Imaging

Allscripts comes in on top in the UK/Ireland, Canada and Oceania. Epic is No. 1 in Southeast Asia and the Middle East. Cerner is first in Africa.

Taipei Veterans General Hospital, a national first-class medical centre and a teaching hospital located in Beitou, Taipei, will utilise the Philips IntelliSite Pathology Solution to transform its pathology tissue examination to digital diagnostics, the Dutch health technology giant announced yesterday. The Philips IntelliSite Pathology Solution allows tissue samples to be remotely viewed within a virtual pathology network across hospital locations, thereby helping TPVGH to establish Taiwan’s first fully digitalised pathology department. WHY IT MATTERS Using the Philips IntelliSite Pathology Solution with the Ultra-Fast Scanner, TPVGH will now be able to digitise tissue samples so that pathologists can review, interpret, analyse, and share digital images. The Image Management System that is part of the Philips IntelliSite Pathology Solution allows them to instantly consult colleagues or conveniently present images during multi-disciplinary team meetings, without the need to physically transport pathology slides or tissue samples. In addition to clinical use, the digital images will be stored in image repositories and used to teach pathology students, or used for medical research, including cancer research. THE LARGER TREND Last month, South Korea’s Seoul St. Mary’s Hospital introduced a digital pathology system that aids in the diagnosis of cancer, also utilising the Philips IntelliSite Pathology Solution, Healthcare IT News reported.  ON THE RECORD “Philips Taiwan continues to help Taiwan in realising intelligent and advanced healthcare solutions,” said Richard Hu, General Manager of Philips Taiwan in a statement. “Digitalising pathology so that tissue samples can be viewed remotely, wherever they are needed, will not only enhance laboratory efficiency and quality but also improve patient safety.”

Sydney Neuroimaging Analysis Centre (SNAC), an Australian company co-located with the University of Sydney’s Brain and Mind Centre, is building AI tools to automate laborious analysis tasks in their research workflow, such as isolating brain images from head scans and segmenting brain lesions. Additional algorithms are in development and being validated for clinical use. One compares how a patient’s brain volume and lesions change over time. Another flags critical brain scans, so radiologists can more quickly attend to urgent cases. The researchers develop their algorithms using the NVIDIA Clara suite of medical imaging tools, as well as cuDNN libraries and TensorRT inference software. WHAT’S THE IMPACT SNAC analyses patient MRI and CT scans acquired at clinical sites around the world. With a training dataset of more than 15,000 three-dimensional CT and MRI images, SNAC is building its deep learning algorithms using the PyTorch and TensorFlow frameworks. One of the centre’s AI models automates the time-consuming task of cleaning up MRI images to isolate the brain from other parts of the head, such as the venous sinuses and fluid-filled compartments around the brain. Using the NVIDIA DGX-1 system for inference, SNAC can speed up this process by at least 10x. Using semi-automatic methods, the process would take SNAC’s analysts 20-30 minutes but now it can be reduced to two or three minutes of pure machine time, while performing better and more consistently than a human. Another tool tackles brain lesion analysis for multiple sclerosis cases. In research and clinical trials, image analysts typically segment brain lesions and determine their volume by manually examining scans — a process that takes up to 15 minutes. AI can shrink the time needed to determine lesion volume to just three seconds. That makes it possible for these metrics to be used in clinical practice as well, where due to time constraints, radiologists often simply eyeball scans to estimate lesion volumes. COLLABORATIONS AND FUTURE PLANS The centre collaborates with I-MED, one of the largest imaging providers in the world, as well as the computational neuroscience team at the University of Sydney’s Brain and Mind Centre. The group also works closely with radiologists at major Australian hospitals to validate its algorithms.  SNAC plans to integrate its analysis tools with systems already used by clinicians, so that once a scan is taken, it is automatically routed to a server and processed. The AI-evaluated scan is then passed on to radiologists’ viewers — giving them the analysis results without altering their workflow. THE LARGER TREND Also located in Sydney, Macquarie University and Macquarie Medical Imaging has recently partnered with GE Healthcare and Fujitsu Australia to research the ways artificial intelligence can help diagnose and monitor brain aneurysms on scans faster and more efficiently. The university will provide clinical expertise for the development and testing of the technology, which is provided by GE Healthcare, while Fujitsu will lead the initiative. Singapore’s National Neuroscience Institute (NNI) is working with local medtech company specialising in AI and surgical robotics, Iota MedTech to develop a system to sort brain scans of head injury patients in order of urgency, Healthcare IT News reported in May this year. The development of the priority sorting system will help ensure that patients requiring immediate medical attention receive the care they need. ON THE RECORD “We often refer to manual annotation as the gold standard for neuroimaging, when it’s actually probably not,” said Tim Wang, director of operations, SNAC in a statement. “In many cases, AI provides a more consistent, less biased evaluation than manual classification or segmentation.”

Tristan van Doormaal, a neurosurgeon at UMC Utrecht in the Netherlands, details how augmented reality and virtual reality can help patients understand their condition better and train residents in different approaches to surgery.

The 10-year initiative will focus on workforce development, education and training for next-generation care delivery.

Accenture's Kaveh Safavi discusses healthcare's outsized spend on labor – and how machine learning, telehealth and consumerism are transforming how work is done and who does it.

Seoul St. Mary’s Hospital, which was established in 2009 and part of the Catholic Medical Centre (CMC), recently announced that it has been operating a digital pathology system that aids in the diagnosis of cancer since last month. The system, known as the Philips IntelliSite Pathology Solution, allows the hospital’s pathology team to obtain accurate diagnosis results through a digitalised method. The system automatically creates, visualises, and manages digital pathology images based on an image management system, including a slide scanner, a server, a storage device, and a viewer. The digitalisation of pathology images will allow pathologists to make diagnoses through computer monitors. In addition, clinicians from other departments in the hospital can digitally access the digital pathology images easily, increasing efficiency and opening more opportunities for collaboration. THE LARGER TREND In 2017, the Philips received FDA clearance to market its IntelliSite Pathology Solution for primary diagnostic use in the US. The solution also received the green light from South Korea’s Ministry of Food and Drug Safety in June 2018. In the US, Mount Sinai and LabCorp announced last month that they would collaborate to establish the Mount Sinai Digital and Artificial Intelligence-Enabled Pathology Centre of Excellence. LabCorp, has implemented the Philips IntelliSite Pathology Solution in four of its laboratories and plans to introduce it to additional laboratories, will use its experience and expertise to lead the integration of digital pathology into clinical practice across Mount Sinai’s hospitals. A report by Frost & Sullivan in February 2019 “sees wider use and better management of the clinical datasets owned by major medical institutions as a major boon for developers of digital pathology tech.” ON THE RECORD “Through the establishment of a digital pathology system, we will be able to shift from previous ways of interpreting analog microscopic image to a more digitalised technology, which will boost efficiency and productivity,” said Lee Youn-soo, a professor at St. Mary’s Department of Pathology in a statement.

A Frost & Sullivan report predicts that as many as 45% of ORs will be integrated with intelligent technologies within the next four years to improve the precision and predictability of surgical services.

Pre-treatment scans were input into a deep-learning model, which analyzed the scans to create an image signature that predicts treatment outcomes.

According to a report by Meticulous Research, the global Vendor Neutral Archive (VNA) market & Picture Archiving and Communication System (PACS) market is expected to grow at a CAGR of 7.2% from 2018 to 2023 to reach $3.87B by 2023. In China, the demand of PACS & VNA is growing, as 70% of the medical procedures need images, including CT, X-ray, and magnetic resonance imaging. Esteban Rubens, Global Enterprise Imaging Principal, Pure Storage, shared his insights on the PACS market in the Asia Pacific market and thoughts on the developments of AI in medical imaging. Trends and developments in the Asia Pacific PACS Market “I think cloud plays a large part of what people are looking at in Asia Pacific, even more so than in the US and Europe. In the US, there's some resistance to moving the traditional areas of healthcare imaging (PACS and VNA) to the public cloud. In Asia, there’s much more interest in leveraging the cloud for imaging and I think the region is more advanced along that journey,” said Rubens. In China, there is massive investment from both the government and the companies specifically around imaging but the focus is more on access to healthcare and bringing healthcare to people who are underserved. Outside of tier one or tier two cities in the country, the healthcare experience can be complicated with long waits and queues. For the case of Japan, there is access to good healthcare but the emphasis is more on early detection and tackling issues related to a rapidly aging population.  “There are a lot of new players in imaging, but they seem to be only capturing the low end of the market. The high end is dominated by the big players from the US and Europe. Looking at the region, GE Healthcare made huge investments in China for decades and are now reaping the rewards. In Japan, Fujifilm dominates their home market. Outside of those two companies, its fragmented elsewhere in Asia and we expect some consolidation to take place in the coming years,” he added. Helping healthcare organisations manage their increasing appetite for data Due to the large volumes of imaging data as well as storage and accessibility concerns, Rubens believes that the best approach for healthcare organisations is a hybrid one, keeping some images on-premise and others in the cloud.  “We have solutions that essentially run in the cloud such as Cloud Block Store, which allows the customer to duplicate the experience they have with a Pure FlashArray on-premise in the cloud regardless of their cloud provider. There is also ObjectEngine, the result of a recent acquisition we made, that allows people to put their backup copies in the cloud with up to 90% data reduction, which is unheard of. This has made it really viable for organisations to do backup and restore to the cloud.” Backup for healthcare organisations is important because of regulatory requirements. Due to the on-going threats from hackers, they would also want to keep as many backups as possible. However, this can get expensive especially if they have to backup petabytes of images. ObjectEngine integrates the client’s backup application and the backup is sent to the cloud where data reduction is applied, which results in lower costs and also the flexibility to restore that data. “Backups are only as useful as the restores, which is where the rubber meets the road. We believe that focusing on the rapid restoration of data from backups is crucial, especially in healthcare. If you get hit by ransomware, for example, you can't afford to wait days or weeks to restore your data and applications, especially on the clinical side,” shared Rubens. AI developments in medical imaging According to Rubens, AI is really good at is detection and segmentation, especially of images and thus a natural fit for medical imaging. In addition, there is a worldwide shortage of radiologists – there is an exponentially growing set of images but not enough trained people to interpret them. “The only way to bridge that gap is technology and AI is such a perfect fit. We’ve trained computers to do triage, take accurate measurements and identify lesions. There are very specific applications for each algorithm and it’s a big job to decide which images to send to which algorithm. So the challenge is now bringing all of that from the research side to the clinical side,” he explained. Leveraging data and reaping the benefits of automation  Regardless of their size, healthcare organisations are holding data and that is very valuable both for patients in terms of improved care as well as for the organisations themselves in terms of efficiency and cost savings. However, approaching data may be a daunting task, especially for smaller healthcare organisations or hospitals. Rubens’ advice for them was to start small: “You don't have to be a huge research hospital with a medical school attached, or an engineering school, to do this kind of research. It's all very democratic, you can download these models from the internet, go to GitHub, open source stuff and then you do tutorials on Coursera so you can start training models with your own data.” “Every organisation has someone who is interested in AI, so it's important to support those people and recognise that what they want to do is important to the organisation's development and goals,” he concluded.