Artificial Intelligence (AI) holds great promise for improving the delivery of healthcare and medicine worldwide, but only if ethics and human rights are put at the heart of its design, deployment, and use, according to the World Health Organization (WHO).

WHO’s report, Ethics and governance of artificial intelligence for health, is the result of two years of consultations held by a panel of international experts appointed by WHO.

“Like all new technology, artificial intelligence holds enormous potential for improving the health of millions of people around the world, but like all technology it can also be misused and cause harm,” said Dr Tedros Adhanom Ghebreyesus, WHO Director-General. “This important new report provides a valuable guide for countries on how to maximize the benefits of AI, while minimizing its risks and avoiding its pitfalls.”

Artificial intelligence can be, and in some wealthy countries is already being used to improve the speed and accuracy of diagnosis and screening for diseases; to assist with clinical care; strengthen health research and drug development, and support diverse public health interventions, such as disease surveillance, outbreak response, and health systems management.

AI could empower patients to take greater control of their own health care and better understand their evolving needs. It could also enable resource-poor countries and rural communities, where patients often have restricted access to health-care workers or medical professionals, to bridge gaps in access to health services.

However, WHO’s new report, published on June 28, cautions against overestimating the benefits of AI for health, especially when this occurs at the expense of core investments and strategies required to achieve universal health coverage.

It also points out that opportunities are linked to challenges and risks, including unethical collection and use of health data; biases encoded in algorithms, and risks of AI to patient safety, cybersecurity, and the environment.

For example, while private and public sector investment in the development and deployment of AI is critical, the unregulated use of AI could subordinate the rights and interests of patients and communities to the powerful commercial interests of technology companies or the interests of governments in surveillance and social control.

The report also emphasizes that systems trained primarily on data collected from individuals in high-income countries may not perform well for individuals in low- and middle-income settings.

AI systems should therefore be carefully designed to reflect the diversity of socio-economic and health-care settings. They should be accompanied by training in digital skills, community engagement and awareness-raising, especially for millions of healthcare workers who will require digital literacy or retraining if their roles and functions are automated, and who must contend with machines that could challenge the decision-making and autonomy of providers and patients.

Ultimately, guided by existing laws and human rights obligations, and new laws and policies that enshrine ethical principles, governments, providers, and designers must work together to address ethics and human rights concerns at every stage of an AI technology’s design, development, and deployment. 

To limit the risks and maximize the opportunities intrinsic to the use of AI for health, WHO provides the following principles as the basis for AI regulation and governance:

Protecting human autonomy: In the context of health care, this means that humans should remain in control of health-care systems and medical decisions; privacy and confidentiality should be protected, and patients must give valid informed consent through appropriate legal frameworks for data protection.

Promoting human well-being and safety and the public interest. The designers of AI technologies should satisfy regulatory requirements for safety, accuracy and efficacy for well-defined use cases or indications. Measures of quality control in practice and quality improvement in the use of AI must be available.

Ensuring transparency, explainability and intelligibility. Transparency requires that sufficient information be published or documented before the design or deployment of an AI technology. Such information must be easily accessible and facilitate meaningful public consultation and debate on how the technology is designed and how it should or should not be used.

Fostering responsibility and accountability. Although AI technologies perform specific tasks, it is the responsibility of stakeholders to ensure that they are used under appropriate conditions and by appropriately trained people. Effective mechanisms should be available for questioning and for redress for individuals and groups that are adversely affected by decisions based on algorithms.

Ensuring inclusiveness and equity. Inclusiveness requires that AI for health be designed to encourage the widest possible equitable use and access, irrespective of age, sex, gender, income, race, ethnicity, sexual orientation, ability or other characteristics protected under human rights codes.

Promoting AI that is responsive and sustainable. Designers, developers and users should continuously and transparently assess AI applications during actual use to determine whether AI responds adequately and appropriately to expectations and requirements. AI systems should also be designed to minimize their environmental consequences and increase energy efficiency. Governments and companies should address anticipated disruptions in the workplace, including training for health-care workers to adapt to the use of AI systems, and potential job losses due to use of automated systems.

“These principles will guide future WHO work to support efforts to ensure that the full potential of AI for healthcare and public health will be used for the benefits of all.”

The post WHO Issues First Global Report In Artificial Intelligence In Health appeared first on Artificial Intelligence.

Growing use of AI for health presents governments, providers, and communities with opportunities and challenges

Artificial Intelligence (AI) holds great promise for improving the delivery of healthcare and medicine worldwide, but only if ethics and human rights are put at the heart of its design, deployment, and use, according to new WHO guidance published today.

The report, Ethics and governance of artificial intelligence for health, is the result of 2 years of consultations held by a panel of international experts appointed by WHO.

“Like all new technology, artificial intelligence holds enormous potential for improving the health of millions of people around the world, but like all technology it can also be misused and cause harm,” said Dr Tedros Adhanom Ghebreyesus, WHO Director-General. “This important new report provides a valuable guide for countries on how to maximize the benefits of AI, while minimizing its risks and avoiding its pitfalls.”

Artificial intelligence can be, and in some wealthy countries is already being used to improve the speed and accuracy of diagnosis and screening for diseases; to assist with clinical care; strengthen health research and drug development, and support diverse public health interventions, such as disease surveillance, outbreak response, and health systems management.

AI could also empower patients to take greater control of their own health care and better understand their evolving needs. It could also enable resource-poor countries and rural communities, where patients often have restricted access to health-care workers or medical professionals, to bridge gaps in access to health services.

However, WHO’s new report cautions against overestimating the benefits of AI for health, especially when this occurs at the expense of core investments and strategies required to achieve universal health coverage.

It also points out that opportunities are linked to challenges and risks, including unethical collection and use of health data; biases encoded in algorithms, and risks of AI to patient safety, cybersecurity, and the environment.      

For example, while private and public sector investment in the development and deployment of AI is critical, the unregulated use of AI could subordinate the rights and interests of patients and communities to the powerful commercial interests of technology companies or the interests of governments in surveillance and social control.

The report also emphasizes that systems trained primarily on data collected from individuals in high-income countries may not perform well for individuals in low- and middle-income settings.

AI systems should therefore be carefully designed to reflect the diversity of socio-economic and health-care settings. They should be accompanied by training in digital skills, community engagement and awareness-raising, especially for millions of healthcare workers who will require digital literacy or retraining if their roles and functions are automated, and who must contend with machines that could challenge the decision-making and autonomy of providers and patients.

Ultimately, guided by existing laws and human rights obligations, and new laws and policies that enshrine ethical principles, governments, providers, and designers must work together to address ethics and human rights concerns at every stage of an AI technology’s design, development, and deployment. 

Six principles to ensure AI works for the public interest in all countries

To limit the risks and maximize the opportunities intrinsic to the use of AI for health, WHO provides the following principles as the basis for AI regulation and governance:

Protecting human autonomy: In the context of health care, this means that humans should remain in control of health-care systems and medical decisions; privacy and confidentiality should be protected, and patients must give valid informed consent through appropriate legal frameworks for data protection.

Promoting human well-being and safety and the public interest. The designers of AI technologies should satisfy regulatory requirements for safety, accuracy and efficacy for well-defined use cases or indications. Measures of quality control in practice and quality improvement in the use of AI must be available.

Ensuring transparency, explainability and intelligibility. Transparency requires that sufficient information be published or documented before the design or deployment of an AI technology. Such information must be easily accessible and facilitate meaningful public consultation and debate on how the technology is designed and how it should or should not be used.

Fostering responsibility and accountability. Although AI technologies perform specific tasks, it is the responsibility of stakeholders to ensure that they are used under appropriate conditions and by appropriately trained people. Effective mechanisms should be available for questioning and for redress for individuals and groups that are adversely affected by decisions based on algorithms.

Ensuring inclusiveness and equity. Inclusiveness requires that AI for health be designed to encourage the widest possible equitable use and access, irrespective of age, sex, gender, income, race, ethnicity, sexual orientation, ability or other characteristics protected under human rights codes.

Promoting AI that is responsive and sustainable. Designers, developers and users should continuously and transparently assess AI applications during actual use to determine whether AI responds adequately and appropriately to expectations and requirements. AI systems should also be designed to minimize their environmental consequences and increase energy efficiency. Governments and companies should address anticipated disruptions in the workplace, including training for health-care workers to adapt to the use of AI systems, and potential job losses due to use of automated systems.         

These principles will guide future WHO work to support efforts to ensure that the full potential of AI for healthcare and public health will be used for the benefits of all.

The post WHO issues first global report on Artificial Intelligence (AI) in health and six guiding principles for its design and use appeared first on Artificial Intelligence.

CommonSpirit Health, Providence, and Northwell Health are among the major health providers building a big data analytics platform to improve care.

Fourteen leading healthcare providers are partnering to form Truveta, a new company that will leverage big data analytics for enhanced care insights.

Providers involved in the effort include AdventHealth, Advocate Aurora Health, Baptist Health of Northeast Florida, Bon Secours Mercy Health, CommonSpirit Health, Hawaii Pacific Health, Henry Ford Health System, Memorial Hermann Health System, Northwell Health, Novant Health, Providence health system, Sentara Healthcare, Tenet Health, and Trinity Health.

Together, these healthcare providers care for tens of millions of patients and operate thousands of care facilities across 40 states. The health providers will govern Truveta’s ethical pursuit of insights from a de-identified dataset.

“The COVID-19 pandemic has shown us how much the world needs to learn faster, so we can better serve our communities,” said Terry Myerson, CEO of Truveta.

“Our vision is to save lives with data. We want to help researchers find cures faster, empower every clinician to be an expert, and help families make the most informed decisions on their care. We believe the Truveta platform can help improve health equity and advance personalized medicine. We are honored to be partnering with innovative and world-class health providers in this pursuit.”

Healthcare organizations today have datasets that are growing at exponential rates, but many don’t have the tools in place to extract meaningful insights from this information. The thoughtful use of big data in healthcare could help providers learn faster, move quicker, and improve patient outcomes.

Truveta will aim to build a new big data analytics platform by structuring, analyzing, and de-identifying data from participating health providers, while carefully protecting privacy and security. The platform will use AI and machine learning to enable unprecedented insights and collaborative learning among organizations.

“For years we have seen the opportunity for diverse health providers to come together with a shared sense of purpose and use our collective data for the common good of humanity. With Truveta, we created a unique model that is led by the health providers yet supported by one of the most talented technical teams to focus on health,” said Dr. Rod Hochman, President and CEO of Providence.

The new company will facilitate innovation in patient care through the creation of the Truveta data platform. The platform will deliver valuable insights from billions of critical data points with a single search, unlocking the power of de-identified data across all diagnoses, geographies, and demographics.

“We believe the cure for certain diseases could lie within the Truveta platform,” said Michael Slubowski, CEO of Trinity Health. “For the first time in the history of health, we have enough data at scale to dramatically advance innovation in healthcare with collective commitment to partner on ethical innovation.”

The current pandemic has demonstrated just how critical it is for the healthcare industry to move quickly in order to effectively serve patients. If an effort like Truveta had existed prior to the start of the pandemic, providers could have learned the best treatment paths from each other faster. Leaders hope that the formation of the company will pave the way for more collaborative care and education going forward.

“Since the start of the COVID-19 pandemic, we have seen displays of humanity and unity during our most trying moments. Together, we share a common mission to improve health equity and believe nobody should be left behind,” said Michael Dowling, President and CEO of Northwell Health. “Truveta is a catalyst for innovation, energizing health providers to modernize how we look at data to benefit patients.”

A Board of Governors will advise Truveta to ensure expertise is drawn from a variety of perspectives for strategic stewardship. Leaders from a diverse set of health providers will provide ongoing strategic, scientific, and operational advice on areas of expertise including ethics and health equity, data integrity, and clinical outcomes to make sure the company operates according to its mission.

“The future of healthcare is collaborative. We in healthcare exist side-by-side in our communities and we need to prioritize cooperation to truly make a difference—now more than we ever have,” said Lloyd Dean, CEO of CommonSpirit Health.

“We have a unique opportunity today to rebuild the healthcare system in our country, so it is better, stronger, and more responsive to the needs of everyone – especially the vulnerable and underserved populations.”

Truveta is inviting health providers, education, and research institutions anywhere in the world to join in creating and leveraging the Truveta platform to deliver the best care for patients.

“Our mission is to improve all people’s lives through excellence in the science and art of healthcare,” said Wright Lassiter III, President and CEO of Henry Ford Health System. “Truveta can uniquely provide the data and insights that will empower us to deliver equitable care with respect and compassion, which are the fundamental rights of those we serve.”

Truveta is committed to maintaining privacy and security, and all data on the platform will be de-identified.

“We know health data is unlike other data. It is the very definition of personal,” said Myerson.
“While we embark on our pursuit to generate knowledge and insights to improve patient care around the world, we must do so with the utmost caution to protect the privacy of patients.”

Healthcare providers expect that the Truveta platform will help enhance care delivery and patient outcomes.

“We see such a valuable opportunity to save lives in partnership with the Truveta platform,” said Alan Sanders, Vice President of Ethics, Trinity Health. “We believe it would be irresponsible to not join Truveta on this mission. It would be a tremendous data waste and withhold valuable contributions to the common good.”

The post Health Providers Partner to Create Big Data Analytics Platform appeared first on Artificial Intelligence.

Dr Nicole Robinson is a research fellow at Monash University working across two disciplines – engineering and medicine. Specifically, robotics and psychology.

She’s led a new research trial offering a glimpse into the near-future of digital healthcare. It found social robots – as helpers – can deliver a multi-session behaviour-change treatment around diet and weight reduction without the need for human intervention.

The results showed a robot-delivered program helped people to cut snack “episodes” by half, and lose, on average, 4.4 kilograms in weight. The trial also found that an autonomous robot-delivered program may be as effective as a human clinician in the same context.

She talks to Lens about social robots, her research, and what the future might – or might not – hold.

Where did this interest in robotics – and robots – begin for you?

I’ve been interested in technology for a very long time. Any time there was a new gadget or tool available, it was always something I wanted to get my hands on. The idea of a social robot really sparked my imagination – it comes from a childhood dream of having robots around and seeing them in a societal context.

But you started off studying people?

I started in behavioural science learning about people, how they behave, and the things they do. Then I started getting more interested in robots, and how robots can interact with people, what they can do, and how they should operate in society.

It was a fantastic crossroad between the two disciplines I was interested in. We can look to behavioural science to find new ways to make robots more effective with people.

What exactly is a social robot?

A social robot is a robot that can communicate and interact with people. The first ideas in robotics would have been that robots would eventually talk to people, interact with them and be able to support them in their daily life whether that is at home, in society or in the workplace.

But you’re looking at them specifically in terms of health and wellbeing interventions with individuals?

Yes. The behavioural intervention side of things is still relatively new. There’s been previous work in digital health and wellbeing programs using smartphones and web-based digital applications. We were interested to see what the utility of social robots might have in the space to support health and wellbeing.

The trial you led was about unhealthy, high-calorie snack foods and drinks. What did you find out?

We developed the robotic health coach using principles from evidence-based practice, but shaped to be delivered by a robot. In the trial, people who did complete the treatment program worked through the steps provided by the robot, and made improvements to their diet intake.

There was an average weight loss of 4.4kg, and people decreased their snack episodes by over 50% after having a conversation with the robot about their behaviour. We’ve found it was effective without the need for human intervention, demonstrating that robots could be important tools to help promote healthy daily behaviour.

Why do you think it worked well?

When we talked to people about what was important to them, people mentioned that the robot is non-judgemental. It was able to have a discussion with a person about a behaviour that is sensitive to them, and talk to them about the steps they could take to make changes. If there is a behaviour that a person is embarrassed about – such as eating a lot of sugary drinks or cans of soda – then it might be easier to talk openly if you don’t feel as if you will be judged.

When you can have that open and honest conversation, and participate in treatment and come up with a plan that works best for you based on your current situation, then that’s something that can help support you to make a change.

Is there a threshold in robotics around attachment? You want people to feel comfortable, but not too comfortable?

Some see robots as a way to complete the steps, create a plan, and make those changes in real life. Others have a much more relational focus in terms of the interactions they had with the robot, such as the robot was very friendly, and provided support and emotional warmth. Robots can appeal to both sides.

The program is designed to prompt self-reflection. The robot goes through a series of questions designed to get people thinking about how they might make a change.

There seems to be a threshold where you want to develop the robot to be personable and friendly enough to work with in a treatment session, but not so much that it misrepresents what it is or what it can do. If we had the robot saying, ‘I understand how you feel’, that would not resonate too well, because people know robots can’t feel, in that sense.

We need to make sure that we make careful scripting choices so the robot appears to be supportive, but not so much that it comes across as fake and sterile.

How attached do some people get? Do they name the robot?

Some people do. Some feel quite fond of social robots – this concept of anthropomorphism where people give inanimate objects a sense of personality, emotion and human-like quality. For some people that feeling can be really quite strong; they can get a real sense of connection and understanding. For others, they see it as a machine, no different from a screen or a tablet or an interactive kiosk. They have no additional feelings. It’s a very individual thing.

It sounds like that strange episode of Black Mirror where a small robot takes over a girl’s life.

That’s definitely science-fiction. The concept comes up a lot when you talk about robotic systems that take on some form of human-like qualities. We’re nowhere near that level of sophistication. Social robots can’t operate to that level of autonomy or intelligence. Robots do what we program them to do.

How might it work for a person who wanted to cut down on soft drinks as part of their healthcare?

The program is designed to prompt self-reflection. The robot goes through a series of questions designed to get people thinking about how they might make a change. The robot might say, ‘if you wanted to make a change, what would you need to do?’ The robot would then be prompted to ask them to think about how they would create a plan to see positive change in their lifestyle.

It’s about asking questions and getting people to think about a goal that would work for them. That’s another fascinating thing. We did not have detailed personalisation in the robot interaction, but laying out these steps and getting people to think out loud, come up with a goal, plan and a strategy that works best, they found that they did improve their dietary intake and achieve some weight loss.

What are the next steps with this technology?

Looking at it in closer detail. Can we apply the social robot coach to other daily health areas? This includes making the robot interaction more personalised to each person, which might help to increase health behaviour for people that may need extra support or encouragement to make a start.

Could it work in addiction?

It’s possible. There’s an increased risk, so it should be paired with clinician support. The robot would then be an additional feature or treatment access point.

Do you anticipate resistance from clinicians to a robot taking on some aspects of health and wellbeing treatments?

The idea of a robot does spark some different interpretations of what that means, and what it would be like in real life. If we were programming a robot to act exactly like a clinician it would not be appropriate, and mostly likely met with some hesitation.

If a robot can be programmed to provide support to treatment, robots then become another technology that can extend the reach of the clinician to provide greater support to both patients and practitioners.

The post Diet and weight loss: Using social robots to support human health and wellbeing appeared first on Artificial Intelligence.

An AI-driven health monitoring and disease detection platform was able to catch the signs of the Wuhan viral outbreak approximately a week before government agencies warned the public, providing a look at how AI can be used to catch disease outbreaks in a timely fashion.

While the official World Health Organization notification of the Wuhan virus went out on January ninth and the US Center for Disease Control and Prevention (CDC) received word of the outbreak on January sixth, the first warning signs of the outbreak were picked up by a Canadian health monitoring system almost a week prior. As Wired reported, the AI-driven health system BlueDot warned its clients about the possible outbreak on December 31st. Bluedot uses AI algorithms to monitor different global news sources and detect patterns in health reports. It also takes into account information on plant and animal disease networks. Using the information it collects, BlueDot epidemiologists then delivers warnings and predictions about possible health risks and outbreaks to its subscribers.

When dealing with an outbreak of disease, early detection is always better. The earlier the detection, the more time health officials have to respond. In the case of the Wuhan virus and other disease outbreaks in China, the Chinese government has often been slow in sharing information with global public health officials. This possesses a problem as the CDC and WHO rely on communications from other government agencies to plan their own responses. However, if an AI system like BlueDot can make accurate predictions based on the information that leaks through across many individual news reports, blogs, and forums, this could potentially enable health organizations to act quicker in response to outbreaks.

According to Kamran Khan, the found of BlueDot, the company doesn’t use social media data when predicting the spread of diseases because the data is too variable and messy to be of use. Instead, news reports, data on known animal disease networks, and airline ticketing data is combined to create a model that predicts where infections begin and where infected people may travel next. BlueDot was correctly able to predict that the Wuhan virus would spread to Taipei, Tokyo, Seoul, and Bangkok within a few days of its manifestation.

BlueDot was launched by Khan in 2014, and the company currently has 40 employees, including data scientists, physicians, and programmers who work together to create the disease surveillance and prediction models. Machine learning algorithms and natural language processing techniques are used to mine data from news reports spanning the globe and covering 65 different languages. Khan said to Wired:

“What we have done is use natural language processing and machine learning to train this engine to recognize whether this is an outbreak of anthrax in Mongolia versus a reunion of the heavy metal band Anthrax.”

After the automated data collection and initial analysis are complete, human analysts double-check the data and ensure that the model’s conclusions seem sound. Finally, a report is generated and sent out to the clients of the application.

BlueDot’s system is far from the first attempt by the AI field to predict the spread of diseases. Data scientists have been using big data and machine learning models to track the spread of various diseases like for some time now, with some attempts being more successful than others. Google tried its own hand at tracking the spread of disease with Google Flu Trends, but its attempts to predict the severity of the 2013 flu seasons were reportedly off by about 140%. Only time will tell if BlueDot can consistently predict the spread of diseases, but if it can it could pave the way for faster, more accurate estimates of disease outbreaks.

The post First Hints Of The Wuhan Virus Outbreak Were Caught By AI appeared first on Artificial Intelligence.

The work will include up to four projects and a new partnership that will integrate with Health Data Research UK’s (HDRUK) existing research programmes.

The Better Care Partnership was established to lead a new data science initiative, which could receive up to £1.2 million in funding.

Initially funded for three years, it will use continuous improvement methods to integrate clinical practice, large scale health data and advanced analytics to deliver insights for improving care for patients across the UK.

In addition, HDRUK will also support one year Catalyst Projects to demonstrate how patient care can be improved through data-driven health and care decision.

The organisation is teaming up with the Health Foundation to support the projects, which could each receive up to £200,000.

Professor Simon Ball, medical director at University Hospitals Birmingham NHS Foundation Trust and national lead for Health Data Research UK’s Better Care priority, said: “As healthcare professionals we make hundreds of decisions a week with our patients. In doing so we aim to decide what will work best for each individual.

“Electronic healthcare records offer the opportunity to combine patients’ data with information on best practice, so that we can reliably deliver high quality care in complex settings and pressured environments.

“Beyond that we can use the resulting data on patients’ outcomes and experience, to continuously learn from, and improve on, everyday practice in ways that are applicable across the NHS.”

Organisations putting forward partnership or project ideas will be expected to demonstrate how they plan to listen to patients and understand their wishes about how their health data will be used.

They will also be expected to show how patients will be involved at all stages. The aim is to use health data responsibly and ethically with a clear focus on improving patient care.

The UK has vast and rich data about people’s health and care, however this is often not available quickly for clinicians or patients to access to support their decision making, according to HDRUK.

This causes delays and, in some cases, prevents the data from being analysed to deliver better care and improve people’s health.  Both the Catalyst Projects and the Better Care Partnership, which will start in May 2020, aim to help address some of these issues.

The closing date for applying for the Catalyst and Better Care Partnership is 11 March, 2020 at 4pm.

The post Health Data Research UK launches £1m data science partnership appeared first on Artificial Intelligence.

Oil and gas pipelines have become polarizing issues in Canada, but supporters and detractors alike can agree that the safer they are, the better. With news emerging last month that the Keystone pipeline leak in North Dakota may have affected almost 10 times the amount of land previously reported, the issue of pipeline failure takes on an added urgency.

Unfortunately, integrity and health are ongoing and serious problems for North America’s pipeline infrastructure. According to the US Department of Transportation (DOT), there have been more than 10,000 pipeline failures in that country alone since 2002. Complicating safety measures are the cost and intensity of labour required to monitor the health of the thousands of kilometres of pipelines that criss-cross Canada and the United States.

In a recent paper in the Journal of Pipeline Systems Engineering and Practice, researchers at Concordia and the Hong Kong Polytechnic University look at the methodologies currently used by industry and academics to predict pipeline failure — and their limitations.

“In many of the existing codes and practices, the focus is on the consequences of what happens when something goes wrong,” says Fuzhan Nasiri, associate professor in the Department of Building, Civil and Environmental Engineering at the Gina Cody School of Engineering and Computer Science.

“Whenever there is a failure, investigators look at the pipeline’s design criteria. But they often ignore the operational aspects and how pipelines can be maintained in order to minimize risks.”

Nasiri, who runs the Sustainable Energy and Infrastructure Systems Engineering Lab, co-authored the paper with his PhD student Kimiya Zakikhani and Hong Kong Polytechnic professor Tarek Zayed.

Safeguarding against corrosion

The researchers identified five failure types: mechanical, the result of design, material or construction defects; operational, due to errors and malfunctions; natural hazard, such as earthquakes, erosion, frost or lightning; third-party, meaning damage inflicted either accidentally or intentionally by a person or group; and corrosion, the deterioration of the pipeline metal due to environmental effects on pipe materials and acidity of oil and gas impurities. This last one is the most common and the most straightforward to mitigate.

Nasiri and his colleagues found that the existing academic literature and industry practices around pipeline failures need to further evolve around available maintenance data. They believe the massive amounts of pipeline failure data available via the DOT’s Pipeline and Hazardous Materials Safety Administration can be used in the assessment process as a complement to manual in-line inspections.

These predictive models, based on decades’ worth of data covering everything from pipeline diameter to metal thickness, pressure, average temperature change, location and timing of failure, could provide failure patterns. These could be used to streamline the overall safety assessment process and reduce costs significantly.

“We can identify trends and patterns based on what has happened in the past,” Nasiri says. “And you could assume that these patterns could be followed in the future, but need certain adjustments with respect to climate and operational conditions. It would be a chance-based model: given variables such as location and operational parameters as well as expected climatic characteristics, we could predict the overall chance of corrosion over a set time span.”

He adds that these models would ideally be consistent and industry-wide, and so transferrable in the event of pipeline ownership change — and that research like his could influence industry practices.

“Failure prediction models developed based on reliability theory should be realistic. Using historical data (with adjustments) gets you closer to what actually happens in reality,” he says.

“They can close the gap of expectations, so both planners and operators can have a better idea of what they could see over the lifespan of their structure.”

The post Concordia researcher hopes to use big data to make pipelines safer appeared first on Artificial Intelligence.

Artificial intelligence is on a par with human experts when it comes to making medical diagnoses based on images, a review has found.

The potential for artificial intelligence in healthcare has caused excitement, with advocates saying it will ease the strain on resources, free up time for doctor-patient interactions and even aid the development of tailored treatment. Last month the government announced £250m of funding for a new NHS artificial intelligence laboratory.

However, experts have warned the latest findings are based on a small number of studies, since the field is littered with poor-quality research.

One burgeoning application is the use of AI in interpreting medical images – a field that relies on deep learning, a sophisticated form of machine learning in which a series of labelled images are fed into algorithms that pick out features within them and learn how to classify similar images. This approach has shown promise in diagnosis of diseases from cancers to eye conditions.

However questions remain about how such deep learning systems measure up to human skills. Now researchers say they have conducted the first comprehensive review of published studies on the issue, and found humans and machines are on a par.

Prof Alastair Denniston, at the University Hospitals Birmingham NHS foundation trust and a co-author of the study, said the results were encouraging but the study was a reality check for some of the hype about AI.

Dr Xiaoxuan Liu, the lead author of the study and from the same NHS trust, agreed. “There are a lot of headlines about AI outperforming humans, but our message is that it can at best be equivalent,” she said.

Writing in the Lancet Digital Health, Denniston, Liu and colleagues reported how they focused on research papers published since 2012 – a pivotal year for deep learning.

An initial search turned up more than 20,000 relevant studies. However, only 14 studies – all based on human disease – reported good quality data, tested the deep learning system with images from a separate dataset to the one used to train it, and showed the same images to human experts.

The team pooled the most promising results from within each of the 14 studies to reveal that deep learning systems correctly detected a disease state 87% of the time – compared with 86% for healthcare professionals – and correctly gave the all-clear 93% of the time, compared with 91% for human experts.

However, the healthcare professionals in these scenarios were not given additional patient information they would have in the real world which could steer their diagnosis.

Prof David Spiegelhalter, the chair of the Winton centre for risk and evidence communication at the University of Cambridge, said the field was awash with poor research.

“This excellent review demonstrates that the massive hype over AI in medicine obscures the lamentable quality of almost all evaluation studies,” he said. “Deep learning can be a powerful and impressive technique, but clinicians and commissioners should be asking the crucial question: what does it actually add to clinical practice?”

However, Denniston remained optimistic about the potential of AI in healthcare, saying such deep learning systems could act as a diagnostic tool and help tackle the backlog of scans and images. What’s more, said Liu, they could prove useful in places which lack experts to interpret images.

Liu said it would be important to use deep learning systems in clinical trials to assess whether patient outcomes improved compared with current practices.

Dr Raj Jena, an oncologist at Addenbrooke’s hospital in Cambridge who was not involved in the study, said deep learning systems would be important in the future, but stressed they needed robust real-world testing. He also said it was important to understand why such systems sometimes make the wrong assessment.

“If you are a deep learning algorithm, when you fail you can often fail in a very unpredictable and spectacular way,” he said.

Since you’re here…

… we have a small favour to ask. More people are reading and supporting The Guardian’s independent, investigative journalism than ever before. And unlike many new organisations, we have chosen an approach that allows us to keep our journalism accessible to all, regardless of where they live or what they can afford. But we need your ongoing support to keep working as we do.

The Guardian will engage with the most critical issues of our time – from the escalating climate catastrophe to widespread inequality to the influence of big tech on our lives. At a time when factual information is a necessity, we believe that each of us, around the world, deserves access to accurate reporting with integrity at its heart.

Our editorial independence means we set our own agenda and voice our own opinions. Guardian journalism is free from commercial and political bias and not influenced by billionaire owners or shareholders. This means we can give a voice to those less heard, explore where others turn away, and rigorously challenge those in power.

We need your support to keep delivering quality journalism, to maintain our openness and to protect our precious independence. Every reader contribution, big or small, is so valuable.

The post AI equal with human experts in medical diagnosis, study finds appeared first on Artificial Intelligence.

The Pittsburgh Health Data Alliance (PHDA) has announced a machine learning research sponsorship from Amazon Web Services (AWS) that would see the alliance aim to advance innovation in areas such as cancer diagnostics, precision medicine, voice-enabled technologies, and medical imaging.

The PHDA is a consortium formed by Pittsburgh’s UPMC hospital, the University of Pittsburgh, and Carnegie Mellon University.

The alliance focuses on the data generated in healthcare, such as patient information in electronic health records, diagnostic imaging, prescriptions, genomic profiles, and insurance records, and is charged with using that data to transform the way diseases are treated and prevented, as well as to better engage patients in their own care.

The PHDA said new machine learning technologies and advances in computing power, such as Amazon SageMaker and Amazon EC2, are making it possible to “rapidly translate insights discovered in the lab into treatments and services that could dramatically improve human health”.   

PHDA scientists from both universities are expected to accelerate research and product commercialization efforts across eight projects through the AWS sponsorship, such as those with the potential to create an individual risk score for every cancer patient.

The ability to create an individual risk score for cancer patients, the alliance said, could enable doctors to better predict the course of a person’s disease and response to treatment.

Other projects the PHDA will undertake as part of the AWS deal include the use of a patient’s verbal and visual cues to diagnose and treat mental health symptoms, and reduce medical diagnostic errors by mining all the data in a patient’s medical record.

Associate dean for research at Pitt’s Swanson School of Engineering and the John A. Swanson Professor of Bioengineering David Vorp and his team are using AWS resources to improve the diagnosis and treatment of abdominal aortic aneurysms, which the alliance said is the 13th-leading cause of death in western countries.

“Currently, clinicians can use only the simple measurements of an aneurysm’s diameter and growth rate to predict the risk of a rupture,” the PHDA said.

“With the latest advances in machine learning, we are developing an algorithm that will provide clinicians with an objective, predictive tool to guide surgical interventions before symptoms appear, improving patient outcomes,” Vorp added.

Similarly, a CMU team led by professor of biological sciences and computational biology Russell Schwartz and Jian Ma will use AWS support to develop algorithms and software tools to better understand the origin and evolution of tumor cells.

According to the PHDA, the CMU project will use machine learning to gain insights into how tumors develop and to predict how they are likely to change and grow in the future.

“Data-driven, genomic methods guided by an understanding of cancers as evolutionary systems have relevance to numerous aspects of clinical cancer care,” Schwartz explained.

“These include determining which precancerous lesions are likely to become cancers, which cancers have a good or bad prognosis, and which of those with bad prognoses might respond long-term to specific therapies.”

“We believe that machine learning can significantly accelerate the progress of medical research and help translate those advances into treatments and improved experiences for patients,” AWS vice president of machine learning Swami Sivasubramanian added. 

“We are excited to bring our machine learning services and cloud computing resources to support the high-impact work being done at the PHDA.”

UPMC Enterprises, which funds the PHDA and focuses on commercializing its research, expects AWS machine learning and artificial intelligence services will help Pittsburgh become the premier hub of technology innovation in healthcare.

The post Pittsburgh Health Data Alliance looks to machine learning for patient care appeared first on Artificial Intelligence.

Few areas of the corporate world are fraught with the conflicting objectives found in employee health. Company health plans are designed to maintain employee health. Healthy employees are more productive. And generous health coverage bolsters recruitment and retention, a key goal in the zero-employment economy.

But benefits are also a nagging cost center. And many employers are uncertain about the legality of analyzing the health data their plans create.

How to balance these often conflicting priorities?

Enter the latest potential solution: data mining platforms and consultants.

Data warehouses are nothing new. That’s where employers first turned when they needed a black hole in space to store the enormous bits of data generated every minute of the work day. But now, as the Big Data industry rushes ahead, driven by its own data, employers suddenly have myriad options for managing and mining those bits stashed away on the cloud. The question is: How do I get the answers I need from my data in a timely fashion with actionable recommendations? Oh, and without running afoul of privacy concerns?

That’s where vendors like Springbuk, Segal Group, Artemis Health and others come in. Their promise to employers: We’ll help you quickly find out what you’re looking for in your data. And we will also bring to your attention trends and issues you didn’t know existed that can generate a better return on your health plan investment.

Employers are signing on for these services despite concerns about just how deeply they can mine health data. A recent Accenture survey found that only 30 percent of respondents were “very confident that they are using new sources of workforce data in a highly responsible way.” But 62 percent said they were already “using new technologies and sources of workforce data today,” and three-quarters were eager to analyze their employee data to grow and transform their businesses, and to unlock their employees’ full potential.

And apart from legal protections around privacy, which remain uncertain, workers don’t seem to be fearful of the Big Data/Big Brother syndrome. More than 90 percent of employees responding to the survey said they were fine with collection of personal data, as long as it “improves their performance or well-being or provides other personal benefits.”

Connecting with employers

Platform builders are finding two routes to employers: Directly to them through their sales teams, and through broker channels. Says Springbuk’s Reasen, “We do believe in the value proposition of the broker model. They represent a strong advocacy at the local level that still exists. They like to be able to offer a tool like ours to get into data warehouses and analyze what’s there for their clients.”

Springbuk just unveiled an upgrade of its health intelligence platform that its executives say will both greatly reduce the time required to mine specific intelligence from health data, and provide clients with customized, curated data-based reports on topics ranging from risk mitigation, care efficiency and drug savings, to steerage procedures and potentially unnecessary procedures.

The upgrade further enhances the platform’s ability to identify members within a plan population “who are at risk of developing health conditions and then get actionable information including appropriate treatment, disease management resources and risk mitigation strategies. At-risk employees are identified based on a proprietary algorithm using a database of existing claims,” the company says.

In other words, the platform both responds rapidly to employer queries about employee health, and anticipates, explores, and issues reports on trends that clients may not be aware of.

“The message around health intelligence, as opposed to health data, is resonating with employers,” says Springbuk’s Rod Reasen, CEO and co-founder. “I just got off a call with a very large organization that represents hundreds of thousands of lives. They want to know why they bought a data warehouse. ‘We thought we’d have access to a lot o f information. but actually it’s just access to a lot of data.’ Our health intelligence platform goes beyond a data warehouse to provide actually actionable intelligence.”

“It’s a question of data mining versus data reporting,” says Segal’s David Searles, vice president and the executive who developed Segal’s data analytics business. “Data mining creates new information from the data. The mining can say, for example, that you have 20 percent of your diabetics who aren’t getting their tests done. It is creating new actionable information from the data you are presented with.”

Data-driven decisions

These new platforms can swiftly adapt to shifting priorities among employees. As opioid abuse continues to take a toll on employee health and the cost of insurance, Segal was asked to examine one client’s population to identify total savings potential for opioid abuse prevention management.

“The client wanted us to quantify enhanced opioid criteria savings to medical and prescription drug programs,” he says. “We analyzed the data and discovered that, by limiting first fills of opioid prescriptions to a 7-day supply, ER-related opioid visits decreased 35.3 percent.”

Demand is strong to mine employee data to evaluate workplace wellness programs. Generally, employers want to reduce their wellness offerings to those programs that engage employees and produce better health outcomes.

“We get a lot of requests to examine the data for return on investments in various programs, both wellness or disease management. While you can’t really do an ROI accurately–no one agrees on a consistent methodology for it–you can determine the effectiveness of the program by looking at the change in biometric data of the participants,” Searles says.

“What we are pushing toward is this: Plan sponsors should use data to actively manage their health plans. They should evaluate their employee profiles. Let’s target a program that addresses conditions that are driving trends.”

One example would be designing a treatment plan for diabetics with coronary disease. “They should be highly motivated and will incur large claims if they don’t improve their condition,” he says.

But first, the company needs to know whether its workforce includes enough diabetics with coronary disease to justify creating such a program. And that’s where the emerging data mining platforms shine.

The new mining platforms have immeasurably reduce the time required for an employer to find the desired information. Springbuk’s Reasen says the chief medical officer for one client told him “it would have taken him a month to come up with the exact [report] we came up with in seconds.”

He adds: “When a user steps in front of a data warehouse, we are asking them to spend time and use their knowledge to extract information. We all have the same amount of time. How do we use it?”

Strategic benefits firm Sequoia Consulting Group is a Springbuk broker client. CEO Greg Golub says his clients especially value the executive reports the platform produces.

“Springbuk is very effective at producing executive reports, tailored for the CFO or HR leader. They do a good job of synthesizing the information into an actionable report. They are focusing on the right stuff.”

Sequoia’s chief marketing officer, Michele Floriani, says being able to offer Springbuk reports to clients has led to positive feedback. “We offer it as a service to our self-insured clients and together we make use of the output and insights to make annual and longer term strategy decisions on plan design. It’s really wonderful. That’s the value to us. It focuses on what matters to the client.”

The post Digging deeper: Health data mining platforms surge ahead appeared first on Artificial Intelligence.