Artificial Intelligence

Also: Sensyne Health announces three-year research alliance with the University of Oxford Big Data Institute; local authorities in England are invited to apply for a share of £1m in funding for digital innovation in social care.

The company will be demonstrating its new patient experience tool built in collaboration with Adobe and Microsoft in a partnership revealed last year.

Companies say cloud-based offering will help customers tap artificial intelligence tools for administrative, technical and physical safeguards.

Companies will show the new tech at HIMSS19.

New technology using AI to tell the difference between harmless moles and dangerous melanomas has hit the market. Created by FotoFinder Systems, Moleanalyzer pro is a portal that lets physicians confirm their skin cancer diagnosis using evaluation techniques, combining specialist expertise with AI and including the option of receiving a second opinion from international skin cancer experts. FotoFinder Systems Global Brand Director Kathrin Niemela told HITNA that the technology aims to aid skin cancer diagnoses. According to the Cancer Council Australia, every year skin cancers account for around 80 per cent of all newly diagnosed cancers in Australia, with GPs seeing more than a million patients per year for skin cancer. In addition, the Australian Government identified that there were 14,320 new cases of melanoma skin cancer diagnosed in 2018, accounting for 10.4 per cent of all new cancer cases diagnosed. “The earlier skin cancer is detected, the better the prognosis. The leisure behaviour of sunbathing in many parts of the world makes early detection of skin cancer more important worldwide,” Niemela said. FotoFinder Systems first calculates and compares size, diameter and structure of moles and quantifies their percentage deviations. Moleanalyzer pro works with deep learning. Its Convolutional Neural Network was ‘trained’ with a large data collection of dermoscopic images and corresponding diagnoses. Through growing experience and its own autonomous rules, it then distinguishes between benign and malignant lesions. “Moleanalyzer pro features the possibility to manually evaluate lesions according to acknowledged checklists and optionally contains an innovative algorithm based on AI, allowing a risk-of-malignancy evaluation,” Niemela said. “In the last few years, the new algorithm has been trained with a large number of dermoscopic images. FotoFinder Systems has an international network of partners who contribute to the training of the algorithm with their pictures of histologically proven lesions.” The analysis then determines a risk assessment score of both melanocytic and non-melanocytic skin lesions, allowing physicians to verify their diagnoses. FotoFinder Systems is working towards making this AI score available for doctors on mobile devices. “When this technology becomes available for mobile devices, rural physicians, for example, who practice far away from clinics or specialist centers can use the Moleanalyzer pro's deep learning algorithm on their mobile phones to get a second opinion on their diagnosis of skin lesions,” Niemela said. The application also allows physicians to request a second opinion from skin cancer experts. “The AI represents a ‘silent virtual colleague’ that delivers a virtual opinion simply, uncomplicatedly and at any time. But together with the human experience delivered by the optional second opinion service, the tool helps to increase diagnostic accuracy.” [Read more: New bid to map AI’s impact in radiology | New AI imaging solution to accelerate critical patient diagnoses] According to Niemela, a man-against-machine study involving 58 dermatologists from 17 nations found that whereas the experts correctly identified 86.6 of malignant skin tumours, Moleanalyzer pro successfully detected 95 per cent. In addition, the technology identified 82.5 per cent of benign naevi correctly, while the experts identified 71.3 per cent as benign. However, Niemela said the technology was not expected to replace specialists. “As fascinating as AI is, it cannot take the place of human experience in the matter of skin cancer. AI will increasingly find its way into dermatology and mole examinations by supporting physicians, not by replacing them,” Niemela said. “Doctors need to combine total body mapping with video documentation of single moles and AI-based evaluation. The combination of these three elements are the pillars of early skin cancer detection. Only a physician with profound knowledge and experience can map this complex process. “In addition, patients do not want to do away with doctors under any circumstances and want to combine high-tech solutions with specialist competence.” And the future potential for AI in skin cancer detection is huge. “The aim of AI is to bundle global knowledge and consistent diagnostic standards – independent of the practice location – all over the world. The combination of human experience and AI can contribute to a drastic improvement in diagnostic accuracy in early skin cancer detection, with the potential for almost 100 per cent accuracy,” Niemela added.

The company will showcase its new Rapid Response Analytics product at the conference and discuss trends including digital health and the practical uses of artificial intelligence.

In an effort to deliver on safety for patients through the establishment of protocols around AI, the Royal Australian and New Zealand College of Radiologists (RANZCR) has created a working group that determines how the technology fits into the world of radiology and healthcare. Composed of practicing radiologists, data scientists, computer scientists and professionals in AI, the working group will explore what AI means for radiologists, the safety of the technology and training needed for doctors to use it. RANZCR President Dr Lance Lawler said that the group will also be working towards evaluating the impacts of the technology, the ethics of it, as well as how it fits into accreditations and regulatory frameworks. “We’re trying to be proactive with it so that we don’t end up in a situation where implications are not thought through,” he added. “Nowhere have we seen anyone seriously looking at these issues, which is often the way that regulations and new technologies come about. We want the technology to be used so that the benefits of it are reaped without being exposed to undue risk.” According to Dr Lawler, AI is of huge potential to radiology as it’s “very good at image recognition and pattern recognition”, which the field uses for image analysis to diagnose and follow up on diseases. But, even with the increasing hype around AI, there has not been enough work done to understand what the technology is, what it will be good for, what the risks of it are and how it can be applied to deliver better quality, lower cost healthcare to Australian patients, he said. As such, the organisation is embarking on a multi-layered approach to the issue. Dr Lawler said it will be looking into how AI is working in other countries, particularly in the US, with the FDA approving certain AI algorithms for use under certain conditions. “We want to understand their approach, and how that applies to us,” he said. On the other hand, it will be looking at bringing to market a training element that educates medical practitioners about the technology. “This new technology will be a tool that doctors use, so they have to know about this technology in order to deliver safe healthcare to their patients. The training element will look into what they need to know, how it will be delivered, how it will be assessed, etc.” The working group is also collaborating with Standards Australia to focus on standards and regulation. “We’re working with Standards Australia to decide what the standards are that need to be met, as well as the minimum standards for the applications of the technology, and then use those standards as the basis for safely regulating it, the way that everything else is regulated in healthcare,” Dr Lawler said.    He addressed the need for a more combined, coherent response from government departments to actively investigate the use of this technology and involving medical academia and healthcare providers in strategy for it. “The ministries and regulatory bodies have been conspicuously quiet with AI, so we’d like to start engaging them in even basic discussions about this technology because it will make a very big impact, it’s only a matter of when.” Is AI replacing radiologists? Dr Lawler also spoke about concerns that roles will change significantly with AI infiltrating the market, in terms of machines replacing work done by radiologists. “There has been a lot of hype around this technology, some resulting in statements made about the future of radiology. They suggest that radiologists will be replaced by machines. Hypes aren’t based on facts and this is just an assumption,” he said. “There’s always going to be work for radiologists and it may just be an issue of reapplying people to different areas. There is fear of the unknown and the purpose of the working group is to put some shape around this whole issue so that people aren’t afraid of it.” At the moment, there’s very little impact of AI on health and radiology because it’s not being used to its full potential and is still under development, according to Dr Lawler. But he said AI is the “next big thing for the industry” and “is a great move for society as a whole”. “This is because there are some things that AI can do better than humans, but there are also some things that humans do better than AI so, we need to find a way to balance the two to deliver better quality healthcare overall,” he said. “And the best way to do that is to get ahead of the curve than be chasing our tails, which happens a lot in healthcare. “We want to get to a point where there’s an accepted and routine use of certain AI tools for some clinical circumstances. For radiology, that may be for breast cancer or lung cancer screening, or comparing responses to treatment – basically in high volume, repetitive cases that machines can do easily.” This article originally appeared on Healthcare IT News Australia.

A new report from the Duke-Margolis Center for Health Policy explores some of the policy changes that should be made to enable safer and more effective deployment of artificial intelligence in healthcare. As AI and machine learning become de facto ingredients in many key clinical technologies, a better understanding of how they can best be leveraged for optimal analytics and decision support is the goal of the study, "Current State and Near-Term Priorities for AI-Enabled Diagnostic Support Software in Health Care." WHY IT MATTERS The Duke report takes stock of the existing legal and regulatory landscape for algorithm-based CDS and diagnostic support software, and lays out some essential priorities to work toward in the years ahead to ensure safe deployment of AI in clinical settings. These aren't just theoretical concerns. AI and ML are making inroads all over healthcare, of course, and current legislation and regulatory policy – whether it's the massive 21st Century Cures Act or FDA's new updates to the Software Pre-Cert Pilot Program – are adequate but still not optimal for a future that promises to evolve at a dizzying pace. The Duke-Margolis paper, meant as a "resource for developers, regulators, clinicians, policy makers, and other stakeholders as they strive to effectively, ethically, and safely incorporate AI as a fundamental component in diagnostic error prevention and other types of CDS," looks at some of the major challenges and opportunities facing AI in the years ahead. Stakeholders like those listed about will need to grapple with big questions, more than a dozen researchers and authors write. Such as: Making a case for the value of more widespread adoption of these technologies. Such evidence would include how the software improves patient outcomes, boosts quality and lowers cost of care, gives clinicians relevant information in a manner they find "useful and trustworthy." Assessing the potential risk of using those products in clinical settings. "The degree to which a software product comes with information that explains how it works and the types of populations used to train the software will have significant impact on regulators’ and clinicians’ assessment of the risk to patients when clinicians use this software," said Duke researchers. "Product labeling may need to be reconsidered and the risks and benefits of continuous learning versus locked models must be discussed." Seeing to it that such systems are deployed in a way that's both flexible and ethical. More and more health systems will need to develop best practices that can mitigate any bias that could be introduced by the training data used to develop software, they explained. That's the only way to ensure that "data-driven AI methods do not perpetuate or exacerbate existing clinical biases." Also, these organizations will have to think hard about the data implications as the products scale up into settings that may be different from initial use cases. And, of course, "new paradigms are needed for how to best protect patient privacy," according to the report. THE LARGER TREND As the technological capabilities and clinical applications of AI-enabled decision support continue to expand, the Duke researchers said more regulatory clarity from agencies such as FDA, which has signaled an appetite for much wider approval of machine learning apps, is needed to protect patients from wanton use of the "black box" algorithms that many have warned about. In addition, there are other major areas that need ironing-out. Among them: proper allowances for patient privacy and data access, and the ability for these fast-emerging technologies demonstrate value and ROI for providers. In all of those, hospitals and health systems have an active role to play. Then there are all sorts of other technical questions that exist – but haven't necessarily been answered, certainly not on a consistent or widespread basis. Such as: how new approaches to labeling different software might improve understanding of its inner workings; how to weigh the relative risks and benefits of locked versus continuously learning models of AI; how to evaluate its performance over time most effectively; how to mitigate data bias; how to assess "algorithmic adaptability" and more. ON THE RECORD "AI is now poised to disrupt health care, with the potential to improve patient outcomes, reduce costs, and enhance work-life balance for health care providers, but a policy process is needed," said Greg Daniel, deputy director for policy at Duke-Margolis, in a statement. "Integrating AI into healthcare safely and effectively will need to be a careful process, requiring policymakers and stakeholders to strike a balance between the essential work of safeguarding patients while ensuring that innovators have access to the tools they need to succeed in making products that improve the public health," he said. "AI-enabled clinical decision support software has the potential help clinicians arrive at a correct diagnosis faster, while enhancing public health and improving clinical outcomes," added Christina Silcox, managing associate at Duke-Margolis and co-author of the report. "To realize AI’s potential in health care, the regulatory, legal, data, and adoption challenges that are slowing safe and effective innovation need to be addressed." Twitter: @MikeMiliardHITN Email the writer: mike.miliard@himssmedia.com Healthcare IT News is a publication of HIMSS Media.

The AI vendor will be discussing these 2019 healthcare and technology trends with attendees at HIMSS19 in Orlando next month.

Artificial intelligence (AI) has disrupted numerous industries in recent years, but for the technology to work effectively, the technology needs to be used right. For the healthcare sector, one of the end goals is to provide better patient outcomes, minimise human errors and alleviate some of the physical and mental burnout felt by healthcare practitioners as a result of the volume of admin work required. A study in the US found that for every hour that physicians spend providing direct clinical facetime to patients, almost two additional hours are spent on desk work. By utilising AI and analytics, this can be reduced, and by extension, so too will the rates of mental illness. For this to happen, however, the industry must first get ready for the AI era by building up skills in reading, working with, analysing and arguing with data – also known as data literacy. Data is the lifeblood of AI; which is what makes AI and analytics the ideal combination. Doctors are constantly receiving data from their patients, often pertaining to the symptoms of an illness or injury and how it can be treated. Healthcare professionals must develop their skills so as to confidently interpret this data and accurately input it into the system to fuel AI. Recent research by Qlik found that only 12 per cent of Australia’s healthcare professionals are data literate, behind the global average of 15 per cent. But, it is important to note that Australia has not been given an opportunity to skill up appropriately. In my time in the healthcare industry, I found that staff learn best when they are given a flexible learning environment to develop their skills in their own time and in a way that does not add to the pressure they are already under. This is why Qlik offers online learning via Qlik Continuous Classroom, where people can login and undertake modules that suit their learning needs at a time that is convenient for them. And for the more junior professionals entering the industry, these types of educational undertakings should be included as part of their structured learning to build the next generation of leaders that are data literate from the start. Embedding data analytics and AI within the healthcare industry is not as easy as just rolling out a software installation; it requires an entire strategy and cultural transformation. The new technology must be ingrained in the ways that doctors and physicians work. In addition to embedding analytics into the workflow, performance KPIs should be put in place to ensure that employees are working towards improving their data literacy and in turn, using it for data-based decision making. It’s also vital that the adoption of analytics is not just restricted to one level i.e. just surgeons, but is embraced by all levels, including admin staff, nurses, GPs, etc. It is the responsibility of the organisation as a whole to drive best practice procedures and provide feedback to ensure that everyone is getting the most out of the analytics platform. This can be done by continuously engaging users in conversation about their data and results, and presenting the results to their peers. MercyAscot in New Zealand has done this extremely well – using the Qlik platform to act on patient feedback quickly and making changes that cover all aspects of the hospital. The organisation recognised the need to more effectively leverage the significant volume of data collated over the years to improve the delivery of services to its patients. Rather than rely on static reports which were often time-delayed, MercyAscot created a system that allows staff to act more quickly on patient feedback to make quality improvements. In addition, connecting to multiple data sources and systems to see how various departments link together was paramount for it to drive optimum efficiencies holistically. MercyAscot’s staff are now interested in the data and openly discuss the numbers and improvements that it can make with using this data. “As an organisation, we were data rich, but information poor. We struggled to process our data fast enough to use it effectively across our business functions and knew this had to change,” MercyAscot Director of Medical Services Dr Lloyd McCann said previously. But it’s not just healthcare professionals that are skilling up to make the most of AI and analytics. The recently launched Data Literacy Project features organisations’ and industry professionals’ real-life stories of how analytics has made an impact on their day-to-day work. As part of this, an online tool was developed to allow organisations to discover their own Corporate Data Literacy score, against which companies can benchmark themselves. I encourage the healthcare industry to see where they stand on this scale and map out their future data literacy path based on individual scores. I hope to soon see a future where the healthcare sector is able to ensure that employees can make full use of the benefits of AI correctly and efficiently.   Charlie Farah is the Healthcare and Public Sector Director at Qlik APAC.