industry news

The UK Government has announced plans to open five new centres of excellence focused on the use of artificial intelligence (AI) for digital pathology and medical imaging.

In line with the country’s Industrial Strategy, the new initiative aims to accelerate disease diagnosis with advances in medical technology. The detection of diseases at an earlier stage is expected to improve patient outcomes.

The new AI medical centres will receive a total of £50m of funding under the government’s Industrial Strategy Challenge Fund. They are set to be led by some of the country’s biggest medical companies including GE Healthcare, Siemens, Philips, Leica, Canon and Roche Diagnostics.

Doctors, businesses and academics will join these facilities to create digital technology-based products for medical use. These products will be designed to enable more personalised treatment.

The investment in large-scale genomics and image analysis is anticipated to offer new insights into the development of complex diseases, which would enable medical professionals to administer the right treatment at the right time.

UK Business Secretary Greg Clark said: “AI has the potential to revolutionise healthcare and improve lives for the better.

“The innovation at these new centres will help diagnose disease earlier to give people more options when it comes to their treatment, and make reporting more efficient, freeing up time for our much-admired NHS staff time to spend on direct patient care.”

One of the AI medical centres, London Medical Imaging and Artificial Intelligence Centre for Value-Based Healthcare, will use AI in medical imaging and associated clinical data for fast and early diagnosis.

The Industrial Centre for AI Research in Digital Diagnostics (I-CAIRD) in Glasgow will focus on finding quick and efficient solutions for healthcare challenges.

In Oxford, National Consortium of Intelligent Medical Imaging (NCIMI) will explore the role of clinical imaging in earlier diagnosis and more personalised care.

The Northern Pathology Imaging Collaborative (NPIC) will be opened in Leeds to work on digital pathology research.

Meanwhile, the Pathology image data Lake for Analytics, Knowledge and Education (PathLAKE) will be based in Coventry and leverage NHS pathology data to bolster health-related AI.

All the centres are scheduled to begin operations next year.

Last month, the UK announced an additional investment of £235m to support the development and commercialisation of quantum technologies.

A new study conducted by the Feinstein Institute for Medical Research in the US has demonstrated that genetic tests can be used to identify schizophrenia patients who are likely to respond to treatment with antipsychotic drugs.

Schizophrenia is a mental disorder that results in delusions, hallucinations, disorganised thoughts and behaviour. It is known to be one of the primary causes of disability in the US.

The disease is commonly treated using antipsychotic drugs, which are chosen through a trial-and-error approach. The treatment does not leverage lab tests that could validate the effectiveness of the therapy.

In the latest study, the researchers assessed the use of genetic tests to predict the response to drugs in patients suffering their first schizophrenic episode. The team used polygenic risk scores for testing.

Feinstein Institute assistant professor Jian-Ping Zhang said: “Polygenic risk scores represent the combined effects of many thousands of genetic variants across the entire genome, and better represent the very complex genetic nature of schizophrenia.”

Findings showed that patients with higher polygenic risk scores were less likely to respond to standard antipsychotic therapy. The data was found to be consistent in two independent cohorts, one of which was evaluated by European researchers.

Feinstein Institute professor Todd Lencz said: “The results we found open the door for ‘precision medicine’ approaches to psychiatry and, more specifically, the use of polygene scores as a new technology for the treatment of psychiatric disorders.”

The team intends to expand the study in order to facilitate clinical guidelines for the use of polygenic risk scores and other predictors such as brain scans in schizophrenia treatment.

Findings from the Feinstein Institute study were published in the American Journal of Psychiatry.

A research team at Hong Kong Polytechnic University (PolyU) has developed a robotic arm to facilitate self-help and upper-limb mobile rehabilitation for stroke patients after discharge from hospital.

Referred to as a mobile exo-neuro-musculo-skeleton, the robotic arm enables intensive and effective self-help rehabilitation exercise.

The lightweight device is said to be the first of its kind to combine exo-skeleton, soft robot and exo-nerve stimulation technologies. It is intended to cater to the increasing need for outpatient rehabilitation service for stroke patients.

PolyU Department of Biomedical Engineering researcher Hu Xiaoling said: “We are confident that with our mobile exo-neuro-musculo-skeleton, stroke patients can conduct rehabilitation training anytime and anywhere, turning the training into part of their daily activities.

“We hope such flexible self-help training can well supplement traditional outpatient rehabilitation services, helping stroke patients achieve a much better rehabilitation progress.”

Designed to be flexible and easy-to-use, the robotic arm is compact in size, has fast responses and requires a minimal power supply.

It comprises different components for the wrist/hand, elbow, and fingers that can be worn separately or together for various functional training needs. The device can also be connected to a mobile application, where users can manage their training.

The exo-skeleton and soft robot components of the device offer external mechanical forces guided by voluntary muscle signals in order to facilitate the desired joint movement for the patients.

PolyU improved the rehabilitation by adding its Neuro-muscular Electrical Stimulation (NMES) technology, which allows the robotic arm to contract user’s muscles when electromyography signals are detected.

When tested in a clinical trial involving ten stroke patients, the robotic arm is reported to have led to better muscle coordination, wrist and finger functions, and lower muscle spasticity following 20 two-hour training sessions.

The researchers plan to collaborate with hospitals and clinics for conducting additional trials.

A new study led by a Swiss research institute École polytechnique fédérale de Lausanne has found that electrical stimulation with a wireless spinal implant could enable chronic paraplegia patients to walk again with support.

The STIMO study, conducted in alliance with the Lausanne University Hospital (CHUV), involved three patients with cervical spinal cord injuries that were sustained many years ago.

When treated with targeted electrical stimulation to the lumbar spinal cord along with weight-assisted therapy, the participants regained movement in their paralysed leg muscles and were able to walk over ground with the aid of crutches or a walker.

In September, a study conducted by researchers in the US found that use of an epidural stimulator and physical therapy could help paralysed patients regain some movement.

The Swiss researchers observed that the spinal implant enabled patients to voluntarily control previously paralysed leg muscles even when the electrical stimulation was turned off. This occurred after training for a few months.

EPFL neuroscientist Grégoire Courtine said: “Our findings are based on a deep understanding of the underlying mechanisms which we gained through years of research on animal models. We were thus able to mimic in real time how the brain naturally activates the spinal cord.

“The exact timing and location of the electrical stimulation are crucial to a patient’s ability to produce an intended movement. It is also this spatiotemporal coincidence that triggers the growth of new nerve connections.”

During the latest study, the researchers delivered precise stimulation by implanting an array of electrodes over the spinal cord. These implants are said to have facilitated the targeting of individual muscle groups in the legs.

According to the team, the new rehabilitation protocols are backed by targeted neurotechnology and improved neurological function by enabling patients to train their natural over-ground walking capabilities.

Based on these findings, medical device startup GTX medical will create a tailored neurotechnology to advance the rehabilitation protocols into a treatment for clinical use.

Researchers at Edinburgh University in the UK have discovered a new approach that leverages mathematical methods devised by Second World War codebreaker Alan Turing for detecting a variety of diseases, including cancer, in advance.

The technique devised to break the Enigma code could be used to help measure the effectiveness of diagnostic tools that are currently in use.

At present, statistical techniques that were developed in the 1980s are used to evaluate the accuracy of diagnostic tests.

They are not capable of measuring the usefulness of a test to determine an individuals’ risk of developing a disease.

Now, researchers at Edinburgh University’s Usher Institute of Population Health Sciences and Informatics believe that the methods devised by Turing could improve these.

Usher Institute of Population Health Sciences and Informatics professor Paul McKeigue told media sources: “Most existing diagnostic tests for identifying people at high risk of cancer or heart disease do not come anywhere near the standards we could hope to see.

“The new era of precision medicine is emerging, and this method should make it easier for researchers and regulatory agencies to decide when a new diagnostic test should be used.”

Alan Turing unveiled the method to break the Enigma code in 1941. His approach investigated the distribution of ‘weights of evidence’ work that establishes the potential outcomes in a given situation to help decide the strategy to crack the code.

The study by Edinburgh University researchers, which is published in Statistical Methods in Medical Research, revealed that applying the same principle is expected to facilitate the development of personalised treatments.

New research has revealed that manipulating human memories may become possible as cyberattackers may be able to steal, spy on, alter or control the memories of people implanted with brain stimulation devices.

The research was published as a report by Russian cybersecurity firm Kaspersky Lab in alliance with the UK’s University of Oxford Functional Neurosurgery Group.

According to the report ‘The Memory Market: Preparing for a future where cyber-threats target your past’, radical threats may not be a reality in the near future, but the vulnerabilities already exist in the form of brain implants, which are used to learn more about memories.

These implants come with connected software and hardware that can be potentially exploited by cyberattackers.

University of Oxford Functional Neurosurgery Group doctoral researcher Laurie Pycroft said: “The prospect of being able to alter and enhance our memories with electrodes may sound like fiction, but it is based on solid science, the foundations of which already exist today.

“Memory prostheses are only a question of time. Collaborating to understand and address emerging risks and vulnerabilities, and doing so while this technology is still relatively new, will pay off in the future.”

For the research, the team specifically analysed the vulnerabilities in deep brain stimulation devices, the latest versions of which have Bluetooth-based management software on commercial smartphones and tablets.

The researchers found a serious vulnerability and multiple misconfigurations on an online surgical management platform. They believe that this may offer access to sensitive data and procedures.

Insecure data transfer from the implants could also allow cyberattacks and lead to changed settings that may cause pain, paralysis or data theft.

Some brain implants are focused on patient safety rather than security, and thereby become exposed to tampering by unwanted parties. The researchers also said that insecure behaviour by medical staff leaves the medical devices vulnerable.

The team noted that addressing all such vulnerabilities is important to mitigate the risk of any cyberattacks in the future.

Kaspersky Lab global research and analysis team junior security researcher Dmitry Galov said: “Current vulnerabilities matter because the technology that exists today is the foundation for what will exist in the future.

“Although no attacks targeting neurostimulators have been observed in the wild, points of weakness exist that will not be hard to exploit.”

A team of researchers at Kanazawa University in Japan has identified a potential quantifiable biomarker for diagnosing autism spectrum disorder (ASD).

They used magnetic brainwave imaging to correlate altered gamma oscillation with the motor response of children with ASD.

This is consistent with previous key hypotheses on the disease that autism results from an excitatory and inhibitory imbalance in the brain, which is associated with repetitive gamma oscillations.

Researchers observed a reduction in this type of brain activity during visual, auditory, and tactile stimulation in individuals with ASD.

Two groups of children aged from five to seven years were formed for the research.

In the first group, children were conventionally diagnosed with ASD, while the second group included children classed as developing typically.

Each of them was given a video-game-like task that included pressing of a button with their right finger when they are relaxed.

The children’s brainwaves were monitored during the task using magnetoencephalography, which records magnetic activity from neurons.

Study corresponding author Mitsuru Kikuchi said: “We measured the button response time, motor-evoked magnetic fields, and motor-related gamma oscillations.

“As found in other studies, the ASD children’s response time was slightly slower and the amplitude in their magnetic fields was a bit decreased. The gamma oscillations were where we saw significant and interesting differences.”

The motor task was developed for child participants who performed the motor task involving the 100 button-press, while a whole-head MEG child system was recorded.

Study first author Kyung-min An said: “These findings may prove to be extremely useful in helping us understand the neurophysiological mechanism behind social and motor control development in children with ASD.

“Using magnetoencephalography in this way gives us a noninvasive and quantifiable biomarker, which is something we are in great need of.”

In a UK first, a team from University College London Hospitals (UCLH), University College London (UCL) and Great Ormond Street Hospital (GOSH) has successfully operated on the spinal cords of two babies with spina bifida still developing in the womb.

The fetoscopic surgery was a success, with surgeons able to repair the defect in the spine of the two babies in separate operations this summer. The hospital reports the mothers and babies are recovering well.

The UK-based institutions partnered with University Hospitals Leuven in Belgium for training before carrying out the operations in the UK.

GOSH lead neurosurgeon Dominic Thompson said: “In spina bifida, the spinal canal does not close completely in the womb, leaving the spinal cord exposed from an early stage in pregnancy. This results in changes to the brain, as well as severe permanent damage to the nerves on the lower half of the body.”

UCLH and Leuven lead foetal surgeon Jan Deprest added: “Operating in the womb involves opening the uterus, exposing the spina bifida without delivering the baby, closing the defect and then repairing the uterus to leave the baby safely inside.”

Until now, pregnant women in Britain could choose to have foetal surgery abroad or postnatal surgery but did not have the option for the surgery in the UK.

Surgeons say conducting this kind of surgery in the womb offers better outcomes than postnatal surgery, as patients with spina bifida are often incapable of walking, and may require a series of operations to drain fluid from the brain later in life.

Research in the US prior to the operations found that prenatal closure was associated with a 50% reduction in the need for surgical shunt placement in newborns and resulted in a significant improvement in motor function at 30 months of age.

UCL Great Ormond Street Institute of Child Health Professor Paolo De Coppi said: “The reduction in the need for shunts is particularly important as long-term follow-up of children that have undergone prenatal closure in the womb suggests that brain function, mobility, and total independence were higher in non-shunted than shunted children aged five.”

UCLH clinical director for Women’s Health Professor Donald Peebles said: “The Centre for Prenatal Therapy programme at UCLH and GOSH has been made possible thanks to generous charitable funding totalling £450,000 from GOSH Children’s Charity and UCLH Charity.

“These vital funds have provided training for the surgical team and will fund surgery for the first 10 patients. While we currently only perform an open foetal surgery approach, we are developing the fetoscopic approach in pre-clinical models and hope this could further minimise maternal complications.”

The team is also researching a minimal access fetoscopic technique, which aims to develop new fetoscopic tools and imaging techniques to support prenatal therapy.