Healthcare Technology Insights for May 1st — May 14th, 2021
Here are the most-read articles from our Digital Health Blogs and Curated Newsletters over the past 2 weeks. Insights by Nrip Nihalani
Vaccination Passports — What are they?
As mass vaccination programs are being rolled out globally, vaccine passports have become a major topic of discussion.
COVID test results and proof of vaccine will be required in many countries for quarantine-free travel, just as it has been for polio and yellow fever vaccinations in the past.
Countries will need to look at convenient and secure ways for verifying COVID-19 test results and vaccination information at airports and borders.
The International Air Transport Association (IATA) has also called for a “global standard to securely record digital proof of vaccination”. They have been promoting the IATA Travel Pass Initiative (https://www.iata.org/en/programs/passenger/travel-pass/)
In February, Qantas completed a trial run of an app for this purpose on an international repatriation flight from Frankfurt to Darwin.
The idea behind the app is that health or border officials and airline staff may be able to easily verify COVID-19 test results and the vaccination history of an individual.
The app links customers with certified testing labs to allow their results to be automatically uploaded onto it.
Similar digital solutions are being developed in several other countries around the world to enable travel again. For instance, travelers from Singapore will receive a notarized certificate following a negative COVID-19 test that they can present at airports around the world.
Another example is France taking part in a month-long trial of a vaccine passport that leverages a smartphone app.
It's important that such digital health technologies, whether apps or chip cards or health tracker add ons, be easy to use. It is important that the process be as seamless as possible for flyers to share the relevant information and get the information validated by the ground and air staff so they can travel internationally, again, safely!
image source: https://foto.wuestenigel.com/person-hands-holding-a-covid-19-passport/
image license:
https://creativecommons.org/licenses/by/2.0/
From www.scoop.it
Analyzing the Essential Attributes of Nationally Issued COVID-19 Contact Tracing Apps
Contact tracing apps are potentially useful tools for supporting national COVID-19 containment strategies. Various national apps with different technical design features have been commissioned and issued by governments worldwide.
Objective: Our goal was to develop and propose an item set that was suitable for describing and monitoring nationally issued COVID-19 contact tracing apps.
This item set could provide a framework for describing the key technical features of such apps and monitoring their use based on widely available information.
Methods: We used an open-source intelligence approach (OSINT) to access a multitude of publicly available sources and collect data and information regarding the development and use of contact tracing apps in different countries over several months (from June 2020 to January 2021). The collected documents were then iteratively analyzed via content analysis methods. During this process, an initial set of subject areas were refined into categories for evaluation (ie, coherent topics), which were then examined for individual features.
These features were paraphrased as items in the form of questions and applied to information materials from a sample of countries (ie, Brazil, China, Finland, France, Germany, Italy, Singapore, South Korea, Spain, and the United Kingdom [England and Wales]). This sample was purposefully selected; our intention was to include the apps of different countries from around the world and to propose a valid item set that can be relatively easily applied by using an OSINT approach.
Results: Our OSINT approach and subsequent analysis of the collected documents resulted in the definition of the following five main categories and associated subcategories:
(1) background information (open-source code, public information, and collaborators);
(2) purpose and workflow (secondary data use and warning process design);
(3) technical information (protocol, tracing technology, exposure notification system, and interoperability);
(4) privacy protection (the entity of trust and anonymity); and
(5) availability and use (release date and the number of downloads).
Based on this structure, a set of items that constituted the evaluation framework was specified. The application of these items to the 10 selected countries revealed differences, especially with regard to the centralization of the entity of trust and the overall transparency of the apps’ technical makeup.
Conclusions: We provide a set of criteria for monitoring and evaluating COVID-19 tracing apps that can be easily applied to publicly issued information. The application of these criteria might help governments to identify design features that promote the successful, widespread adoption of COVID-19 tracing apps among target populations and across national boundaries.
read the study at https://mhealth.jmir.org/2021/3/e27232
Our insights :
Where a lot of studies falter, is they dont focus on ease of use as a primary criteria of evaluation. Digital Health tools for far too long have faced criticism due to the ease of use factor.
It takes several iterations for any app/tool to become easy to use when the use cases contain a lot of data input. As such, contact tracing tools will do well by being built over surveillance and data collection platforms like MediXcel Lite and Commcare.
The data collection platforms must also focus on contact tracing as a type of app they generate along with the longitudinal and case based apps they currently allow.
Shortcomings with the AI Tools and Devices Preventing COVID-19?
Since the start of the pandemic, new technologies have been developed to help reduce the spread of the infection.
Some of the most common safety measures today include measuring a person’s temperature, covering your nose and mouth with a mask, contact tracing, disinfection, and social distancing. Many businesses have adopted various technologies, including those with artificial intelligence (AI) underneath, helping to adhere to the COVID-19 safety measures.
As an example, numerous airlines, hotels, subways, shopping malls, and other institutions are already using thermal cameras to measure an individual’s temperature before people are allowed entry. In its turn, public transport in France relies on AI-based surveillance cameras to monitor whether or not people are social-distancing or wearing masks. Another example is requiring the download of contact-tracing apps delivered by governments across the globe.
However, there are a number of issues.
While many of these solutions help to ensure that COVID-19 prevention practices are observed, many of them have flaws or limits. In this article, we will cover some of the issues creating obstacles for fighting the pandemic.
Issue #1. Manual temperature scanning is tricky
Issue #2. Monitoring crowds is even more complex
Issue #3. Contact tracing leads to privacy concerns
Issue #4. UV rays harm eyes and skin
Issue #5. UVC robots are extremely expensive
Issue #6. No integration, no compliance, no transparency
Regardless of the safety measures in place and existing issues, innovations are already playing a vital role in the fight against COVID-19. By improving on existing technology, we can make everyone safer as we all adjust to the new normal.
read the details at https://www.altoros.com/blog/whats-wrong-with-ai-tools-and-devices-preventing-covid-19/
Our insight :
Yes, there are issues with some of the innovations being used. But a faster response is a useful response. I found this post extremely well researched and accurate , and not necessarily negetive. We need criticism of good intentions to make them better. This post does that. These is a valuable list of some shortcomings and some mistakes which will be worked on and improved. Sometimes by changing the system, sometimes by changing the financial model, and sometimes by changing behaviour and mindset.
The future of healthcare contains a lot of AI. That bit is true.
Estimating COVID Severity Based on Mutations in the SARS-CoV-2 Genome
Numerous studies demonstrate frequent mutations in the genome of SARS-CoV-2. Our goal was to statistically link mutations to severe disease outcome.
We found that automated machine learning, such as the method of Tsamardinos and coworkers used here, is a versatile and effective tool to find salient features in large and noisy databases, such as the fast-growing collection of SARS-CoV-2 genomes.
In this work, we used machine learning techniques to select mutation signatures associated with severe SARS-CoV-2 infections. We grouped patients into 2 major categories (“mild” and “severe”) by grouping the 179 outcome designations in the GISAID database.
A protocol combined of logistic regression and feature selection algorithms revealed that mutation signatures of about twenty mutations can be used to separate the two groups. The mutation signature is in good agreement with the variants well known from previous genome sequencing studies, including Spike protein variants V1176F and S477N that co-occur with DG14G mutations and account for a large proportion of fast spreading SARS-CoV-2 variants. UTR mutations were also selected as part of the best mutation signatures. The mutations identified here are also part of previous, statistically derived mutation profiles.
An online prediction platform was set up that can assign a probabilistic measure of infection severity to SARS-CoV-2 sequences, including a qualitative index of the strength of the diagnosis. The data confirm that machine learning methods can be conveniently used to select genomic mutations associated with disease severity, but one has to be cautious that such statistical associations — like common sequence signatures, or marker fingerprints in general — are by no means causal relations, unless confirmed by experiments.
Our plans are to update the predictions server in regular time intervals. While this project was underway more than 100 thousand sequences were deposited in public databases, and importantly, new variants emerged in the UK and in South Africa that are not yet included in the current datasets. Also, in addition to mutations, we plan to include also insertions and deletions which will hopefully further improve the predictive power of the server.
The study was funded by the Hungarian Ministry for Innovation and Technology (MIT), within the framework of the Bionic thematic program of the Semmelweis University.
Read the entire study at https://www.biorxiv.org/content/10.1101/2021.04.01.438063v1.full
Access the online portal mentioned above at https://covidoutcome.com/
Our insight :
I love studies like this. Each one builds upon the value provided by the previous one. AI in Healthcare keeps getting better. and that opens up the door for Healthcare to become more accurate, and eventually faster.
Key takeaways -
Artificial intelligence is an effective tool for uncovering hidden associations in large medical datasets.
The mutation signature of the virus be used as an indicator of the severity of the disease
AI algorithm that can detect the presence of COVID-19 disease in Chest X Rays
“ATMAN AI”, an Artificial Intelligence algorithm that can detect the presence of COVID-19 disease in Chest X Rays, has been developed to combat COVID fatalities involving the lung. ATMAN AI is used for chest X-ray screening as a triaging tool in Covid-19 diagnosis, a method for rapid identification and assessment of lung involvement. This is a joint effort of the DRDO Centre for Artificial Intelligence and Robotics (CAIR), 5C Network & HCG Academics. This will be utilized by online diagnostic startup 5C Network with the support of HCG Academics across India.
Triaging COVID suspect patients using X-Ray is fast, cost-effective, and efficient. It can be a very useful tool especially in smaller towns in India owing to the lack of easy access to CT scans there.
This will also reduce the existing burden on radiologists and make CT machines which are being used for COVID be used for other diseases and illness owing to overload for CT scans.
The novel feature namely “Believable AI” along with existing ResNet models has improved the accuracy of the software and since it is a machine learning tool, the accuracy will improve continually.
Chest X-Rays of RT-PCR positive hospitalized patients in various stages of disease involvement were retrospectively analyzed using Deep Learning & Convolutional Neural Network models by an indigenously developed deep learning application by CAIR-DRDO for COVID -19 screening using digital chest X-Rays. The algorithm showed an accuracy of 96.73%.
read more at http://indiaai.gov.in/news/drdo-cair-5g-network-and-hcg-academics-develop-atman-ai
Our insight:
Utilizing algorithms for chest X-ray is an effective triaging tool. Once perfected these can accessible by people in remote areas. Thus offering significant improvements in the care process as encountered in rural and remote areas.
Similar methods are being used/experimented on by a variety of labs and digital health companies, for predominant respiratory diseases.
Plus91 has developed similar technology for different Pneumonia and TB.
Please use the comments section below, or the form on the right in case you wish to simply obtain some specific information or wish to collaborate and use our help with building out or scaling health tech in Brazil.
Need help selecting which conferences to attend or simply to get accurate information about one (Medical and life sciences only), check out our trusted friends at http://medicaleventsguide.com/
Keep reading our curations at https://www.scoop.it/t/healthcare-technology and http://hcsm.plus91.in
Follow Plus91 Technologies on Twitter [https://www.twitter.com/plus91