‘Selfies’ could be used to detect heart disease: new research uses artificial intelligence to analyse facial photos

Curated on Posted on by Stefaan Verhulst

European Society of Cardiology: “Sending a “selfie” to the doctor could be a cheap and simple way of detecting heart disease, according to the authors of a new study published today (Friday) in the European Heart Journal [1].

The study is the first to show that it’s possible to use a deep learning computer algorithm to detect coronary artery disease (CAD) by analysing four photographs of a person’s face.

Although the algorithm needs to be developed further and tested in larger groups of people from different ethnic backgrounds, the researchers say it has the potential to be used as a screening tool that could identify possible heart disease in people in the general population or in high-risk groups, who could be referred for further clinical investigations.

“To our knowledge, this is the first work demonstrating that artificial intelligence can be used to analyse faces to detect heart disease. It is a step towards the development of a deep learning-based tool that could be used to assess the risk of heart disease, either in outpatient clinics or by means of patients taking ‘selfies’ to perform their own screening. This could guide further diagnostic testing or a clinical visit,” said Professor Zhe Zheng, who led the research and is vice director of the National Center for Cardiovascular Diseases and vice president of Fuwai Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, People’s Republic of China.

He continued: “Our ultimate goal is to develop a self-reported application for high risk communities to assess heart disease risk in advance of visiting a clinic. This could be a cheap, simple and effective of identifying patients who need further investigation. However, the algorithm requires further refinement and external validation in other populations and ethnicities.”

It is known already that certain facial features are associated with an increased risk of heart disease. These include thinning or grey hair, wrinkles, ear lobe crease, xanthelasmata (small, yellow deposits of cholesterol underneath the skin, usually around the eyelids) and arcus corneae (fat and cholesterol deposits that appear as a hazy white, grey or blue opaque ring in the outer edges of the cornea). However, they are difficult for humans to use successfully to predict and quantify heart disease risk.

Prof. Zheng, Professor Xiang-Yang Ji, who is director of the Brain and Cognition Institute in the Department of Automation at Tsinghua University, Beijing, and other colleagues enrolled 5,796 patients from eight hospitals in China to the study between July 2017 and March 2019. The patients were undergoing imaging procedures to investigate their blood vessels, such as coronary angiography or coronary computed tomography angiography (CCTA). They were divided randomly into training (5,216 patients, 90%) or validation (580, 10%) groups.

Trained research nurses took four facial photos with digital cameras: one frontal, two profiles and one view of the top of the head. They also interviewed the patients to collect data on socioeconomic status, lifestyle and medical history. Radiologists reviewed the patients’ angiograms and assessed the degree of heart disease depending on how many blood vessels were narrowed by 50% or more (≥ 50% stenosis), and their location. This information was used to create, train and validate the deep learning algorithm….(More)”.

Designing Governance as Collective Intelligence

Curated on Posted on by Stefaan Verhulst

Paper by Hamed Khaledi: “This research models governance as a collective intelligence process, particularly as a collective design process. The outcome of this process is a solution to a problem. The solution can be a decision, a policy, a product, a financial plan, etc. The quality (value) of the outcome solution reflects the quality (performance) of the process. Using an analytical model, I identify five mediators (channels) through which, different factors and features can affect the quality of the outcome and thus the process. Based on this model, I propose an asymmetric response surface method that introduces factors to the experimental model considering their plausible effects.

As a proof of concept, I implemented a generic collective design process in a web application and measured the effects of several factors on its performance through online experiments. The results demonstrate the effectiveness of the proposed method. They also show that approval voting is significantly superior to plurality voting. Some studies assert that not the design process, but the designers drive the quality of the outcome. However, this study shows that the characteristics of the design process (e.g. voting schemes) as well as the designers (e.g. expertise and gender) can significantly affect the quality of the outcome. Hence, the outcome quality can be used as an indicator of the performance of the process. This enables us to evaluate and compare governance mechanisms objectively free from fairness criteria….(More)”.

Prediction paradigm: the human price of instrumentalism

Curated on Posted on by Stefaan Verhulst

Editorial by Karamjit S. Gill at AI&Society: “Reflecting on the rise of instrumentalism, we learn how it has travelled across the academic boundary to the high-tech culture of Silicon Valley. At its core lies the prediction paradigm. Under the cloak of inevitability of technology, we are being offered the prediction paradigm as the technological dream of public safety, national security, fraud detection, and even disease control and diagnosis. For example, there are offers of facial recognition systems for predicting behaviour of citizens, offers of surveillance drones for ’biometric readings’, ‘Predictive Policing’ is offered as an effective tool to predict and reduce crime rates. A recent critical review of the prediction technology (Coalition for Critical Technology 2020), brings to our notice the discriminatory consequences of predicting “criminality” using biometric and/or criminal legal data.

The review outlines the specific ways crime prediction technology reproduces, naturalizes and amplifies discriminatory outcomes, and why exclusively technical criteria are insufficient for evaluating their risks. We learn that neither predication architectures nor machine learning programs are neutral, they often uncritically inherit, accept and incorporate dominant cultural and belief systems, which are then normalised. For example, “Predictions” based on finding correlations between facial features and criminality are accepted as valid, interpreted as the product of intelligent and “objective” technical assessments. Furthermore, the data from predictive outcomes and recommendations are fed back into the system, thereby reproducing and confirming biased correlations. The consequence of this feedback loop, especially in facial recognition architectures, combined with a belief in “evidence based” diagnosis, is that it leads to ‘widespread mischaracterizations of criminal justice data’ that ‘justifies the exclusion and repression of marginalized populations through the construction of “risky” or “deviant” profiles’…(More).

The Rise of the Data Poor: The COVID-19 Pandemic Seen From the Margins

Curated on Posted on by Stefaan Verhulst

Essay by Stefania Milan and Emiliano Treré: “Quantification is central to the narration of the COVID-19 pandemic. Numbers determine the existence of the problem and affect our ability to care and contribute to relief efforts. Yet many communities at the margins, including many areas of the Global South, are virtually absent from this number-based narration of the pandemic. This essay builds on critical data studies to warn against the universalization of problems, narratives, and responses to the virus. To this end, it explores two types of data gaps and the corresponding “data poor.” The first gap concerns the data poverty perduring in low-income countries and jeopardizing their ability to adequately respond to the pandemic. The second affects vulnerable populations within a variety of geopolitical and socio-political contexts, whereby data poverty constitutes a dangerous form of invisibility which perpetuates various forms of inequality. But, even during the pandemic, the disempowered manage to create innovative forms of solidarity from below that partially mitigate the negative effects of their invisibility….(More)”.

Mapping socioeconomic indicators using social media advertising data

Curated on Posted on by Stefaan Verhulst

Paper by Ingmar Weber et al: “The United Nations Sustainable Development Goals (SDGs) are a global consensus on the world’s most pressing challenges. They come with a set of 232 indicators against which countries should regularly monitor their progress, ensuring that everyone is represented in up-to-date data that can be used to make decisions to improve people’s lives. However, existing data sources to measure progress on the SDGs are often outdated or lacking appropriate disaggregation. We evaluate the value that anonymous, publicly accessible advertising data from Facebook can provide in mapping socio-economic development in two low and middle income countries, the Philippines and India. Concretely, we show that audience estimates of how many Facebook users in a given location use particular device types, such as Android vs. iOS devices, or particular connection types, such as 2G vs. 4G, provide strong signals for modeling regional variation in the Wealth Index (WI), derived from the Demographic and Health Survey (DHS). We further show that, surprisingly, the predictive power of these digital connectivity features is roughly equal at both the high and low ends of the WI spectrum. Finally we show how such data can be used to create gender-disaggregated predictions, but that these predictions only appear plausible in contexts with gender equal Facebook usage, such as the Philippines, but not in contexts with large gender Facebook gaps, such as India….(More)”.

When Mini-Publics and Maxi-Publics Coincide: Ireland’s National Debate on Abortion

Curated on Posted on by Stefaan Verhulst

Paper by David Farrell et al: “Ireland’s Citizens’ Assembly (CA) of 2016–18 was tasked with making recommendations on abortion. This paper shows that from the outset its members were in large part in favour of the liberalisation of abortion (though a fair proportion were undecided), that over the course of its deliberations the CA as a whole moved in a more liberal direction on the issue, but that its position was largely reflected in the subsequent referendum vote by the population as a whole….(More)”

Going Beyond the Smart City? Implementing Technopolitical Platforms for Urban Democracy in Madrid and Barcelona

Curated on Posted on by Stefaan Verhulst

Paper by Adrian Smith & Pedro Prieto Martín: “Digital platforms for urban democracy are analyzed in Madrid and Barcelona. These platforms permit citizens to debate urban issues with other citizens; to propose developments, plans, and policies for city authorities; and to influence how city budgets are spent. Contrasting with neoliberal assumptions about Smart Citizenship, the technopolitics discourse underpinning these developments recognizes that the technologies facilitating participation have themselves to be developed democratically. That is, technopolitical platforms are built and operate as open, commons-based processes for learning, reflection, and adaptation. These features prove vital to platform implementation consistent with aspirations for citizen engagement and activism….(More)”.

How the Administrative State Got to This Challenging Place

Curated on Posted on by Stefaan Verhulst

Essay by Peter Strauss: “This essay has been written to set the context for a future issue of Daedalus, the quarterly of the American Academy of Arts and Sciences, addressing the prospects of American administrative law in the Twenty-first Century. It recounts the growth of American government over the centuries since its founding, in response to the profound changes in the technology, economy, and scientific understandings it must deal with, under a Constitution written for the governance of a dispersed agrarian population operating with hand tools in a localized economy. It then suggests profound challenges of the present day facing administrative law’s development: the transition from processes of the paper age to those of the digital age; the steadily growing centralization of decision in an opaque, political presidency, displacing the focused knowledge and expertise of agencies Congress created to pursue particular governmental ends; the thickening, as well, of the political layer within agencies themselves, threatening similar displacements; and the revival in the courts of highly formalized analytic techniques inviting a return to the forms of government those who wrote the Constitution might themselves have imagined. The essay will not be published until months after the November election. While President Trump’s first term in office has sharply illustrated an imbalance in American governance between law and politics and law, reason and unreason, that imbalance is hardly new; it has been growing for decades. There lie the challenges….(More)”

The $100 Million Nudge: Increasing Tax Compliance of Businesses and the Self-Employed using a Natural Field Experiment

Curated on Posted on by Stefaan Verhulst

Paper by Justin E. Holz et al: “This paper uses a natural field experiment to examine the effectiveness of specific nudges on tax compliance amongst firms and the self-employed in the Dominican Republic. In collaboration with the Dominican Republic’s tax authority, we designed messages for more than 28,000 self-employed workers and over 56,000 firms. Leveraging administrative tax data, we find evidence that our nudges (increasing the salience of prison sentences or public disclosure of tax evaders) have large effects on increasing tax compliance, primarily working through the channel of decreasing claimed tax exemptions. Interestingly, we find that firms are more impacted than the self-employed, and that firm size is critically linked to nudge effectiveness: larger firms are considerably more influenced by nudges than smaller firms. We find this latter result noteworthy given the paucity of evidence showing significant behavioral impacts of nudges amongst the largest players in a market. Overall, our messages increased tax revenue by $193 million (roughly 0.23% of the Dominican Republic’s GDP in 2018), with over $100 million constituting income that the government would not have received without our field experimental nudges….(More)”.

Digital technologies in the public-health response to COVID-19

Curated on Posted on by Stefaan Verhulst

Paper by Jobie Budd et al in Nature Medicine: “Digital technologies are being harnessed to support the public-health response to COVID-19 worldwide, including population surveillance, case identification, contact tracing and evaluation of interventions on the basis of mobility data and communication with the public. These rapid responses leverage billions of mobile phones, large online datasets, connected devices, relatively low-cost computing resources and advances in machine learning and natural language processing. This Review aims to capture the breadth of digital innovations for the public-health response to COVID-19 worldwide and their limitations, and barriers to their implementation, including legal, ethical and privacy barriers, as well as organizational and workforce barriers. The future of public health is likely to become increasingly digital, and we review the need for the alignment of international strategies for the regulation, evaluation and use of digital technologies to strengthen pandemic management, and future preparedness for COVID-19 and other infectious diseases….(More)”.