By Paul Nicolaus
April 11, 2018 | It’s an understatement to say that smartphones have become widespread. Today, the global human population stands at over 7 billion, according to the U.S. Census Bureau. Meanwhile, the number of mobile connections circulating on Earth has climbed over 8 billion. And some believe this prevalence of handheld tech tools could help tackle some of the world’s peskiest problems.
Infectious diseases like chikungunya, dengue, and Zika, for example, continue to spread from mosquitos to humans and often occur in parts of the world without the type of basic infrastructure that exists in modern clinical diagnostic settings. Because these illnesses are carried by the same type of mosquito and share similar early symptoms, they can be easily misdiagnosed.
Fast and accurate diagnosis is needed to determine the type of infection, move forward with appropriate treatment, and better contain these mosquito-borne illnesses. The challenge is that traditional methods require large, expensive tools that can take days to produce results. Polymerase chain reaction (PCR) testing involves shipping a sample to a lab, extracting DNA or RNA, and amplifying the genetic material to an amount that can be detected.
The process involves raising and lowering the temperature of the sample many times over, which takes both machines and power, and the complexity and expense have made it difficult to move related devices beyond the walls of a lab. But smartphone-based detection tools are offering new promise by tapping into the power of handheld technology that already exists across the globe.
Digitalizing the Process
UK-based smartphone diagnostics specialist Novarum DX (part of BBI Solutions) has developed an app that turns a smartphone into a diagnostic reader and provides test results from the point of care. The company recently announced it has been granted a patent for its smartphone diagnostic technology in the United States, which builds upon previous patent success in countries like China, Japan, and Russia.
“What we do is we turn a smartphone into a reader that will recognize and measure the results of a diagnostic test,” explained Novarum founder Neil Polwart. “So instead of relying on human eyes, which are all different and variable and see what they want to see, we take all of that subjectivity and ambiguity out of the equation and use the camera on a smartphone.”
From a user perspective, the process is similar to scanning a QR code. “You see a template on the screen, you hold it over the test roughly lined up but not perfectly lined up, and then we will capture that test information, correct for the fact that you’re not holding the phone in exactly the right position, and then extract the data that’s there to make a measurement from it,” he said.
Additional information can be recorded as well, such as the sample ID, the patient ID, the GPS coordinates of where the result was recorded, the date, or the time. This can be either automatically collected, or the user can enter it in. And once it’s on the phone, he added, the information can go anywhere in the world.
Because the image processing is performed directly in the app, it can work without an Internet connection and without running up a massive data bill, he explained. Collecting additional information (such as date, time, and sample ID) works without a connection, too, and if users want a database of authorized users or patients, these can be cached locally on the device when there is a connection and become available off-line later.
“We even go one step further than this,” he noted. If there is a result that needs to be sent elsewhere as quickly as possible as well as more detailed information that would take up more bandwidth—like the test image or the profile used for analysis—this data payload can be split in two. The urgent bit can be included in a small package of data sent as soon as there is a connection of any type, and the more detailed bit sent only when there is a WiFi connection. Those two sets of data can then be matched up when they arrive at company servers.
This Novarum technology is predominantly being used with lateral flow tests that look for the presence of lines, but the company has also done work measuring different colors. More recently, there have been efforts to read data from digital screens. “So basically anything where you can read the result by eye,” Polwart said, “we’ll try and read it with a phone instead” in order to make it easier for the user and gain the benefits of connectivity.
In terms of mosquito-borne diseases, the technology can be used in conjunction with a test to capture the data and record that result. “They’re often being performed in relatively resource-poor settings where perhaps the person performing the test doesn’t have as high of an educational standard as the equivalent person performing the test in the U.S. or the UK might,” he explained, “so we can help provide pictorial instructions—we call it test choreography—in an app as well.”
Rather than relying on someone to follow a set of complicated instructions, pictorial or even animated instructions show how to collect the sample and handle any preparation. Because the test itself needs to run for several minutes, a timer is built into the app, too. Together, these measures are intended to reduce human error and lead to a better end result. On top of that, recording where that result was captured can help build maps that identify where the greatest areas of infection are located.
“We’re very keen that people have something very portable that fits in their pocket,” he said. Ideally, it’s actually the phone a person already owns so that there is no additional hardware cost. Because there is nothing extra to carry around, it is possible to put a packet of tests and a phone in your pocket and go visit the patient without carrying any extra equipment.
“When you see some of the healthcare workers in remote settings traveling by bicycle to get to patients, then you can start to see that even something as small as a laptop computer becomes an extra burden for them to carry,” Polwart said. In this sense, using something they have with them anyway is a solution that fits into this realm of remote or low-resource settings. “It’s about trying to bring the maximum benefit with the least inconvenience to the user,” he added.
New Tools Emerging
Other tools, while not yet commercially available, are seeking to move away from the heating and cooling associated with traditional PCR methods. Researchers at Sandia National Laboratories a multi-mission laboratory operated by National Technology and Engineering Solutions of Sandia LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration, have developed a mobile platform that uses loop-mediated isothermal amplification (LAMP) to indicate traces of a virus when blood, urine, or saliva samples are heated. If a sample contains a virus, the molecules glow a certain color following the amplification process. And it is possible to have multiple viruses glow with different colors.
Within about half an hour, the smartphone detects the color and intensity of the LAMP assay in order to determine whether or not the sample contains a virus. In that sense, LAMP can detect viruses down to copy number almost as good as PCR, according to Aashish Priye, a postdoctoral scientist at Sandia National Laboratories. “I say almost because it is generally one log less sensitive than PCR,” he noted. As a proof of concept, the platform was used to detect the presence of chikungunya, dengue, and Zika virus, and the results have been published in Scientific Reports (doi:10.1038/srep44778).
LAMP does have its limitations, including the decreased sensitivity compared with PCR and the complex biochemistry involved. And while it can be useful to know whether a virus is present or not, Priye acknowledged that a yes/no result is insufficient when a researcher needs to determine a virus’s stage. Despite these limitations, he sees the upside of going this route.
There are many commercial systems that handle point-of-care diagnostics, he explained, and most of them are PCR-based. There are now efforts underway to make portable PCR machines that can be carried in your hand and used in areas without medical facilities, but Priye argues that the complexity and cost involved still means that these types of solutions are not suitable for widespread distribution in low-resource environments. A smartphone-based diagnostic is different from other point-of-care diagnostics in the sense that all of the hardware and the software are bought right off the shelf and then integrated with the amplification chemistry of LAMP.
“The reason we did the smartphone is because we wanted to make it inexpensive,” he said. Imagine a setting that’s off-grid and without hospitals or medical facilities, Priye explained. A smartphone can take images of blood, analyze it, and send the information back to a lab where technical professionals can determine whether that area is at risk of a viral epidemic. A decade ago this was not possible, but now millions of people have powerful cameras in their pockets. “We’re trying to harness that inexpensive, already present camera on the smartphone,” he said.
Meanwhile, a multidisciplinary group of researchers with the University of Illinois at Urbana-Champaign and the University of Washington at Tacoma has developed a platform to diagnose infectious disease at the point-of-care using of a smartphone as the detection tool. The method is similar to Sandia’s, explained Brian Cunningham, University of Illinois electrical and computer engineering professor, but does differ in a couple of notable ways. Sandia’s assay is conducted in plastic tubes, he pointed out, and the national lab has its own proprietary variant of the LAMP assay. Cunningham and colleagues, on the other hand, have made use of a microfluidic chip.
The technology uses a test cartridge that looks like a credit card, and a cradle to hold the phone in place. Inside of the card is a silicon microfluidic chip that has, printed within it, the nucleic acid reagents, or “primers,” required for detecting specific DNA sequences associated with pathogens. “So we analyze the genome of the pathogens and then look for unique stretches of DNA within their sequence that we can target with one of these primers,” he said. “In that way, we can have separate tests for several viral or bacterial pathogens and then test for all of them at the same time with a single droplet of test sample.”
The smartphone cradle positions the camera to take a fluorescent image of that chip. Inside the cradle is a set of light-emitting diodes that bathe the chip in blue light, and a positive test result for one of the pathogens turns areas within the chip bright green. This green glow is captured by the phone’s camera, and image processing software roughly quantifies the amount of green fluorescent to make a determination as to whether each of those pathogens is present.
There are two distinct generations of cradle and chip that the researchers have demonstrated, according to Cunningham. One iteration allowed for a yes/no response for the presence of up to eight pathogens. In this chip, the test sample has been processed off-chip using a commercially available sample preparation kit. A different system has been used to process a whole blood sample and then estimate the concentration of the virus present for up to four pathogens at a time in about 30 minutes. In this system, the microfluidic cartridge has a second stage; the first stage is for lysing a whole blood sample, and the second is for detection.
It is this second system that has been used to detect and quantify the presence of chikungunya, dengue, and Zika virus, and those findings have been published in Biomedical Microdevices. (doi: 10.1007/s10544-017-0209-9). “We did not invent the LAMP process,” Cunningham explained, “but we needed to identify and test novel LAMP primers for our tests. We designed and built the phone’s cradle and the microfludiic cartridges used for our work.”
The researchers measured the test samples using both the conventional lab method and their smartphone-based system, which yielded results that were nearly equivalent for each of the viruses, he said, both in terms of the sensitivity and the selectivity. “But then, of course, we’re doing the measurement with a small, handheld, and inexpensive instrument,” he added.
Smartphones have enough “computational horsepower” in them nowadays that they can do real image processing, Cunningham said. The capabilities of smartphone cameras have been engineered to the point that their pixel resolution and low-light capabilities can now produce results “that are equivalent to those taken with laboratory instruments.” It’s not essential that a smartphone camera be used, he added, “but since it’s already an integral part of the instrument that does communication and computation so well, we can provide that capability at very low cost in terms of the hardware.”
He sees the reduced turnaround time as a significant advantage of mobile systems used for pathogen detection. When the process involves collecting a sample, sending it to a laboratory, and then waiting for those results to come back, infectious disease has the opportunity to spread to others during that lag time or patients can be lost to follow up. A faster test result, and the ability to share that information using cloud-based service systems, can lead to faster reaction time and improved disease management.
Toward Commercialization
Cunningham noted a number of efforts to move forward since his research group’s findings were published. “There are several engineering improvements and image processing improvements that we’re incorporating,” he said, but declined to share specifics. “We are working with an eye toward making the limits of detection even lower than they currently are” and making the microfluidic cartridge more automated and easier to use.
The work conducted to date has been carried out through the efforts of graduate and undergraduate students with support from a National Science Foundation (NSF) grant that has funded several prototypes. “We have a team of software engineers who are writing an app and trying to make it user-friendly in connection with cloud-based service systems,” he said, “so I would say at this stage it’s a working prototype” with capabilities that are still being developed and demonstrated for additional uses.
Austin, Texas-based Reliant Immune Diagnostics took a license to the patents and the pending applications last year, he said, and the company has been working to raise investment dollars needed to pursue the development of a commercial product for health applications.
Similarly, since the Scientific Reports paper came out last year, Priye and colleagues have revised their prototype. And he sees room for improvement in a number of areas, such as the chemistry behind the amplification reactions and the timer design. There is a need to design six primers to amplify the target DNA with LAMP, and because they are fixed these primers interact in very complicated ways. “We have to come up with an optimized way of designing these primers so they interact with themselves less and with the target more,” he said.
The reactions are handled in custom-made wells, and this is yet another element that could use further development. “I think the housing for the reagents of those wells can be further optimized to enable better detection,” he said. “We can come up with better science so that there are no gases, there are no bubbles, and so the imaging and detection can be easier.”
Sandia has entered into a contract with Maxim Biotech, Inc., Priye added, and the company is working toward optimizing the platform and taking it to the next stage of commercialization.
“Smartphones are ubiquitous,” said Priye. There are smartphones in places in West Africa and India, for example, where people don’t even have toilets and clean water, so this is a device that is available to the rural-most area. “I think it’s time that these devices move out of the laboratory,” he said, “and find the places where they’re needed the most.”
Paul Nicolaus is a freelance writer specializing in science, technology, and health. Learn more at www.nicolauswriting.com