Algorithm Uses Phone Camera, AI to Detect Blood Oxygen Levels

 Ai driven algorithm captures blood oxygen levels on Smartphone.  Note 'Hyperspectral Learning'.

Algorithm Uses Phone Camera to Detect Blood Oxygen Levels

By Purdue University, June 9, 2023

A smartphone camera paired with an AI-driven algorithm can capture blood oxygen data faster and more efficiently than highly specialized equipment.

The researchers used a computational approach that they described as “hyperspectral learning.”

Purdue University researchers have developed an algorithm to improve the speed and accuracy of medical diagnoses using smartphone sensors.

Although smartphone cameras capture only red, green, and blue wavelengths of light in each pixel, the researchers were able to reconstruct the full spectrum of visible light in each pixel of an image taken by a smartphone camera using deep learning, statistical techniques, and an understanding of light-tissue interactions.

The researchers found their technology produced information about blood oxygen levels in study participants' eyelids faster than commercially available hyperspectral imaging equipment, while being less expensive and just as accurate.

From Purdue University

View Full Article   

Towards Talking Concrete

Novel infrastructure idea.

'Talking' Concrete Could Help Prevent Traffic Jams, Cut Carbon Emissions

By Purdue University News

March 16, 2023

From beneath a concrete pour, this black circular sensor transmits data about the concrete's strength levels through a cord plugged into an above-ground handheld device called a data logger.

Credit: Rebecca McElhoe/Purdue University

U.S. interstates are preparing to test sensors developed by Purdue University researchers that could help prevent congestion and lower carbon emissions.

The sensors enable concrete pavement to relay data about its strength and repair requirements to engineers, forgoing the need to test samples; this should reduce construction time and repair frequency while cutting emissions from vehicles waiting to bypass construction sites.

The devices communicate to engineers through a smartphone application when the pavement is sufficiently strong to tolerate heavy traffic.

More than half of U.S. states with concrete interstate pavement will deploy the sensors as part of a Federal Highway Administration study, with Indiana and Texas already testing them in highway paving projects.

The WaveLogix company will manufacture the technology for commercialization as the REBEL Concrete Strength Sensing System.

From Purdue University News

View Full Article

Forestry Management Efforts at Purdue

More on Forestry and LiDAR, Purdue work there. integrated with AI.  Considerable detail.

Integration Leads to Leap in Tech for Forest Inventory, Management

Purdue University News,  Elizabeth K. Gardner, June 6, 2022

Purdue University researchers located, tallied, and measured more than 1,000 trees in just hours by combining aerial and ground-based mobile mapping sensors and systems. The project, part of Purdue's Digital Forestry Initiative, employs manned aircraft, unmanned aerial drones, and backpack-mounted systems that integrate cameras with light detection and ranging (LiDAR) units, along with sensors including global navigation satellite systems and inertial navigation systems. The researchers also created a machine learning algorithm to analyze the data. Purdue's Songlin Fei called the scheme "a groundbreaking development on our path to using technology for a quick, accurate inventory of the global forest ecosystem, which will improve our ability to prevent forest fires, detect disease, perform accurate carbon counting, and make informed forest management decisions."  ... 

AI Taking Over Gadgets

Now lets make them as secure as possible.    And the AI does not have to be created on the edge, it can  be encrypted and fed to devices.    We will see both.  

How AI Is Taking Over Our Gadgets   By The Wall Street Journal. July 1, 2021

 Artificial intelligence (AI) is taking over "edge" devices, including smartphones, automobiles, drones, home appliances, industrial sensors and actuators, and other devices with a microchip and some memory.

An example of this is Apple's Siri assistant, which soon will begin processing voice on iPhones using a "neural engine," rather than sending an audio recording to the cloud for processing. While AI in the cloud can be trained on the fly and improved over time, edge AI systems are pre-trained and updated periodically.

Purdue University's Elisa Bertino said one edge device could collect data and then pair with a more powerful device that can integrate data from a variety of sensors, like how a smartwatch with a heart-rate monitor transmits the data to the user's smartphone for additional data analysis.

Gartner's Eric Goodness said edge AI systems could become powerful enough to gather data and use it to train their own algorithms.  ... '

Towards a Quantum Workspace

Hmmm, why a 'Quantum Age', is it not just a computing age using new tools?     But as a structure for training in a new space, OK. 

Building a Workforce for the Quantum Age  By Purdue University

With the emergence of quantum technology, Purdue University is working to build a quantum workforce, using an array of tools including an upcoming summer school, online "micromasters" degree, adapted coursework, clubs, and seminars.

 "Quantum has the potential to be revolutionary technology," says David Stewart, managing director of the Purdue Quantum Science and Engineering Institute. "But right now, there just aren't enough people to do the work that's needed."

Purdue is pioneering ways to give engineers and other scientists a workable quantum background quickly.

"We are developing programs to train the next generation of quantum scientists," says Alexandra Boltasseva, workforce development lead for the Quantum Science Center. "The more education that's available, the more people and events students have access to, the more likely they are to connect and spark cross-institutional collaboration, which will lead to future advances. Our vision is to equip scientists and engineers from all sorts of different disciplines to participate in the quantum workforce and open more doors for our young people."

From Purdue University

Quantum Wavelength Switch

Interesting direction outlines.   Note the number of users in the quantum network being key.

Quantum Internet Closer to Reality, Thanks to This Switch

By Purdue University News,  March 3, 2021

A new study has found that using a programmable wavelength-selective switch can help increase the number of users in a quantum network without increasing photon loss from the switching device.

Engineers at Purdue University and the U.S. Department of Energy's Oak Ridge National Laboratory have overcome an obstacle in the development of a quantum Internet via a programmable switch that adjusts the amount of data sent to each user by selecting and rerouting wavelengths of light carrying data channels.

This could increase user numbers without incurring greater photon and data loss as the network expands. The approach removes the need to physically interchange many fixed optical filters tuned to individual wavelengths, by simply programming the switch to route data-carrying wavelengths to each new user.

The switch also can be programmed to modify bandwidth based on user needs.  ... " 

Computational Models Help Translate Drugs From Animal Studies

Could be a considerable development.    We need better drug development methods now.  And expect considerable pushback here too.

Computational Model Could Improve Success in Translating Drugs From Animal Studies to Humans
Purdue University News
By Kayla Wiles

Researchers from Purdue University and the Massachusetts Institute of Technology created a computational model to help translate drug development from animal studies to people. The TransComp-R model consolidates thousands of measurements from an animal model to a small number of data coordinates for human comparison, highlighting the most relevant sources of biological divergence; scientists could then train other models to anticipate human response to treatment. The researchers used TransComp-R to find an overlooked biological mechanism that may underpin human resistance to the inflammatory bowel disease drug infliximab. Purdue's Doug Brubaker said, "The modeling framework itself can be repurposed to different kinds of animals, different disease areas, and different questions. Figuring out when what we see in animals doesn't track with what's happening in humans could save a lot of time, cost, and effort in the drug development process overall."

How Have People Responded to Pandemic?

Fascinating piece, recall using data from 'discovered' cameras to determine  city walking traffic patterns.   Also using data from public cameras, with scrubbed facial images, could be acquired for similar analysis.  Today with a increased attention to privacy,  much of this data would have to be further scrubbed of signs of identity.   Still very useful patterns of  usage and behavior of people over time in the pandemic could be derived.   Note the use of 120K cameras here to generate data worldwide.

How Have People Have Responded to Covid-19 Restrictions Around the World?
Purdue University News
By Kayla Wiles

Purdue University engineers have built a website that pools live public videos and images from about 30,000 network cameras in more than 100 countries, to make it easier to analyze responses to Covid-19 restriction policies. The system automatically discovers cameras in public spaces, then an algorithm saves image data and downloads videos every 10 minutes or so, to be sent to cloud data centers for processing. The discovered cameras are part of a system called the Continuous Analysis of Many CAMeras, which taps roughly 120,000 cameras worldwide. The data also is helping to construct models for human interactions and disease proliferation.  ... ' 

Multi-Agent VR for Task Application?

Might this be useful for simulating multi-agent complex, interactive tasks in VR?    We encountered such problems when assigning multiple people to do a job and they needed to be informed of the location, status, actions ...   of others in the team.  This inhibited the use of VR solutions.   Like too that this could be done with simple devices.

Novel System Allows Untethered Multi-Player VR
Purdue University News  By Chris Adam

Purdue University researchers have created a virtual reality (VR) system that allows untethered multi-player gameplay on smartphones. The Coterie system manages the rendering of high-resolution virtual scenes to fulfill the quality of experience of VR, facilitating 4K resolutions on commodity smartphones and accommodating up to 10 players to engage in the same VR application at once. Purdue's Y. Charlie Hu said Coterie "opens the door for enterprise applications such as employee training, collaboration and operations, healthcare applications such as surgical training, as well as education and military applications."

Sensing Tech to Improve ML Precision

A photo of the sensing model created at Purdue that works with ML for applications like e-cars and home design. Sensing Technology Could Improve Machine Learning Precision for Manufacturing, Electric Vehicles, Smart Homes

Purdue University News
Chris Adam
December 11, 2019

Purdue University researchers have developed a sensor that tracks electrical current, which could be used with machine learning to track currents in applications ranging from electric cars to manufacturing and home design. The device can transmit the current data it captures to any computing system via Bluetooth, USB, or other techniques. The researchers used a machine learning algorithm with the sensor to help it gather and interpret energy data. Purdue's Byunghoo Jung said, "This sensor could be used with machine learning to train manufacturing robots, provide precise tips for homeowners on cutting down their energy usage, or help diagnose issues with electric vehicles and scooters."... "

Cloud Based CyberCrime Evidence Detection

Links to other approaches I have recently seen in this area.

CIT Researchers Develop System to Detect Cloud-Based Cybercrime Evidence
Purdue University News
John O'Malley

Researchers at the Purdue Polytechnic Institute's Department of Computer and Information Technology (CIT) have developed a forensic model that uses machine learning to collect digital evidence related to unlawful activities on cloud storage apps. The technology performs real-time identification and analysis of cybercrime-related incidents via transactions uploaded to the apps. An app user uploading a media file causes the system to implement deep learning models to scan images for cybercrime evidence and report illegal activities through a forensic evidence collection system. Cloud service providers can compile alert logs, block associated accounts, and notify law enforcement based on a cloud search warrant request. CIT's Fahad Salamh said, "It is important to automate the process of digital forensic and incident response in order to cope with advanced technology and sophisticated hiding techniques and to reduce the mass storage of digital evidence on cases involving cloud storage applications."  ... ' 

Fighting Cybercrime in the Cloud

Evidence for Cybercrime in the cloud.

Machine Learning Advances Tool to Fight Cybercrime in the Cloud
Purdue University News
By Chris Adam

Purdue University researchers used machine learning to develop a cloud forensic model that collects digital evidence associated with illegal activities in cloud storage applications. The system deploys deep learning models to classify child exploitation, illegal drug trafficking, and illegal firearms transactions uploaded to cloud storage applications, and to automatically report detection of any such illegal activities via a forensic evidence collection system. The researchers tested the system on more than 1,500 images, and found that the model accurately classified an image about 96% of the time. Said Purdue’s Fahad Salamh, "It is important to automate the process of digital forensic and incident response in order to cope with advanced technology and sophisticated hiding techniques and to reduce the mass storage of digital evidence on cases involving cloud storage applications." .... "

Listening for Indicators of Failure

Not an uncommon thing that is done by human engineers responsible for maintenance.  Now proposed to be more autonomous.

'Listening' to Engine Blades to Stop Failures, Disasters   By Purdue University News

This new method can measure and monitor blade vibration at the same time.
Purdue University researchers have developed a monitoring system that can detect one of the most common causes of premature blade failure in gas turbine engines: rotor forced response vibration.

Purdue University researchers have developed a monitoring system to identify the sound of rotor-forced response vibration, a common cause of premature blade malfunction in gas turbine engines.

The system employs multiple unsteady pressure sensors to listen for specific pressure waves associated with turbine engine blade vibration. Gas turbine blades are typically very lightly damped, and can function like a tuning fork; their blades can produce a specific frequency or tone when they resonate, which the pressure sensors are designed to pick up.

Said Purdue's Nicole Key, "With the help of big data analytics, the blade vibration information can be used to predict the possible engine failure and optimize preventive maintenance schedules."

From Purdue University News
View Full Article

Collecting Soil Samples

More autonomous, continuous and precise means of gathering data to make decisions.

Autonomous robots collect precise soil samples, help farmers improve yields, reduce environmental impact   by Drew Schumacher, Purdue University

A pair of Purdue University graduates have founded a company called Rogo Ag LLC, a pioneering agricultural technology company that has developed an autonomous robot called Smartcore. Smartcore is designed to collect accurate, repeatable soil samples in fields and bring the samples directly to the farmers or growers.  ...

Collecting precise soil samples is essential for farmers because a small amount of soil determines the amount of nutrients needed for acres of crops and can determine crop yields.

The U.S. Department of Agriculture reports that soil testing can help farmers increase yields, reduce production costs and prevent surface and groundwater pollution.

"Smartcore," an autonomous robot developed by a pair of Purdue University College of Engineering graduates, is designed to collect accurate, repeatable soil samples in fields and bring to the edge of the field for shipment to the lab. Troy Fiechter and Drew Schumacher founded Rogo, officially Rogo Ag LLC, a startup to advance the technology and move it to the public.  .... " 

Planning Cities of the Future with AI

Like to see much more about this.... planning is tough, and rarely done completely.    Analytic and AI methods require considerable attention to detail and process.  Doing planning well with these methods is probably difficult.  Visualization does help.

AI, Robots, Data Software Helping Create New Approach for Planning Cities of the Future 
Purdue University News in ACM
By Chris Adam

Purdue University researchers have developed a unique strategy to plan future cities, by streamlining building information modeling software through new approaches to data. Purdue's Jiansong Zhang said the methodology facilitates full software development based on data from industry foundation classes, "for any task in the life cycle of an [architecture, engineering, and construction] project." Zhang added that the researchers created a visualization program deployed via the new technique. Said Zhang, "The new method can help eliminate missing or inconsistent information during software development." The data encompasses all sectors, functions, and life-cycle stages of software development for construction projects. .... "

New Senses from States

The use of simple states would suggest we could do some interesting things here. 

Spidey Senses’ Could Help Autonomous Machines See Better
Purdue University News
By Kayla Wiles

Researchers at Purdue University, Nanyang Technology University in Singapore, and ETH Zurich in Switzerland have developed sensors inspired by spiders, bats, birds, and other animals to help autonomous machines navigate in dangerous environments and avoid accidents. In nature, once a particular level of force activates the mechanoreceptors of an animal, they compute information by switching from one state to another. The researchers designed their mechanosensors to do the same, and to use these on/off states to interpret signals. Said Purdue researcher Andres Arrieta, "With the help of machine learning algorithms, we could train these sensors to function autonomously with minimum energy consumption."

Shape Structuralizing for Design and More

Liked this, especially as it could be connected to supply chains,  procurement, compliance ....

Smart software tool could pave the way for changing how things get designed, made, and supplied      Purdue.edu

WEST LAFAYETTE, Ind. — If you don’t like wandering through big-box stores trying to find the furniture you need, and then struggle to assemble it, researchers have proposed a solution: Smart software that helps you design your own furniture, 3D print the joints and assemble the whole structure at home.

“This work has implications for changing how the supply chain works,” said Karthik Ramani, Purdue University’s Donald W. Feddersen Professor of Mechanical Engineering. “A tool like this turns consumers into producers.”

The researchers presented the tool, called “Shape Structuralizer,” at the 2019 ACM CHI Conference on Human Factors in Computing Systems in Glasgow, Scotland on Monday (May 6).

There’s one problem: The industries for such a tool – a way to pick up parts to 3D print your designs or have the parts shipped to you – don’t exist yet. ....

A YouTube video of the tool is available at https://www.youtube.com/watch?v=G1dSmxgI7wA&feature=youtu.be.  ... 

“We don’t know yet what form the ideas we have demonstrated will take when they’re available to consumers. But we’ve demonstrated one way that a tool can adapt to the user rather than the other way around,” Ramani said. “This would localize manufacturing and cut tedious steps between design and fabrication.”

Shape Structuralizer is a step toward creating these industries by making structure design “open-source,” similarly to how Apple provides a platform for anyone to make apps, or how Tesla made its patents on electric car technology freely available to the public.  ... " 

Purdue Hummingbird Drone

Flight bio-mimicry and the potential for swarming task completion.  AI being used to determine useful patterns supporting flight.

Tiny drone uses A.I. to learn from nature’s best pilot, the hummingbird   By Georgina Torbet

One of nature’s most remarkable creations is the hummingbird, which flaps its wings up to 80 times per second and which can hover in place and fly in any direction. Now scientists have used machine learning algorithms to study the way these birds fly in order to replicate their abilities in drones.

The robot, developed by researchers at Purdue University, has artificial intelligence (A.I.) which learns from hummingbird simulations and applies its findings to the movements of its flexible flapping wings.  .... " 

Securing the Body Internet

As the body internet continues to expand,   hacking of devices and cnnections is possible, , but has never been done.   The system of devices and connections can be hacked.     Now work at Purdue shows how the signals involved can be restricted.

Your body has internet—and now it can't be hacked
by Kayla Wiles, Purdue University

The human body carries electrical signals well. Now these signals can be contained within close proximity of the body, thanks to new technology developed by Purdue University engineers.... :  

Someone could hack into your pacemaker or insulin pump and potentially kill you, just by intercepting and analyzing wireless signals. This hasn't happened in real life yet, but researchers have been demonstrating for at least a decade that it's possible.

Before the first crime happens, Purdue University engineers have tightened security on the "internet of body." Now, the network you didn't know you had is only accessible by you and your devices, thanks to technology that keeps communication signals within the body itself.

The work appears in the journal Scientific Reports. Study authors include Shreyas Sen, an assistant professor of electrical and computer engineering at Purdue, and his students, Debayan Das, Shovan Maity and Baibhab Chatterjee.

"We're connecting more and more devices to the human body network, from smart watches and fitness trackers to head-mounted virtual reality displays," said Sen, who specializes in sensing and communication systems.

"The challenge has not only been keeping this communication within the body so that no one can intercept it, but also getting higher bandwidth and less battery consumption," he said.

Body fluids carry electrical signals very well. So far, so-called "body area networks" have used Bluetooth technology to send signals on and around the body. These electromagnetic waves can be picked up within at least a 10-meter radius of a person.  ... "  ...

Electronic IOT Stickers

Simple, cheap?.  Easily attachable and removable.  Sensing contextual environment.  Security?

Electronic stickers on a set of blocks. Electronic Stickers to Streamline Large-Scale 'Internet of Things'

Purdue University News     By Kayla Wiles

Purdue University and University of Virginia researchers have developed a fabrication method that makes tiny, thin-filmed electronic circuits that are peelable from a surface and which enable any object to sense its environment or to be controlled via the application of a high-tech sticker. The fabrication technique, called "transfer printing," reduces manufacturing costs by using a single wafer to build a nearly infinite number of thin films holding electronic circuits. The peeling is made possible by inserting a ductile metal layer, such as nickel, between the electronic film and a silicon wafer, and then submerging the entire device in water. These thin-film electronics can then be trimmed and pasted onto any surface, giving that object electronic features. "We've optimized this process so that we can delaminate electronic films from wafers in a defect-free manner," says Purdue's Chi Hwan Lee.   .... "