Connect with us

Spread & Containment

CAREER Award: Teaching machines the art of human decision-making

Children attend pre-school and kindergarten to develop the skills they need to thrive. They practice sharing and making friends. They learn to use words…

Published

on

Children attend pre-school and kindergarten to develop the skills they need to thrive. They practice sharing and making friends. They learn to use words and numbers. They gain confidence through movement and self-control. Teachers have an array of techniques to help children develop these skills.

Credit: Tom Cogill

Children attend pre-school and kindergarten to develop the skills they need to thrive. They practice sharing and making friends. They learn to use words and numbers. They gain confidence through movement and self-control. Teachers have an array of techniques to help children develop these skills.

But how should they help students who struggle? There’s no shortage of opinions from educators, parents and policy-makers about effective ways to teach children, based on observations of what happens in classrooms and children’s behavior in and out of school. These observations may be missing something important, however – factors that are unseen, unrecognized or unreported but are powerful influences on a student’s performance.

Jundong Li, University of Virginia assistant professor of electrical and computer engineering, computer science and data science, is conducting research that could help teachers and administrators more accurately determine which learning methods are best for their youngest pupils.

Li has earned a prestigious National Science Foundation CAREER award to better understand cause and effect in human decision-making in the era of big data. Li will use his $600,000 five-year award to develop a suite of sophisticated algorithms and mathematical models, informed by human experience and intuition, to find cause-and-effect relationships in a huge amount of data. His work has the potential for broad applications in public health and medicine in addition to education.

The CAREER program, one of the NSF’s most prestigious awards for early-career faculty, recognizes the recipient’s potential for leadership in research and education. Li’s award recognizes his expertise in data mining, machine learning, and artificial intelligence, which are part of a research strength area for the Charles L. Brown Department of Electrical and Computer Engineering within UVA’s School of Engineering and Applied Science.

“The basic problem here is that machine learning and data mining alone are often insufficient to make decisions for humans,” Li said. “Typically, given a large amount of data, machine learning models can find correlations and then use those correlations to make inferences and predict outcomes.”

Because machines cannot really understand human needs, expectations and behaviors, their predictions and recommendations may be based on spurious correlations.

“We all know that correlation does not necessarily imply causation,” Li said. “In order to make a decision, typically we need to have a better understanding of what is cause and what is outcome. We want to find causal relations between variables at play.” This means creating what Li calls a causal inference model, which quantifies the strength of cause-and-effect relationships between different variables and uses the strongest to make a decision. 

Nowadays, research to make machine learning algorithms and models better at reasoning is largely data-driven, Li said. “For my CAREER award project, I want to incorporate prior human knowledge into these algorithms, to give the model the benefit of human wisdom as it processes data and interprets decision-making scenarios.”

Li has had preliminary success in his proposed approach, working in the public health arena supported by a RAPID grant from the UVA’s Global Infectious Diseases Institute. Li collaborated with Daniel Mietchen, formerly with UVA’s School of Data Science and now a researcher at the Fraunhofer Institute for Biomedical Engineering in Germany, to assess the impact of COVID-19 related policies on outbreaks. Three members of Li’s research group assisted with the study.

The team’s model shows how COVID-19 policies such as social distancing affected outbreaks at the county level, taking into account people’s vigilance over the virus over time. 

A county government may issue policies to enforce social distancing at an early stage of the pandemic, but if residents in the county tend to be more alert to COVID-19, they likely would have a lower probability of infection. In this case, vigilance is a confounding variable, influencing both the “treatment,” or the policy of social distancing, and the outcome, or the number of individuals who get sick.

Publicly available information online provided an important resource. For example, the team used the popularity of Google searches about COVID-19 at different time periods as a measure of residents’ vigilance. Using this indicator and others, the team developed a framework that captures information from different time periods and handles information among counties to estimate how various policies affected COVID-19 outbreaks. The framework shows the cause and effect of policies at different degrees of specificity, from a category of policies with a certain goal, to a single policy.

The team members presented the results of their study in a research paper, Assessing the Causal Impact of COVID-19 Related Policies on Outbreak Dynamics: A Case Study in the US, published in the proceedings of the Association for Computing Machinery’s Web Conference 2022 in April.

“Our web conference paper captures outbreak dynamics more accurately than statistical methods alone,” Li said. “Additionally, our assessment of policies is more consistent with existing epidemiological studies of COVID-19. This suggests that public health officials can use our framework when randomized controlled trials, the gold standard of cause-and-effect estimation, are not feasible.”

Li’s next step involves collaboration with individuals who are knowledgeable about the application areas, such as medical doctors, public health officials and experts in learning and development. He will also identify publicly available data in the areas of health and education that he can mine to further test and develop his decision-making framework.

Ultimately, Li envisions developing sophisticated algorithms that will pinpoint cause and effect, so physicians can use them to customize treatments based on patient information, and decision-makers can plug the algorithms into their own data systems to deliver policies that improve their constituents’ health, economic well-being and quality of life.


Read More

Continue Reading

Spread & Containment

Joined up thinking needed for joined up data plans

Joined up data could transform the pharmaceutical industry and help create a healthier Europe for decades to come
The post Joined up thinking needed for…

Published

on

Joined up data could transform the pharmaceutical industry and help create a healthier Europe for decades to come – but the route to change is far from smooth sailing.

Without careful consideration and full stakeholder input, the EU’s plans for a connected data system could end up being counterproductive.

That’s the view of the European Federation of Pharmaceutical Industries and Associations (EFPIA), which has published a list of recommendations aimed at helping the sector get the most of out of the data it holds.

“If the European Health Data Space (EHDS) and the rules surrounding access to the data are not carefully thought through, with the involvement of all stakeholders, there could be unintended consequences that limit the utility of the data for developing innovative medicines,” said the organisation.

Huge potential

The EFPIA Recommendations on a Connected Data System in Europe, published at the end of April, welcomes the proposals, which are part of the European Strategy for Data, to create common data spaces.

Said the authors: “A connected health data ecosystem has the potential to empower more effective and efficient research and development of new treatments and diagnostics. It would also ensure better planning and delivery of patient-centred care through personalised medicine.

“This, combined with value-based healthcare, can result in better allocation of resources and more sustainable healthcare systems.”

The value of this approach, which places real-world data in the hands of the right people at the right time, was demonstrated in abundance over the last few years, they went on.

It was, they explained, stakeholders from across the healthcare ecosystem coming together to share insights, whether from clinic, research, or genomics, that changed the course of the COVID-19 pandemic.

Applying the same ethos to healthcare in general, then, could give the drug development sector all the information it needs to contribute to a fitter, healthier Europe.

“For the research-based industry, access to data is critical at every step. From accelerating drug discovery to understanding patients’ behaviours and the outcome of treatment, the availability of data is essential to testing hypotheses, identifying trends and assessing proposed treatments,” they said, adding that improved access to, and transmission of, health data could “transform the pharmaceutical industry”.

“A connected health data ecosystem has the potential to empower more effective and efficient research and development of new treatments and diagnostics. It would also ensure better planning and delivery of patient-centred care through personalised medicine.”

 

Significant challenges

While EHDS is a lofty ambition, bringing it to fruition will not be without its challenges, both practical and regulatory.

As the EFPIA paper points out, health data is currently held in a wide range of repositories, from clinical notes and electronic health records to insurance claims, patient registries, patient-reported outcomes records, and continuous patient monitoring data from apps and wearables.

Unlocking their value, then, requires a high level of interoperability between different IT systems, providers, data sources, and software, all based in different countries with different levels of infrastructure maturity.

“Healthcare system information must be better connected. This will allow stakeholders to use this data for optimising and improving health outcomes,” said the paper, adding that interoperability was a “critical enabler of the digital transformation of healthcare in Europe”.

Conflicting national laws could be another important barrier to data access and use. Varying interpretations of the General Data Protection Regulation (GDPR), for example, present challenges for clinical development of innovative medicines, said the authors.

“Conflicting interpretations of Article 9 of the GDPR, and the additional limitations on processing of health and genomic data that member states have enacted under this article, cause significant delays in study start-up and patient enrolment.

“Some member states take the position that the only lawful basis for processing health data is when individuals have given their consent for its collection and use. Others… take the position that processing this health data, when necessary for scientific research, is lawful.”

EU Data Protection Supervisors, the paper recommends, must reach a common understanding of key GDPR terms if citizens are to enjoy the same rights across the EU.

Practical solutions

The EFPIA paper makes a number of recommendations on how the EU could embrace the full potential of the proposed EHDS.

First, it says that developing a shared understanding of the relevant requirements in digital health is essential, and calls for an EU-wide approach to how data is accessed, pooled, compared and used, while also protecting privacy.

In terms of possible solutions, it points to the use of Federated Data Networks (FDN), in which separate networks share mutual RWD resources.

“In an FDN, data is not moved from its host source, though hybrid models can exist with local and central data hosting. The research question or query moves to where the data is originally hosted, with results aggregated centrally or delivered to the researcher,” said the authors.

This, they went on, could unlock the power of data in primary or secondary care settings, in clinical care decision-making, and in research, whilst preserving the privacy of the RWD at a local level.

Common data models (CDM), which standardise the logical infrastructure of software systems to enable interoperability, are also required.

“CDM is essentially a construct, a means to an end to help organise RWD into a common structure, formats, and terminologies across diverse, heterogeneous, and multiple source datasets,” said the paper.

“It addresses a central need to be able to curate data for analysis on a contemporaneous and continuous basis (not on a per study basis) or for largescale, geographically diverse, network studies of multiple data sources.”

 Joined up approach to joined up data

Ultimately, building a usable EU-wide health data system requires input from all stakeholders, and decisions on FDNs and CDMs should be taken internationally, as a sector.

Because, as the EFPIA says, we all have one goal: using the power of data to improve the health of the citizens of Europe.

About the author

Amanda Barrell is a freelance health and medical education journalist, editor, and copywriter. She has worked on projects for pharma, charities and agencies, and has written extensively for patients, HCPs and the public.

The post Joined up thinking needed for joined up data plans appeared first on .

Read More

Continue Reading

Government

Weekly investment update – The soft underbelly of hard inflation data

Warnings by the US and Chinese authorities have underscored the dilemma of conflicting inflation and growth data, with energy and tight labour markets…

Published

on

By

Warnings by the US and Chinese authorities have underscored the dilemma of conflicting inflation and growth data, with energy and tight labour markets pushing up producer and consumer prices amid creeping signs of softening growth. This has put global monetary policy, and markets in risky assets, in a bind.

The Dow Jones Industrial Average fell for the seventh consecutive week last week, while the benchmark US Treasury 10-year yield hovered around 3.0% (almost double the 1.6% of a year ago). Commodity prices came under selling pressure as risk aversion among investors mounted. Safe-haven flows pushed up the US dollar, driving its trade-weighted index to near two-decade highs (see Exhibit 1).

Policy warnings…

China fanned market worries early last week, with Premier Li Keqiang warning that the domestic jobs situation was getting ‘complicated and grave’. The country’s zero-Covid policy is taking a heavy toll on the local economy with negative spillover effects globally. While Shanghai’s lockdown may be wound down soon, other major cities (including Beijing) are facing renewed restrictions.

US Federal Reserve Chair Jerome Powell issued a warning mid-week: The Fed could not guarantee a ‘soft landing’ as it looked to get runaway inflation back to its 2% target amid a tight US labour market. The US Senate nonetheless overwhelmingly confirmed Powell for a second term, signalling monetary policy continuity.

Earlier in the week, former Fed Chair Ben Bernanke warned about the risk of stagflation in an interview with The New York Times.

Aggravated by hard inflation data…

US consumer price inflation was 8.3% YoY in April, down slightly from 8.5% in March. However, core inflation (which excludes food and energy prices), rose on the month from 0.3% to 0.6%, a level still too high for the Fed’s comfort.

Services inflation was particularly strong, rising by 0.7% MoM in April, marking the biggest monthly gain since August 1990. Underscoring continued robust consumer demand, retail sales rose by 0.9% vs the prior month, though this marks the third month in a row that the growth rate has decelerated. 

The prospects for inflation to fall back to the Fed’s 2% target anytime soon may not be good: High wage growth – hourly earnings rose at around 5% YoY – could continue to fuel inflation in the near term. We note that services inflation tends to be much stickier than other index components.

From the Fed’s perspective, these price pressures could in turn drive inflation expectations higher.

The market perceives the latest inflation report as sealing a 50bp rate rise at the June and July meetings of Fed policymakers. It also boosts the chances of the Fed persisting in its aggressive tightening stance at later meetings. A key question is the extent to which – and when – higher interest rates will hit real incomes and crimp demand growth, slowing the economy overall. 

The high services inflation data also suggests labour market tightness would have to ease significantly to bring wage growth back to levels that are acceptable to the Fed. We believe something will have to give. If not, the Fed may have to tap harder on the brakes down the line.

The ECB continues to move closer towards a hawkish policy, with the market now expecting its asset purchasing programme (APP) to end in July, to be followed by a 25bp rate rise soon after. Underpinning the ECB’s policy tightening stance is strong inflation, which rose by 7.4% YoY in April (same as in March), and falling unemployment (the jobless rate hit a record low of 6.8% in March).

The war in Ukraine has added to the upside risks to inflation via food and energy price increases and supply bottlenecks. In addition to higher inflation, the ECB also appears to be concerned about the spillover effects from wage increases. An increasing number of policymakers has spoken out recently in favour of an initial rate rise as soon as July.

And creeping signs of slower growth

Indications of weakening growth momentum have appeared, most noticeably in the UK where GDP growth contracted unexpectedly by 0.1% MoM in March.

In the eurozone, industrial production shrank by 1.8% MoM in March and manufacturing output was down by 1.6%. The main culprit was disruption caused by the war in Ukraine. The weakness was concentrated in Germany, whose supply chains are more integrated with eastern Europe. Its car sector is missing components produced in Ukraine.

Even in the US, recent data showed signs of slowing growth. Jobless claims filings showed an increase in initial claims; the May Senior Loan Officer Opinion survey recorded a drop in demand for mortgages; the University of Michigan consumer sentiment May index hit its lowest level since the start of the pandemic; and the May Empire State Manufacturing survey plunged.

China also released weak data, with industrial output, fixed-asset investment and retail sales all showing year-on-year declines. The property market’s woes deepened, with new home sales and starts falling precipitously.

Investment implications

Mr. Bernanke’s warning of stagflation underscores the dilemma facing policymakers and financial markets: Inflation and growth data are sending conflicting signals. Parts of the US yield curve are inverted, pointing to some risk of an economic recession.

The slowdown concerns are linked to inflation forcing the Fed to tighten policy into restrictive territory and turning weaker growth into a contraction.

The situation is similar in the eurozone: inflation is at its highest ever and could lead the ECB to take stronger measures, exacerbating headwinds from weak Chinese activity and a Russia-induced energy supply shock.

Against the backdrop of the continuing Ukrainian conflict and prolonged supply-chain disruptions, we do not favour sovereign bonds and European equities at this point. We prefer commodities, Japanese and emerging market equities, including Chinese stocks.


Disclaimer

Please note that articles may contain technical language. For this reason, they may not be suitable for readers without professional investment experience.

Any views expressed here are those of the author as of the date of publication, are based on available information, and are subject to change without notice. Individual portfolio management teams may hold different views and may take different investment decisions for different clients. The views expressed in this podcast do not in any way constitute investment advice.

The value of investments and the income they generate may go down as well as up and it is possible that investors will not recover their initial outlay. Past performance is no guarantee for future returns.

Investing in emerging markets, or specialised or restricted sectors is likely to be subject to a higher-than-average volatility due to a high degree of concentration, greater uncertainty because less information is available, there is less liquidity or due to greater sensitivity to changes in market conditions (social, political and economic conditions). Some emerging markets offer less security than the majority of international developed markets. For this reason, services for portfolio transactions, liquidation and conservation on behalf of funds invested in emerging markets may carry greater risk.

Writen by Chi Lo. The post Weekly investment update – The soft underbelly of hard inflation data appeared first on Investors' Corner - The official blog of BNP Paribas Asset Management, the sustainable investor for a changing world.

Read More

Continue Reading

Spread & Containment

Moderna’s HIV vaccine prepped for trials in Africa

Moderna has joined forces with non-profit organisation IAVI on a third phase 1 trial of its candidate HIV
The post Moderna’s HIV vaccine prepped for…

Published

on

Moderna has joined forces with non-profit organisation IAVI on a third phase 1 trial of its candidate HIV vaccine in Africa, where the burden of the virus is still being keenly felt.

IAVI (the International AIDS Vaccine Initiative) has started screening subjects to be included in the study, called IAVI G003, at centres in Rwanda and South Africa, said the biotech.

Moderna’s vaccines deliver HIV-specific antigens discovered by researchers at IAVI and Scripps Research that have already been tested in a proof-of-concept study carried out last year using an adjuvant protein vaccine approach.

There are hopes that its mRNA approach, which proved so effective against COVID-19, could succeed where traditional vaccine technologies have failed in HIV.

One candidate – mRNA-1644 – has already shown its potential in an earlier phase 1 trial (IAVI G001) run in the US. It codes for an antigen called eOD-GT8 60mer and, in the study, stimulated a targeted B-cell immune response in 97% of vaccine recipients.

Moderna says that B-cell activation should lead to the induction of broadly neutralising antibodies (bnAbs), widely considered to be a goal of an efficacious HIV vaccine, but that immunising with eOD-GT8 60mer alone will almost certainly not be sufficient.

The biotech is looking at a combination regimen of vaccines targeting different HIV immunogens such as Core-g28v2 60mer to try to boost the immune response further against HIV and improve the protective efficacy.

Earlier this year, the first healthy volunteers were dosed with mRNA-1644 in a second phase 1 trial (IAVI G002), which is being funded in part by the Bill & Melinda Gates Foundation and is being carried out in US populations.

IAVI G003 will enrol 18 healthy HIV-negative adult volunteers who will receive two doses of the eOD-GT8 60mer mRNA shot. They will be followed for six months to gauge the safety and immunogenicity of the vaccine.

Moderna said the trial is a “first-in-Africa” study, evaluating an mRNA-delivered HIV immunogen in Africa with African researchers leading the project.

Despite more than 30 years of research, the tendency of the virus to mutate means that classical approaches to vaccine design have been ineffective, and at least four prior vaccine candidates have failed in clinical trials.

In February, one of the front-runner candidates in the decades-long quest to find an HIV vaccine – Johnson & Johnson – reported that its candidate failed a phase 2b trial.

The Ad26.Mos4.HIV vaccine – which uses the same adenoviral technology as J&J’s COVID-19 vaccine and targets four HIV antigens – showed that the shot was safe but unable to meet its target of reducing transmission of HIV by 50%.

And last year, the HVTN 702 study of two co-administered HIV candidate vaccines from Sanofi Pasteur and GlaxoSmithKline, combined with GSK’s adjuvant MF59, was also discontinued due to a lack of efficacy.

The post Moderna’s HIV vaccine prepped for trials in Africa appeared first on .

Read More

Continue Reading

Trending