Author: Carmen Morcillo Pérez

Mark Girolami, Chief Scientist at The Alan Turing Institute, said “The TRIC-DT will help to democratise access to digital twin technology by providing open and reproducible computational and social tools. These tools will be available for digital twin development and deployment as a national service – freely accessible to the UK research and innovation communities. Working alongside a range of partners, the creation our first TRIC will build on the wealth of digital twin activity and investment.”

The scanned patient suffers from epilepsy and had no lesions identified on their 3T MRI scan and different drugs did not stop their seizures. The scan in the 7T went smoothly and the patient could go home on the same day.

The success of the world-first scan means the clinical team will be able to accept clinical referrals for patients requiring general anaesthetic, and research studies of vulnerable patients who might require anaesthetic intervention, can be supported.

One research study investigating MRI planning before insertion of deep brain stimulators (DBS) is now ready to start. Preparatory works over two years included a testing programme developed with industry partners to determine the safe conditions of usage for anaesthetic and physiological monitoring equipment at 7T.

The scan was facilitated through a collaboration between the King’s Imaging team, the anaesthetic department from Evelina London Children’s Hospital together with colleagues across Guy’s & St Thomas’ NHS Foundation Trust, Philips Healthcare – MR patient monitoring and Dräger – MR anaesthesia workstation. The team has also been awarded a GSTT ‘Learning from Excellence’ Award for the teamwork that allowed the first scan.

MONAI, the domain-specific, open-source medical imaging AI framework that drives research breakthroughs and accelerates AI into clinical impact, has now been downloaded by over 1M data scientists, developers, researchers, and clinicians. The 1M mark represents a major milestone for the medical open network for AI, which has powered numerous research breakthroughs and introduced new developer tools in the last year.

The most notable is Auto3DSeg, a low-code framework. It enables data scientists and researchers of any skill to train models that can quickly segment regions of interest in 3D imaging modalities like CT and MRI. Researchers in the MONAI community finished atop the leaderboard using Auto3DSeg in numerous segmentation competitions like the MICCAI INSTANCE22 and HECKTOR22 challenges.

Swin-UNETR, another research breakthrough from the MONAI community, was published in CVPR with authors from Vanderbilt University and NVIDIA. Swim-UNETR is a state-of-the-art, 3D transformer-based model with self-supervised pretraining.

Over 700 GitHub projects are based on MONAI today. With the release of MONAI 1.0 in September 2022, you can now access 21 of the leading medical imaging models for tasks such as MRI segmentation, breast-density classification, and pathology tumor detection in MONAI Model Zoo.

Last year, MONAI Label expanded into the field of pathology. New features, sample applications, and viewer integrations for annotating pathology images create a starting point for pathologists and data scientists to work together to use the benefits of deep learning.

“MONAI will enable pathologists and scientists to build accurate models without knowing anything about AI. This is an important step in making AI a universal tool for pathology research,”  said  Lee A.D. Cooper, PhD, associate professor of pathology and director, Computational Pathology at the Northwestern University Feinberg School of Medicine.

A collaborative artificial intelligence project between King’s College London, Guy’s and St Thomas’ and Siemens Healthineers, is developing a new MRI scan on the UK’s first MAGNETOM Free.Max scanner, for pregnant women, to give more information to obstetricians about babies as close to birth as possible to optimally prepare for birth.

Study lead, Dr Jana Hutter, Senior Lecturer in the School of Biomedical Engineering & Imaging Sciences, has scanned more than 95 pregnant women for the Meerkat project which aims to develop a self-driving ‘one-click’ exam which also corrects for fetal movement in the womb. Normally when an MRI is performed on a pregnant woman, there are different ‘scanning steps’ which are done one-by-one, but this is not adapted neither to the movement of the baby nor uterine practice contractions for example.

Dr Hutter is using AI to adapt in real-time depending on how the baby is moving. In this way, radiographers have almost like a ‘rolling-feed’ of the baby – in real-time – and can stop when needing to look at something in more detail. For instance, if the fetus is moving a lot, the scanner will automatically change to a dynamic capture mode able to catch the baby’s motion. And if the baby is not moving a lot, the time is used for high-resolution brain scans. Instead of always being suboptimal, the fetal scan thus adapts in real-time. In-line with other ongoing research in fetal MRI at St Thomas’ Hospital, the team are looking at a wide range of pregnancy complications such as pre-eclampsia and preterm birth as well as any clinical referral eg for anomalies in the babies’ brains.

This project requires very close collaboration with many experts, obstetricians to radiographers, midwives, radiologists as well as technical experts from Siemens Healthineers. While 1.5Tesla or 3Tesla would be the normal magnetic field strength for an MRI system, which does yield similar information, the researchers have found using the lower magnetic field strength provides very rich information. Questions we are trying to address next are: will the placenta support the baby for the full pregnancy? Is it safe for the woman to have a vaginal birth?

The lower-field strength is particularly relevant in high-risk populations, such as for women with chronic hypertension, diabetes and bigger women. Due to the design of this scanner, there is a wider opening, meaning scans can more easily be performed on obese patients allowing for higher quality scans when patients have a high percentage of abdominal fat. There are also impressive opportunities for pregnant women who are late in gestation, as radiographers can get all this information very close to birth.

Dr Raphael Tomi-Tricot, Senior MR Collaborations Scientist at Siemens Healthineers and visiting research fellow at the School of Biomedical Engineering & Imaging Sciences said the Perinatal imaging department at King’s College London has a long-standing history of successful studies and investigations on fetal MRI, with very innovative ideas.

“Setting up this new set of developments on the MAGNETOM Free.Max scanner in such a short time is a real feat and it highlights the quality of the research there, with the well-being of the patient always at heart.”

With the involvement of clinicians, the researchers are now looking at new cohorts, such as those with Crohn’s disease, women with hyperemesis gravidarum (extreme vomiting during pregnancy), and other populations which are less reachable, as well as other higher-risk groups that can benefit from the accessible and shorter scan.