 2025, for felling the Sycamore Gap Tree, Macfarlane said: “The historical-psychological echoes and rhymes of this event are many. The inter of ecocide and toxic masculinity is nothing new.” “I thought immediately of the first story of mindless tree-felling by two glory-seeking males: the account in the Epic of Gilgamesh (a text c. 4400 years old in its Sumerian form), of how Gilgamesh and Enkidu travelled to the Sacred Cedar Wood, slew its guardian spirit Humbaba, felled the tallest cedar in the forest and took its lumber — and Humbaba’s head — back to Uruk as trophies.” Robert Macfarlane’s award-winning books, including Is a River Alive? (2025), Underland (2019), Landmarks (2015), The Old Ways (2012) and The Wild Places (2007), have been widely adapted for music, film, television, radio and theatre. He has previously written operas, plays, and films including River (2022) and Mountain (2017), both narrated by Willem Dafoe. As a lyricist, he has written songs and albums with musicians including Cosmo Sheldrake, Karine Polwart and Johnny Flynn, with whom he has released two albums, Lost In The Cedar Wood (2021) and The Moon Also Rises (2023) and an EP, Six Signs (2022). In 2022, with the actor-director Simon McBurney he co-adapted Susan Cooper's classic fantasy novel The Dark Is Rising into a twelve-part BBC audio drama series. The World Tree world premieres at Helsinki’s Temppeliaukio Church on 18th November 2025, with additional performances that week in Kotka, Nurmijärvi and Vihti. The text in this work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. Images, including our videos, are Copyright ©University of Cambridge and licensors/contributors as identified. All rights reserved. We make our image and video content available in a number of ways – on our main website under its Terms and conditions, and on a range of channels including social media that permit your use and sharing of our content under their respective Terms.

The earliest evidence of an internal ‘GPS’ system in an animal has been identified by researchers, which could help explain how some modern birds and fish evolved the ability to use the Earth’s magnetic field to navigate long distances. The tiny magnetic fossils – dating from 97 million years ago – were buried in ancient seafloor sediments, left behind by a mysterious, unidentified organism. Shaped like spearheads, spindles, bullets and needles, and no larger than a bacterial cell, scientists believe these ‘magnetofossils’ are biological in origin, but they don’t know what creature made them, or why. Now, researchers have now solved part of the mystery and found that these fossils may have served as an animal GPS, enabling organisms to read Earth’s magnetic field like a map. The researchers, from the University of Cambridge and the Helmholtz-Zentrum Berlin, captured the first 3D images of the fossils’ magnetic structure, and revealed features optimised to detect both the direction and strength of Earth’s magnetic field, which would have aided navigation. “Whatever creature made these magnetofossils, we now know it was most likely capable of accurate navigation,” said Professor Rich Harrison from Cambridge’s Department of Earth Sciences, who co-led the research. The discovery provides the first direct evidence that animals have been navigating using the Earth’s magnetic field for at least 97 million years. It may also offer insights into how animals evolved this ability, known as ‘magnetoreception’. The results are reported in the journal Communications Earth & Environment. Life has evolved a range of extraordinary senses, and magnetoreception is one of the least understood. Birds, fish, and insects use the Earth’s magnetic field to navigate vast distances, but how they do this is still unclear. One theory is that tiny magnetite crystals within the body align with the Earth's magnetic field, acting like microscopic compass needles. Certain bacteria found in lakes and other bodies of water possess a primitive form of magnetoreception. Chains of tiny magnetic particles inside the bacteria allow them to line up with the magnetic field, helping them swim to their preferred depth in the water column. “At just 50–100 nanometres wide, these particles are the perfect compass needles,” said Harrison. “If you want to create the most efficient magnetic sense, smaller is better.” But the magnetofossils the team studied for the current study are 10 to 20 times larger than the magnetic particles used by bacteria, and were retrieved from a site in the North Atlantic Ocean. Previously, some researchers had argued that ‘giant’ magnetofossils may have served as protective spines. However, model simulations have suggested that they might also possess advanced magnetic properties, something Harrison wanted to explore further. “It looks like this creature was carefully controlling the shape and structure of these fossils, and we wanted to know why,” he said. The researchers applied a new technique to visualise the fossil’s internal structure, revealing how magnetic moments (tiny magnetic fields generated by spinning electrons) are arranged inside the magnetofossil. Until now, scientists had been unable to capture 3D magnetic images of larger particles, such as giant magnetofossils, because X-rays couldn’t penetrate them. The research was made possible using a technique developed by co-author Claire Donnelly at the Max Planck Institute in Germany and carried out at the Diamond X-ray facility in Oxford. “That we were able to map the internal magnetic structure with magnetic tomography was already a great result, but the fact that the results provide insight into the navigation of creatures millions of years ago is really exciting,” said Donnelly. Their images revealed an intricate magnetic configuration, with magnetic moments swirling around a central line running through the fossil’s interior, forming a tornado-like vortex pattern. This vortex magnetism provides ideal properties for navigation, said Harrison, generating a ‘wobble’ in response to tiny changes in the strength of the magnetic field that translate into detailed map information. “This magnetic particle not only detects latitude by sensing the tilt of Earth’s magnetic field but also measures its strength, which can change with longitude,” he said. The geometry of this vortex structure is highly stable, meaning it can resist small environmental disturbances that may otherwise disrupt navigation. “If nature developed a GPS, a particle that can be relied upon to navigate thousands of kilometres across the ocean, then it would be something like this,” he said. In solving the enduring mystery over the fossils’ function, the work also helps narrow the search for the animal that made them. “The next question is what made these fossils,” said Harrison. “This tells us we need to look for a migratory animal that was common enough in the oceans to leave abundant fossil remains.” Harrison suggests that eels could be a potential candidate, since they evolved around 100 million years ago and remain one of the least understood and elusive animals. European and American eels travel thousands of kilometres from freshwater rivers to spawn in the Sargasso Sea. Though they can sense Earth’s magnetic field, how they do so is unclear. Magnetite particles have been detected in eels but not yet imaged directly in their cells and tissues, partly because of their tiny size and the fact they could be hidden anywhere in the body. Harrison worked closely with Sergio Valencia from Helmholtz-Zentrum Berlin in designing the research. "This was a truly international collaboration involving experts from different fields, all working together to shed light on the possible functionality of these magnetofossils,” said Valencia. Despite their as-yet-unknown host, “giant magnetofossils mark a key step in tracing how animals evolved basic bacterial magnetoreception into highly-specialised, GPS-like navigation systems,” Harrison said. The research was supported in part by the European Union, the European Research Council and the Royal Society. Rich Harrison is a Fellow of St Catharine’s College, Cambridge. Reference: Richard J. Harrison et al. ‘Magnetic vector tomography reveals giant magnetofossils are optimised for magnetointensity reception.’ Communications Earth & Environment (2025). DOI: 10.1038/s43247-025-02721-3 The text in this work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. Images, including our videos, are Copyright ©University of Cambridge and licensors/contributors as identified. All rights reserved. We make our image and video content available in a number of ways – on our main website under its Terms and conditions, and on a range of channels including social media that permit your use and sharing of our content under their respective Terms.

Cambridge Dictionary has named 'parasocial' as the Word of the Year for 2025. The word is defined by the Cambridge Dictionary as ‘involving or relating to a connection that someone feels between themselves and a famous person they do not know, a character in a book, film, TV series, etc, or an artificial intelligence’. Cambridge University Press & Assessment, the publisher of the Cambridge Dictionary, says that the year was marked by interest in the one-sided parasocial relationships that people form with celebrities, influencers and AI chatbots. 'Parasocial' is one of several AI-related words that were added or updated in the Cambridge Dictionary this year, including ‘slop’, meaning ‘content on the internet that is of very low quality, especially when it is created by AI’. Simone Schnall, Professor of Experimental Social Psychology at the University of Cambridge, said: “Parasocial is the perfect Word of the Year. "The rise of parasocial relationships has redefined fandom, celebrity and, with AI, how ordinary people interact online. "We’ve entered an age where many people form unhealthy and intense parasocial relationships with influencers. This leads to a sense that people ‘know’ those they form parasocial bonds with, can trust them and even to extreme forms of loyalty. Yet it’s completely one sided." The term dates back to 1956, when University of Chicago sociologists Donald Horton and Richard Wohl observed television viewers engaged in ‘para-social’ relationships with on-screen personalities, resembling those they formed with ‘real’ family and friends. They noted how the rapidly expanding medium of television brought the faces of actors directly into viewers’ homes, making them fixtures in people’s lives. Colin McIntosh of the Cambridge Dictionary said: “Parasocial captures the 2025 zeitgeist. It's a great example of how language changes. “What was once a specialist academic term has become mainstream. Millions of people are engaged in parasocial relationships; many more are simply intrigued by their rise. “The data reflects that, with the Cambridge Dictionary website seeing spikes in lookups for ‘parasocial’.” Adapted from the Cambridge University Press & Assessment website. The text in this work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. Images, including our videos, are Copyright ©University of Cambridge and licensors/contributors as identified. All rights reserved. We make our image and video content available in a number of ways – on our main website under its Terms and conditions, and on a range of channels including social media that permit your use and sharing of our content under their respective Terms.

Deep brain stimulation – implants in the brain that act as a kind of ‘pacemaker’ – has led to clinical improvements in half of the participants with treatment-resistant severe depression in an ‘open label’ trial. Our study hasn’t just highlighted this promise, it’s given us a potential and much-needed objective marker to say which patients will respond best Valerie Voon Significantly, the study, led by researchers in the UK and China, identified a telltale signature of brain activity that predicted how well individual patients responded to the treatment. This could be used in future to target the treatment at those patients most likely to benefit. Major depressive disorder is one of the most common disabling mental health problems worldwide. While antidepressants and cognitive therapies help many patients, there are high rates of treatment resistance. Treatments will fail for between three and five in 10 patients with depression. Over the past few decades, a technique known as deep brain stimulation (DBS) has begun to be used to treat a range of conditions, most successfully for patients with Parkinson’s disease. The technique involves the insertion of thin electrodes deep into the brain that transmit mild electrical stimulation to correct errant brain activity. Open label trial of deep brain stimulation In a study published today in Nature Communications, researchers trialled DBS in 26 patients recruited from Ruijin Hospital, Shanghai Jiaotong University School of Medicine in China, all of whom had treatment-resistant depression. The trial was open label, which means that both researchers and the patients were aware that DBS was being administered. The team applied stimulation to two areas of the brain. The first was the bed nucleus of the stria terminalis (BNST), an extension of the amygdala that is involved in regulating stress, anxiety, fear and social behaviours, particularly in response to long-term stresses and fears. The second area was the nucleus accumbens, which is involved in how the brain processes rewards, and is a key area for motivation, pleasure, and reinforcement. What the researchers found Half of the patients (13 out of 26) saw significant improvements, as measured on different scores for depression- and anxiety-related symptoms along with clinically relevant quality of life and disability scores. Nine of these patients (35% of the study cohort) achieved remission, meaning a near-complete elimination of their symptoms. The researchers recorded brain electrical activity from the DBS electrodes in the BNST and scalp EEG (electroencephalogram) and found brain activity at a specific frequency range (4–8 Hz), known as theta activity, to be clinically important. Theta activity in the BNST correlated with how severe a patient’s depression was and how anxious they felt on a daily basis – those patients with higher levels of theta activity experienced worse symptoms. People with lower levels of theta activity in this brain region before surgery tended to improve more and report greater improvements in quality of life at three, six and 12 months, though only in relation to depression and anxiety, not to loss of pleasure (known as anhedonia). Similarly, those patients with greater ‘coherence’ between the BNST and the prefrontal cortex in theta frequences – in other words, those patients where these two regions were most closely synchronised – were also likely to have better outcomes. The prefrontal cortex is involved in emotional regulation, and greater coherence implies better communication between these two regions. Technique shows 'real promise' Professor Valerie Voon from the Department of Psychiatry at the University of Cambridge and Fudan University, China, said: “Deep brain stimulation shows real promise at tackling treatment-resistant depression, which can have a huge impact on people’s lives. But our study hasn’t just highlighted this promise, it’s given us a potential and much-needed objective marker to say which patients will respond best.” Dr Linbin Wang from the Department of Psychiatry at the University of Cambridge added: “We found that brain activity at a particular frequency – theta brainwaves – could tell us which patients would have the best response to DBS treatment in the BNST brain region. This could help us personalise treatment for individual patients in future.” During the trial, DBS reduced BNST theta activity, and this reduction matched improvements in symptoms of depression and anxiety. This opens up the possibility of using a ‘closed-loop system’ that uses real-time feedback to adjust the electrical stimulation, say the researchers. Professor Valerie Voon added: “Because theta activity tracks anxiety states in real time, it means that if activity is high, we can say ‘OK, this person is an anxious state, we need to turn up stimulation’. Likewise, if theta activity is low, we can turn down the stimulation.” Professor Bomin Sun, the neurosurgeon who led the study at Shanghai Jiao Tong University School of Medicine, said “This is the largest study to show that deep brain stimulation of the BNST and nucleus accumbens can treat depression. Depression is a major public health problem in China and globally. This study not only tells us how the brain is impaired in depression, it also highlights potential of DBS for depression.” The team also found psychological measures that indicated how well a patient would respond to treatment. Participants were shown a series of images, some pleasant (such as puppies), some neutral (such as furniture), and some negative (such as accidents). Patients with the strongest reaction to the negative images were least likely to see benefits from DBS. The research was funded by the National Natural Science Foundation of China, the Science and Technology Commission of Shanghai Municipality. Professor Voon and her team were also funded by the UK Medical Research Council. Alongside this study, the researchers carried out a double-blinded, randomised controlled trial of DBS for treatment-resistant depression. Such trials are considered the ‘gold standard’, as neither researchers nor patients are aware which treatment is being administered, removing the possibility of a placebo effect. The results of this trial will be published shortly. Reference Wang, L & Zhang, Y et al. Prefrontal–Bed Nucleus of the Stria Terminalis Physiological and Neuropsychological Biomarkers Predict Therapeutic Outcomes in Depression. Nat Comms; 18 Nov 2025; DOI: 10.1038/s41467-025-65179-z The text in this work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. Images, including our videos, are Copyright ©University of Cambridge and licensors/contributors as identified. All rights reserved. We make our image and video content available in a number of ways – on our main website under its Terms and conditions, and on a range of channels including social media that permit your use and sharing of our content under their respective Terms.

Dr Chris Macdonald's groundbreaking Virtual Reality public speaking platform won the Technological Innovation of the Year category at the Times Higher Education Awards 2025. Those who use the VR platform can practise in a stadium in front of 10,000 animated spectators, with loud noises, stadium lights, and flashing cameras. Dr Chris MacDonald Dr Macdonald – a Fellow at Lucy Cavendish College – was among a number of University staff recognised for their innovative contribution to research and academia during a ceremony at the Edinburgh International Conference Centre. Cambridge’s PROFILE trial team were shortlisted in the Research Project of the Year: STEM category, for their transformative work on early biologic therapy for newly diagnosed Crohn’s disease. And Technician Development Advisor John Nicolson was shortlisted for Outstanding Technician of the Year, for his leadership and commitment to the technical community. The PROFILE team were shortlisted at the THE Awards. 1 of 2 John Nicolson was shortlisted at the THE Awards. 2 of 2 Prev Next Dr Macdonald’s free, first-of-its-kind platform provides Virtual Reality (VR) training environments, as well as support from an AI coach, to build confidence in people with public speaking anxiety and transform them into skilled and confident presenters. He said: "Prior to a presentation, most students tend to practise on their own, in a highly controlled environment – normally in their bedrooms – to an audience of zero. As a result, it will feel like a significant step up when they present to even a small group of people, and even a subtle audience gesture can throw them off. "By contrast, those who use my VR platform can practise in a different venue every night to a wide range of increasingly distracting audiences and fear-inducing scenarios. They can, for example, practise in a stadium in front of 10,000 animated spectators, with loud noises, stadium lights, and flashing cameras. Accordingly, a subsequent presentation to a small group can feel like a step down. I call this overexposure therapy. It could be thought of as psychological weight training. And it has been shown to build extra adaptability, grit, and resilience." He added: "The goal in my lab is simple but ambitious: build high-impact transformational tools and make them free to all. By working with truly visionary philanthropists, I believe we can make that a reality and transform millions of lives." The text in this work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. Images, including our videos, are Copyright ©University of Cambridge and licensors/contributors as identified. All rights reserved. We make our image and video content available in a number of ways – on our main website under its Terms and conditions, and on a range of channels including social media that permit your use and sharing of our content under their respective Terms.

Researchers have achieved a new level of control over the atomic structure of a family of materials known as halide perovskites, creating a finely tuned ‘energy sandwich’ that could transform how solar cells, LEDs and lasers are made. Due to their remarkable ability to absorb and emit light, and because they are cheaper and can be configured to convert more of the solar spectrum into energy than silicon, perovskites have long been touted as a potential replacement for silicon in solar cells, LEDs and quantum technologies. However, their instability and durability have, so far, largely limited perovskite devices to the laboratory. In addition, scientists have struggled to precisely control the thickness of perovskite films, and control how different perovskite layers interact when stacked together – an important step in building functional, multi-layered structures. Now, a team of researchers led by the University of Cambridge has found a new way to grow ultra-thin layers of perovskite films so their atoms line up perfectly, which could enable more powerful, durable and efficient devices. The researchers used a vapour-based technique to grow three-dimensional and two-dimensional perovskites one layer at a time, which enabled them to control the thicknesses of the films down to fractions of an atom. Their results, reported in the journal Science, could open the door to usable perovskite devices that can be produced at scale, using a process like that used to make commercial semiconductors. Each layer in a semiconductor ‘sandwich’ does a different job in moving electrons and their positively-charged counterparts – called holes – around and determines how the semiconductors absorb or emit light. Together, the layers act like one-way streets that guide the electric charges in opposite directions, preventing them from bumping back into each other and wasting energy as heat. In other widely used semiconductors, such as silicon or newer materials such as gallium nitride, the properties of the individual layers can be fine-tuned using various methods. But perovskites, despite their excellent performance, have so far proved difficult to control in layered devices, due in part to their ‘chaotic’ atomic structure. “A lot of perovskite research uses solution processing, which is messy and hard to control,” Professor Sam Stranks from the Department of Chemical Engineering and Biotechnology, who co-led the research. “By switching to vapour processing — the same method used for standard semiconductors — we can get that same degree of atomic control, but with materials that are much more forgiving.” The researchers used a combination of three-dimensional and two-dimensional perovskites to create and control their atomically-tuned stacks, a phenomenon known as epitaxial growth. This fine control let the team directly observe how the light given off by the material changes depending on whether it’s a single layer, a double layer, or thicker. “The hope was we could grow a perfect perovskite crystal where we change the chemical composition layer by layer, and that’s what we did,” said co-first author Dr Yang Lu from Cambridge’s Department of Chemical Engineering and Biotechnology and Cavendish Laboratory. “It’s like building a semiconductor from the ground up, one atomic layer after another, but with materials that are much easier and cheaper to process.” The researchers also found they could engineer the junctions between the layers to control whether electrons and holes stayed together or apart — a key factor in how efficiently a material emits light. “We’ve reached a level of tunability that wasn’t even on our radar when we started,” said Professor Sir Richard Friend from the Cavendish Laboratory, who co-led the research. “We can now decide what kind of junction we want — one that holds charges together or one that pulls them apart — just by slightly changing the growth conditions.” The researchers found they could tune the energy difference between the layers by more than half an electron volt, and in some cases, extend the lifetime of electrons and holes to over 10 microseconds: far longer than usual. The team says this level of precision could pave the way for scalable, high-performance devices that use light in new ways, from lasers and detectors to next-generation quantum technologies. “Changing the composition and performance of perovskites at will – and probing these changes – is a real achievement and reflects the amount of time and investment we’ve made here at Cambridge,” said Stranks. “But more importantly, it shows how we can make working semiconductors from perovskites, which could one day revolutionise how we make cheap electronics and solar cells.” The research was supported in part by the Royal Society, the European Research Council, the Simons Foundation, and the Engineering and Physical Sciences Research Council (EPSRC), part of UK Research and Innovation (UKRI). Richard Friend is a Fellow of St John’s College, Cambridge. Sam Stranks is a Fellow of Clare College, Cambridge. Reference: Yang Lu, Young-Kwang Jung et al. ‘Layer-by-layer epitaxial growth of perovskite heterostructures with tunable band offsets.’ Science (2025). DOI: 10.1126/science.adx5685 For more information on energy-related research in Cambridge, please visit the Energy IRC, which brings together Cambridge’s research knowledge and expertise, in collaboration with global partners, to create solutions for a sustainable and resilient energy landscape for generations to come. The text in this work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. Images, including our videos, are Copyright ©University of Cambridge and licensors/contributors as identified. All rights reserved. We make our image and video content available in a number of ways – on our main website under its Terms and conditions, and on a range of channels including social media that permit your use and sharing of our content under their respective Terms.

The University of Cambridge has been recognised for the work it does to ensure care experienced students are supported while studying for their degrees. The University has received the Quality Mark from the National Network for the Education of Care Leavers (NNECL) in recognition of its plans to improve student experience and outcomes throughout their time with the University. The University recognises that this work is ongoing Mike Nicholson Mike Nicholson, Director of Recruitment, Admissions and Participation welcomed the news: "The University is delighted to receive the NNECL Quality Mark in recognition of the valuable activity that takes place across the collegiate University to support students from care experienced backgrounds navigate their time at Cambridge. The University recognises that this work is ongoing, and is grateful that making this application has both identified those areas where we already have strengths in the support we offer whilst also helping us see scope for further action and improvement." The NNECL Quality Mark enables institutions to assess their current practice, identify gaps in provision and establish areas for further development. Several Universities and FE Colleges have been awarded it over recent years. Sian Edwards, Programme Manager at the NNECL, praised the University's approach: "We are delighted to award the University of Cambridge the NNECL Quality Mark in recognition of their commitment to supporting students from care experienced and estranged backgrounds. Navigating the complexities of a devolved college model presents unique challenges, yet the University has demonstrated dedication to developing and embedding consistent, inclusive practices across the institution. Their whole-university approach ensures that care experienced and estranged learners are supported to achieve positive outcomes. Congratulations to everyone at the University of Cambridge on this well-deserved recognition." Students attending the university who are care-experienced, or estranged from family, are entitled to the maximum level of financial support available through the Cambridge Bursary Scheme. With other grants and awards added, this can amount to more than £8,000 a year. Colleges will also provide accommodation all year round as these students often do not have homes to return to out of term-time. In 2018, the University of Cambridge was one of the first signatories to the government’s Care Leaver Covenant which commits partners to providing educational and career opportunities to young people exiting the care system. Around 38 care experienced students were admitted to an undergraduate degree course or to the University’s Foundation Year in 2025 but the true number might be higher as this figure is based on students who declare their status at registration and might not include all those who are estranged from family. The Principal of Homerton College, Lord Simon Woolley, was brought up by foster parents in the 1960s and 70s. He has spoken at a number of events and conferences aimed at breaking down barriers for care experienced teenagers: “It is imperative that both the University and the Colleges recognise the challenges for those who have been in care and have had the amazing journey to be here at Cambridge. We need to acknowledge them but also ensure we have an infrastructure in place that helps them go from A to B and on to further success. The NNECL Quality Mark is a recognition that we are on the right track." Widening participation co-ordinator, Kirstyn Kedaitis, accepted the Quality Mark on behalf of the University. She says: ‘The action plan we submitted in our NNECL Quality Mark application maps out our goals for improvement over the next three years. We have already started a staff project centred around best practice in supporting students from highly under-represented groups, including backgrounds of care experience and estrangement. In January, we will complement this by establishing a working group focused specifically on addressing our Quality Mark action plan. This working group will include representation from Cambridge’s care experienced and estranged student community.’ More information about the NNECL’s Quality Mark can be found here. Information and guidance on the support available to students at Cambridge who are care experienced or estranged from family can be found here. The text in this work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. Images, including our videos, are Copyright ©University of Cambridge and licensors/contributors as identified. All rights reserved. We make our image and video content available in a number of ways – on our main website under its Terms and conditions, and on a range of channels including social media that permit your use and sharing of our content under their respective Terms.