IoT In The Lab Includes Digital Cages And Instrument Sensors
By Deborah Borfitz
April 17, 2020 | The cost and complexity of proving out discoveries in the lab on human study subjects has given rise to a sensor revolution for the monitoring of everything from the environment and health of mice to the temperature and humidity levels in freezers, fridges, and incubators. Internet of Things (IoT) measurement systems are effectively catapulting animal studies out of the horse-and-buggy era and eliminating preventable holdups due to instruments being out of date or incorrectly calibrated, according to IoT enthusiasts working in an assortment of lab settings.
Nicholas Hertz, Ph.D., co-founder and chief scientific officer at biotechnology company Mitokinin, had the digital vivarium system of Vium installed about two years ago to record minute changes in mice that are indicative of neurodegenerative disease but would otherwise be undetectable. Many promising findings in the lab aren’t translating into treatments for people because traditional measurement platforms are unable to “tease out very subtle phenotypes,” he says.
As described by a representative at Vium, the digital vivarium system employs sensors and computer vision to produce data on animals without having to move them out of their home cage. The platform provides continuous real-time measurement of motion, breathing rate, and percent time running on a wheel.
The collected biomarkers get analyzed in the cloud before being sent back to researchers in real time, the spokesperson continues. When the company started in 2015, it functioned as a contract research organization but interest in the technology quickly swelled. As of the end of 2019, five of the top 10 pharma companies were clients of Vium.
The system allows mice to be monitored around-the-clock and not just during the day when it’s most convenient for their human handlers, says Hertz. Mice are nocturnal, so taking measurements on them when they’re supposed to be sleeping can skew results. The mice are also placed into the digital cages well before experiments begin so they can get acclimated to the system, he adds.
Among the ways Mitokinin has used the digital vivarium system is to monitor—via a web portal and 24-hour live camera streaming—the effect of drugs on mice, including whether they appear healthy or sick or are experiencing any severe defects in behavior, Hertz says. The seven scientists using the system can log in from home on the weekend and, if the mice are doing well, save themselves a trip to the lab.
Beyond the convenience factor, the digital cage system facilitates behavior analysis for neurodegeneration using the rotarod test, one of the oldest used means of assessing the effects of a drug on animal behavior, says Hertz. It measures how long it takes for the mice to fall off a circular rod turning at a constant or increasing speed.
The problem with the rotarod test is that results can be hard to reproduce even a week later, Hertz says. “If a mouse thinks it’s going to going to fall off this rotating rod and die then it will of course try to stay on as stringently as possible, but if it knows there’s a soft pad below to land on it may not hold on at all.” Mice also tend to get tested during the day when they’re accustomed to sleeping, which doesn’t produce an accurate picture of their behavior, he adds.
Since Vium monitors mice 24/7, subtleties of their behavior can be more easily detected, he explains. This is one reason some academic groups have tried to create similar 24-hour monitoring systems on their own for years.
The beauty of working with Vium is that its platform is “iterative,” says Hertz. The company also helps with data analysis, providing a rationale for the phenotypes Mitokinin researchers detect.
Mitokinin is using the digital vivarium system several different ways, including continuous behavioral measurement for long-term, longitudinal animal studies lasting about six months, says Hertz. Alternately, it’s used for acute studies where the mice might be quite sick and need to be closely watched “for the moment their bodies are failing.”
One valued feature of the system is that it can measure breathing rate, Hertz continues. An alert is triggered if a mouse has a slow breathing rate so researchers can immediately provide needed care.
The long-term animal studies are for neurodegenerative diseases such as Parkinson’s and Alzheimer’s, he says. Sleep and activity trackers are monitoring the same behaviors in their human research counterparts and the results from the devices tend to differ from self-reporting. The Vium system can help remove that sort of bias in mouse studies by allowing for continuous sleep-and-activity tracking.
Aha moments with the Vium system are commonplace, says Hertz. For example, researchers now know that different neurotransmitters, and not just dopamine-producing ones, inhibit genetic expression of the hyperactivity phenotype. That would have been difficult to discover using traditional behavior quantification methods.
Any research lab that does mouse work could potentially benefit from using the digital cages, Hertz says. This would include cancer researchers, who might want to monitor survival and tumor shrinkage as well as behavioral changes suggestive of improved quality of life in mice getting an experimental drug. But it is especially well suited to aging-related studies to reveal minute changes in mobility.
Hertz says he is aware of no other such commercial system but is a big fan of Vium. “When we’ve had trouble analyzing data, they’d put one of their best scientists on it.” While it is customary to wait six months to see drug effects, detectable changes were found after only two weeks with one of the animal disease models developed with Vium, he notes.
Peace Of Mind
At Repertoire Immune Medicines (previously Cogen Immune Medicines), the IoT in the lab is for ongoing monitoring of temperature and other ambient conditions in freezers, fridges, and incubators. “If they’re not accurate, everything we do can go literally to waste,” says Lab Operations Manager Demet Aybar, a chemist by training.
Depending on the cell lines being grown and stored, the “right” temperature might be as low as minus 140 degrees Celsius (cryogenic freezer) to a steady 37 degrees (CO2 incubators). Protein and assay kits also need to stay in precise storage temperatures.
A trio of sensors provided by Elemental Machines helps Aybar ensure that favorable conditions are maintained, she says. That’s no small feat, given that the responsibility covers 100 machines on two floors occupying 34,000 square feet of space in Cambridge, Massachusetts. “We have two more companies within our facility that we are monitoring, and I am overseeing their day-to-day operations as well,” she explains.
The sensors utilize copper-based thermocouples that read the analog temperature in the machines, digitalize them and transfer the data via a Bluetooth gateway to the cloud, and then return the metadata to researchers on a dashboard with actionable insights, Aybar says. The Element-A sensor monitors the environment (humidity) and Element-T the temperature, while Element-D connects OEM instruments to the cloud to automate the collection of maintenance and calibration records and critical performance and utilization metrics.
Following the recent merger that created Repertoire Immune Medicines, the same system was also implemented at a second site in Cambridge that also has a lot of instruments but all of them are older models without a built-in communication board. Worry quickly turned to relief when the solution turned out to be a call to a few manufacturing companies to have the boards installed at minimal cost.
Advantages of using the sensors are twofold, Aybar says. They give her “peace of mind” that she will be immediately notified by computer (alerts by phone are not yet an option) if anything is amiss, “even a tiny bit of a temperature change.” That translates into financial savings from temperature failings for frequently used, high-cost items such as assay kits and cell lines. “It’s light, very user-friendly technology and I encourage all of my colleagues to use it.”
Later this year, Aybar says Elemental Machines is expected to release sensor fusion technology using artificial intelligence that will provide additional metrics such as compression cycles and how often doors on machines open and close—and to predict when, and why, instruments are going to have a problem.
As far as Aybar is concerned, Elemental Machines is fast approaching the capabilities of Thermo Scientific’s high-end InSight Analytics, powered by KLATU but at a more attractive price point. She has previous experience with Coris sensor technology, she says, but it offered no practical way to get the probe in tissue culture incubators to take readings.
According to an Elemental Machines representative, the company has a library of nearly 200 instruments that it can connect to. Its internal market research indicates that in the U.S. alone more than 200 million scientific instruments are not directly connected to the internet at present.