Data Science newsletter – June 24, 2021

Newsletter features journalism, research papers and tools/software for June 24, 2021

 

4 ways machine learning is fixing to finetune clinical nutrition

AI in Healthcare, Dave Pearson


from

Clinical nutritionists won’t be left out of the medical AI revolution, as researchers are exploring use cases for augmented diet optimization, food image recognition, risk prediction and diet pattern analysis.

The state of the science is described in a paper published this month in Current Surgery Reports.

Applications for AI and other digital technologies are “still young, [but] there is much promise for growth and disruption in the future,” write multidisciplinary specialists at UCLA Health, San José State University and the Mayo Clinic.


Supporting open and reproducible science

Stanford University, Stanford News


from

Open science is a broad goal that includes making data, data analysis, scientific processes and published results easier to access, understand and reproduce. It’s an appealing concept but, in practice, open science is difficult and, often, the costs seem to exceed the benefits. Recognizing both the shortfalls and the promise of open science, Stanford University’s Center for Open and REproducible Science (CORES) – which is part of Stanford Data Science – hopes to make the practice of open science easier, more accessible and more rewarding.

Since its launch in September 2020, CORES has been hard at work on the center’s first major efforts. These include developing a guide for open science practices at Stanford – called the “Open by Design” handbook – and producing workshops and a lecture series to help people learn about and contribute to open science across the university.


For the first time, researchers visualize metabolic process at the single-cell level

University of Chicago, UChicago News


from

Understanding cellular metabolism – how a cell uses energy – could be key to treating a wide array of diseases, including vascular diseases and cancer.

While many techniques can measure these processes among tens of thousands of cells, researchers have been unable to measure them at the single-cell level.

Researchers at the University of Chicago’s Pritzker School of Molecular Engineering and Biological Sciences Division have developed a combined imaging and machine learning technique that can, for the first time, measure a metabolic process at both the cellular and sub-cellular levels.

Using a genetically encoded biosensor paired with artificial intelligence, the researchers were able to measure glycolysis, the process of turning glucose into energy, of single endothelial cells, the cells that line blood vessels.


UCLA Receives $4.8M NIH Grant to Improve Genetic Risk Estimates

Health IT Analytics, Jill McKeon


from

UCLA Health will receive a $4.8M National Institutes of Health (NIH) grant to improve upon genetic disease risk estimates for diverse populations. The grant will enable researchers to establish polygenic risk scores (PRS) for particular diseases using genomic datasets and form a multi-center research consortium.

The funding comes from NIH’s National Human Genome Research Institute (NHGRI) and the National Cancer Institute (NCI). Researchers will focus on finding innovative ways to calculate polygenic risk scores specifically for people of mixed ancestries, or “admixed ancestry.” These populations tend to be overlooked when it comes to biomedical research, which is the reasoning behind the project’s focus.

“More than 30% of individuals living in the U.S. self-identify as having admixed ancestry, usually defined as those with recent ancestry from two or more continental sources, such as African Americans and/or Latinx individuals,” Bogdan Pasaniuc, PhD, an associate professor at the David Geffen School of Medicine at UCLA, said in a press release.


It took a pandemic, but the US finally has (some) centralized medical data

MIT Technology Review, Cat Ferguson


from

At the beginning of the pandemic, a group of researchers funded by the US National Institutes of Health, or NIH, realized that many questions about covid-19 would be impossible to answer without breaking down barriers to data sharing. So they developed a framework for combining actual patient records from different institutions in a way that could be both private and useful.

The result is the National COVID Cohort Collaborative (N3C), which collects medical records from millions of patients around the country, cleans them, and then grants access to groups studying everything from when to use a ventilator to how covid affects menstrual cycles.

“It’s just shocking that we had no harmonized, aggregate health data for research in the face of a pandemic,” says Melissa Haendel, a professor of research informatics at the University of Colorado Anschutz Medical Campus and one of the co-leads of N3C. “We never would have gotten everyone to give us this degree of data outside the context of a pandemic, but now that we’ve done it, it’s a demonstration that clinical data can be harmonized and shared broadly in a secure way, and a transparent way.”


Stanford researchers develop new software for designing sustainable cities

Stanford University, Stanford News


from

New technology could help cities around the world improve people’s lives while saving billions of dollars. The free, open-source software developed by the Stanford Natural Capital Project creates maps to visualize the links between nature and human wellbeing. City planners and developers can use the software to visualize where investments in nature, such as parks and marshlands, can maximize benefits to people, like protection from flooding and improved health.

“This software helps design cities that are better for both people and nature,” said Anne Guerry, Chief Strategy Officer and Lead Scientist at the Natural Capital Project.


‘Nobody is catching it’: Algorithms widely used in hospitals are rife with bias

STAT, Casey Ross


from

The algorithms carry out an array of crucial tasks: helping emergency rooms nationwide triage patients, predicting who will develop diabetes, and flagging patients who need more help to manage their medical conditions.

But instead of making health care delivery more objective and precise, a new report finds, these algorithms — some of which have been in use for many years — are often making it more biased along racial and economic lines.

Researchers at the University of Chicago found that pervasive algorithmic bias is infecting countless daily decisions about how patients are treated by hospitals, insurers, and other businesses. Their report points to a gaping hole in oversight that is allowing deeply flawed products to seep into care with little or no vetting, in some cases perpetuating inequitable treatment for more than a decade before being discovered.


Right to Repair shows no signs of slowing down

U.S. PIRG


from

Our work to make sure that people have what they need to fix modern equipment has continued to grow. Here’s our latest update on work across the country — and across the world — to stand up to manufacturers who restrict repair.


Congressman Introduces National Right-to-Repair Bill

Gizmodo, Alyse Stanley


from

The right-to-repair movement has made it to Congress. On Thursday, Congressman Joseph Morelle of New York filed legislation that would make it easier for consumers to fix their broken gadgets without having to fork over even more money to the original manufacturers.

If passed, the Fair Repair Act would require manufacturers to give device owners and third-party repair shops access to replacement parts, diagnostic information, and tools needed to repair their electronics. To date, most right-to-repair legislation has been introduced at the state level, but this bill would establish a nationwide standard.


CDS Guest Editorial: Active Learning for Optimal Experiment Design in High Energy Physics

Medium, NYU Center for Data Science; Irina Esjepo Morales, Kyle Cranmer, Lukas Heinrich, Gilles Louppe


from

At NYU, we are developing an agent called “Excursion” that uses active learning and gaussian processes to find level sets of computationally expensive black box functions. Excursion is supported by GPyTorch.

We use a gaussian process as a surrogate for the black box function, then we calculate a function named acquisition function which uses the gaussian process posterior to calculate how worth it it is to feed a candidate point in the grid to the black box function. See it in action below!


The Pandemic Has Changed How Business Schools Operate for Good

Bloomberg Businessweek, Chris Stokel-Walker


from

Business schools tend to be steeped in tradition—focused on the long term and slow to respond to societal trends. The pandemic turned that on its head, forcing institutions to tear up the rule book to ensure classes could go ahead amid lockdowns and travel restrictions. “What the pandemic has done is free up space for innovation,” says Kathy Harvey, associate dean of the MBA and executive degrees at the University of Oxford’s Saïd Business School. Some of the adjustments are likely to stick:
relates to The Pandemic Has Changed How Business Schools Operate for Good
Illustration: Nichole Shinn
Smaller Classes

The MBA program at Oxford Saïd has cut its class sizes in half, from 80 students to 40.


How scientists are embracing NFTs

Nature, News, Nicola Jones


from

Is a trend of auctioning non-fungible tokens based on scientific data a fascinating art fad, an environmental disaster or the future of monetized genomics?


Former Air Force combat pilot brings drone soccer to America

Air Force Times, Caitlin O'Brien


from

A veteran Air Force combat pilot is working to make drones and robotics accessible to students through the emerging sport of drone soccer.

The sport takes the typical after-school robotics club to the next level and may look slightly familiar to Harry Potter’s Quidditch. Thanks to Maj. Kyle Sanders, it is making headway in the western US.


Google downsizes health division, reorganizes consumer-health team

MobiHealthNews, Laura Lovett


from

Google is reportedly shaking things up in its health division. The company is moving more than 130 of its roughly 700 Google Health employees to Search and its new Fitbit group, according to an Insider report.

Insider’s sources say that none of the teams left at Google Health will focus on consumer tech. This news comes roughly six months after Google officially purchased Fitbit for roughly $2.1 billion, following over a year of regulatory probes.

At the time of the M&A, Rick Osterloh, Google’s SVP of devices and services, said that Fitbit would contribute its wearable product experience and tech platform, and Google would bring its software, hardware and AI expertise to the table.


Students may now pursue a minor in computational biology

University of Colorado Boulder, Colorado Arts and Sciences Magazine


from

The minor emerged from CU Boulder’s interdisciplinary BioFrontiers Institute. Faculty had seen success in its graduate-level IQ Biology PhD program, but they wanted to address a growing gap in opportunities for undergraduates to learn about these innovative ideas.

Rather than follow the program designs of peer universities, which in many cases add some computer science coursework to a biology major or vice versa, CU Boulder’s faculty opted to make the program as interdisciplinary, cross-departmental, cross-college and collaborative as possible.

“There was a fateful meeting in the biotech building where we got 10 faculty from across campus in the room and we asked, ‘If we designed a computational biology degree from the ground up, focusing on the most important ideas in the field, what would our ideal program look like?’” says Aaron Clauset, associate professor of computer science and faculty director of the minor. “‘How would it train young scholars for the next century of biological research?’ It was an inspiring meeting.”


Deadlines



ICML 2021 Participation Grants

“As part of diversity and inclusion efforts, ICML 2021 is happy to provide grants to lower barriers of access to the conference. The grants are intended for individuals for whom attending the conference would cause a financial burden or create risks to their safety or privacy.” Deadline for applications is July 2.

SPONSORED CONTENT

Assets  




The eScience Institute’s Data Science for Social Good program is now accepting applications for student fellows and project leads for the 2021 summer session. Fellows will work with academic researchers, data scientists and public stakeholder groups on data-intensive research projects that will leverage data science approaches to address societal challenges in areas such as public policy, environmental impacts and more. Student applications due 2/15 – learn more and apply here. DSSG is also soliciting project proposals from academic researchers, public agencies, nonprofit entities and industry who are looking for an opportunity to work closely with data science professionals and students on focused, collaborative projects to make better use of their data. Proposal submissions are due 2/22.

 


Tools & Resources



Some tips on how to go from academia to a Data Science position

reddit/r/datascience, 7 comments


from

I read a lot of posts about how to land a data science position and what the biggest differences are between doing research in academia and working as a Data Scientist. Since I did the same journey myself, a couple of years ago, I thought it might be helpful if I summarised my experience and my thoughts about how to successfully do the transfer.


Careers


Full-time positions outside academia

AI Researcher



Sportlogiq; Montreal, QC, Canada
Internships and other temporary positions

DataSquad Project Management



UCLA Data Science Center DataSquad; Los Angeles, CA
Postdocs

Fellowship, Data Science



American Heart Association; Dallas, TX
Full-time, non-tenured academic positions

Research Assistant III Non-Lab



Harvard T.H. Chan School of Public Health, Department of Biostatistics; Boston, MA

Leave a Comment

Your email address will not be published.