NHSX reported it had enleagued Microsoft, Palantir and Google to build a datastore specific to the Covid-19 crisis response at the end of March. Amazon Web Services (AWS) was also said, at the time, to be about to become a participant in the data analytics programme, and has become one since.

Civil liberties organisations Privacy International, Big Brother Watch, medConfidential, Foxglove and the Open Rights Group have been vocal in their objections to the US companies’ involvement in the NHS database. Silicon Valley data mining company Palantir, controversial in the US because of its proximity to the FBI and the CIA, has fallen under particular scrutiny from Privacy International and its partners.

OpenDemocracy, a media organisation, had been about to initiate legal proceedings against the government to have the contracts with Amazon, Microsoft, Google, Palantir and UK artificial intelligence (AI) firm Faculty published. Before they could do so, the government today handed the contracts over to OpenDemocracy and law firm Foxglove.

OpenDemocracy has now published the contracts for Google, Faculty, Palantir and Microsoft on its website. Faculty and Palantir are in its sights especially.

The organisation stated that: “The contracts show that companies involved in the NHS datastore project, including Faculty and Palantir, were originally granted intellectual property rights (including the creation of databases), and were allowed to train their models and profit off their unprecedented access to NHS data.

“Government lawyers have now claimed that a subsequent (undisclosed) amendment to the contract with Faculty has cured this problem, however they have not released the further contract. OpenDemocracy and Foxglove are demanding its immediate release.”

Meanwhile, Foxglove stated on its site: “We’re still analysing the documents, but at first sight it looks like they’ve backed down significantly and disclosed a significant amount of information. It also appears that our pressure may have already forced them to introduce new safeguards to stop private companies profiteering from NHS data.

“Now we need to assess whether the new safeguards go far enough. And this will all inform an overdue public debate: do we want big tech companies, Faculty and Palantir bedding down with our NHS for the long haul?”

In its statement, OpenDemocracy underlined Palantir’s role in abetting the CIA’s intelligence operations in Afghanistan and Iraq. And in supporting the US Immigration and Customs Enforcement Agency.

Palantir’s CEO, Alex Karp, is an avowed socialist, but civil liberties organisations prefer to draw attention to founder Peter Thiel’s alleged support for US president Donald Trump, as OpenDemocracy does in its statement about the release of the contracts.

According to the contract, Palantir Technologies UK will be paid £1 for the data integration work it is doing on the NHS Covid-19 datastore, through its Foundry technology. The NHS – the buyer is listed as NHS Arden & GEM CSU – will also pay Amazon Web Services on a monthly basis for the duration of the project since the Foundry software is hosted on AWS.

Palantir’s services include “ingestion of mutually agreed data sources and further integration into a data ontology”, configuration of Foundry and user training.

According to the government contract with Faculty, the latter will provide “strategic support to the NHSX AI Lab”. OpenDemocracy emphasises that the company is run by Mark Warner, brother of Ben Warner, who ran the data operation for the Dominic Cummings-led Vote Leave campaign.

The Faculty contract has a value of £930,000.

Faculty’s role is, in part, to provide “short- and long-term forecasts about the spread of Covid-19 and the impact that has on resources across the healthcare system. The short-term forecasts will be generated through extrapolations of data that exists, and long-term forecasts will be generated using simulation… The Faculty team will be collaborating with an epidemiological research group, at Oxford BDI [Big Data Institute], to develop their simulation tool to meet the objectives of the Covid response group.”

Faculty is also working with Palantir to create a dashboard for NHS decision-makers at national, country, regional trust and hospital levels.

The post NHS Covid-19 datastore contracts published under pressure from privacy groups appeared first on Artificial Intelligence.

Civil liberties campaigning group Privacy International is questioning the role of Silicon Valley data mining company Palantir in the NHS’s data analysis programme.

NHSX and NHS England Improvement reported at the end of March that they had engaged Palantir alongside Microsoft, Google and London-based artificial intelligence (AI) firm Faculty to build a “data platform” to make the anti-Covid-19 response as efficient and effective as possible.

Data will be gathered from sources including 111 calls and Covid-19 test results, according to a BBC report by Leo Kelion, technology desk editor.

Privacy International, Big Brother Watch, medConfidential, Foxglove and Open Rights Group have sent Palantir 10 questions about their work with the NHS during the public health crisis, and published them on the PI website.

The four privacy campaigning groups say they are “primarily interested in if and how Palantir will retain and use the data analysis gleaned from this work with the NHS”, adding: “We are interested in whether the company will use this analysis to strengthen its own proprietary systems.”

Among the 10 questions are: “Is Palantir obtaining access to any databases and/or records held by the NHS, such as online prescription systems, patient records, general practitioners’ files, etc?”, “Will Palantir retain the NHS data analysis or insights gleaned from this contract once this exercise is over?” and “Will Palantir be able to use the product trained under the agreement with NHS to improve other future products provided by Palantir?”

Privacy International said: “As Palantir says, in the spirit of ‘open and critical discussion’, we have asked them for key details about their current work with the NHS. It would be misleading and cynical for Palantir to offer services to the NHS without being fully transparent about how the company may benefit from the data analysis gleaned in this work, which they can then go on to profit from and strengthen their proprietary systems.”

Silkie Carlo, director of Big Brother Watch, said in the Privacy International statement: “It is unacceptable that a large-scale project involving patient data is being pursued with Palantir in absence of stakeholder engagement or public transparency. Palantir and NHSX must be fully open and transparent about the ‘Covid-19 datastore’, the nature of contracts, the use of patient data, the confidentiality of 111 calls, and make details of any predictive analytics and anonymisation techniques available for public audit at the soonest possibility.” 

Palantir Technologies UK is providing the NHS project with software, Palantir Foundry, that makes the front-end data platform to a data store specific to the current pandemic. Palantir Foundry is said to enable disparate data to be cleansed and integrated.

Palantir is controversial because it has been involved in such activities as the US Immigration and Customs Enforcement programme. Some of its own employees protested against this in 2019.

The Palo Alto-based firm was co-founded by leading Silicon Valley venture capitalist Peter Thiel in 2003. And although its co-founder and CEO, Alex Karp, is a self-described socialist, the company’s client base in the CIA and FBI has made it controversial among civil libertarians.

Its name is derived from the crystal ball palantirs in Tolkien’s Lord of the Rings, one of which is wielded by Sauron, the dark lord of the fantasy tale.

The post Privacy International puts Palantir in the dock for NHS data analysis work appeared first on Artificial Intelligence.

On 8^th August 2019, Health Secretary Matt Hancock allocated an additional £250 million to be invested in an artificial intelligence (AI) laboratory that will lead to a better ability to screen for cancer, identify patients most at risk of diseases such as heart disease or dementia, build systems to detect people at risk of post-operative infections and more.

On the face of it, this announcement looks like very good news. Hancock commented: “We are on the cusp of a huge health tech revolution that could transform patient experience by making the NHS a truly predictive, preventive and personalised health and care service.” He also emphasised his determination to: “bring the benefits of technology to patients and staff, so the impact of our NHS Long Term Plan and this immediate, multimillion-pound cash injection are felt by all.”

Whilst some examples of the potential uses of AI have been given both in administrative and clinical contexts (which also include predicting patients most likely not to show for appointment and inspecting existing algorithms already used by the NHS to ensure patient confidentiality is protected), there are many other examples of applications of AI to support patient care. For example, the monitoring of Type 1 diabetic patients and of those that have heart conditions via body-worn devices can bring about transformational improvements in the individual’s health and also reductions in the cost to the taxpayer.

With higher patient expectations and increases in life expectancy, a growing number of citizens require pre-emptive advice to promote better health. Leveraging the insight trapped in the UK population’s medical data can make the difference. But with more data and complexity than ever, unlocking this insight is becoming increasingly difficult. Consequently, opportunities for preventive measures and the most efficient corrective care are not always being taken.

So how could AI in collaboration with other technologies improve the NHS?

AI with end-to-end process automation to improve preventative healthcare

As life expectancy rises and pressure on NHS resources grows, investing in ways to educate citizens with pre-emptive advice to staying healthy is growing in importance. To succeed, businesses need an easy, accurate and reliable way to create and incorporate predictive analytics and decisions into every process and interaction. Coupled with other leading technologies, such as interactive business process management, robotic automation and context-sensitive transparent guidance and decisions, AI should bring about both improvements in patient care at the same time as similar enhancements in operational efficiency.

AI with analytics to make cost savings and improve efficiencies

The use of AI and analytics can inform on trends on overtime and temporary staff working patterns, plus identification of likely increases in demand to help set the right numbers of doctors and nurses along with other infrastructure provision. Integration of data from different sources within the NHS and agencies outside of it could also inform where different supply options for beds provide the best value for money.

AI with external source data to inform policy

AI could be used to bring together disparate data sources to indicate the best options for how to improve patient care. For example, with sufficient online information to notify on likely shortfalls in public sector rehabilitation beds, private-sector resources could be taken advantage of more to take the strain of the public-sector. The NHS could analyse the particular needs of a patient based on case history, clinical guidance and rules with AI to decide on the best course of action.

Furthermore, data from social welfare can be used to inform policy at both a macro and local level, as health- care and social welfare are so inextricably linked – what happens in one domain often gives rise to demands on the other. The delivery of social welfare by local government versus centralised provision of healthcare has previously caused issues, so orchestration of inter-agency sharing of information is imperative.

The government says AI is already being developed in some hospitals, successfully predicting cancer survival rates and cutting the number of missed appointments. It is motivating that this technology is already saving lives, however, it is clear AI alone will not result in the desired outcomes. It needs to be part of the greater plan by also taking into account technologies such as digital process automation, smart use of data and a focus on patients to ensure this investment delivers on its promises.

The post NHS: Why AI investment is just one piece of the puzzle 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.