Researchers develop a pipeline to enable fast, accurate image processing for precision medicine
One of the main tools doctors use to detect diseases and injuries in cases ranging from multiple sclerosis to broken bones is magnetic resonance imaging (MRI). However, the results of an MRI scan take hours or days to interpret and analyze. This means that if a more detailed investigation is needed, or there is a problem with the scan, the patient needs to return for a follow-up.
A new, supercomputing-powered, real-time analysis system may change that.
Researchers from the Texas Advanced Computing Center (TACC), The University of Texas Health Science Center (UTHSC) and Philips Healthcare, have developed a new, automated platform capable of returning in-depth analyses of MRI scans in minutes, thereby minimizing patient callbacks, saving millions of dollars annually, and advancing precision medicine.
The team presented a proof-of-concept demonstration of the platform at the International Conference on Biomedical and Health Informatics this week in Orlando, Florida.
The platform they developed combines the imaging capabilities of the Philips MRI scanner with the processing power of the Stampede supercomputer – one of the fastest in the world – using the TACC-developed Agave API Platform infrastructure to facilitate communication, data transfer, and job control between the two.
An API, or Application Program Interface, is a set of protocols and tools that specify how software components should interact. Agave manages the execution of the computing jobs and handles the flow of data from site to site. It has been used for a range of problems, from plant genomics to molecular simulations, and allows researchers to access cyberinfrastructure resources like Stampede via the web.
“The Agave Platform brings the power of high-performance computing into the clinic,” said William (Joe) Allen, a life science researcher for TACC and lead author on the paper. “This gives radiologists and other clinical staff the means to provide real-time quality control, precision medicine, and overall better care to the patient.”
For their demonstration project, staff at UTHSC performed MRI scans on a patient with a cartilage disorder to assess the state of the disease. Data from the MRI was passed through a proxy server to Stampede where it ran the GRAPE (GRAphical Pipelines Environment) analysis tool. Created by researchers at UTHSC, GRAPE characterizes the scanned tissue and returns pertinent information that can be used to do adaptive scanning – essentially telling a clinician to look more closely at a region of interest, thus accelerating the discovery of pathologies.
The researchers demonstrated the system’s effectiveness using a T1 mapping process, which converts raw data to useful imagery. The transformation involves computationally-intensive data analyses and is therefore a reasonable demonstration of a typical workflow for real-time, quantitative MRI.
A full circuit, from MRI scan to supercomputer and back, took approximately five minutes to complete and was accomplished without any additional inputs or interventions. The system is designed to alert the scanner operator to redo a corrupted scan if the patient moves, or initiate additional scans as needed, while adding only minimal time to the overall scanning process.
“We are very excited by this fruitful collaboration with TACC,” said Refaat Gabr, an assistant professor of Diagnostic and Interventional Imaging at UTHSC and the lead researcher on the project. “By integrating the computational power of TACC, we plan to build a completely adaptive scan environment to study multiple sclerosis and other diseases.”
Ponnada Narayana, Gabr’s co-principal investigator and the director of Magnetic Resonance Research at The University of Texas Medical School at Houston, elaborated.
“Another potential of this technology is the extraction of quantitative, information-based texture analysis of MRI,” he said. “There are a few thousand textures that can be quantified on MRI. These textures can be combined using appropriate mathematical models for radiomics. Combining radiomics with genetic profiles, referred to as radiogenomics, has the potential to predict outcomes in a number diseases, including cancer, and is a cornerstone of precision medicine.”
According to Allen, “science as a service” platforms like Agave will enable doctors to capture many kinds of biomedical data in real time and turn them into actionable insights.
“Here, we demonstrated this is possible for MRI. But this same idea could be extended to virtually any medical device that gathers patient data,” he said. “In a world of big health data and an almost limitless capacity to compute, there is little reason not to leverage high-performance computing resources in the clinic.”
Learn more: REAL-TIME MRI ANALYSIS POWERED BY SUPERCOMPUTERS
Technology has promised to transform health care for years now. Multiple apps, devices, and other e-health approaches are being created to help the patient increase their awareness, education and accountability in their own health. In the not-so-distant future, technology will be able to continuously monitor, track and even diagnose a patient remotely.
“An overall trend in your health is very different from a single data point collected at a visit to a doctor’s office,” said Mark Benden, PhD, CPE, associate professor in the Department of Environmental and Occupational Health and director of the Ergonomics Center at the Texas A&M School of Public Health. “Knowing the trends will greatly improve both care and prevention.” In fact, according to a 2013 report, 73 percent of physicians think that health information technology will—at least in the long term—improve health care quality.
Technology may aid in taking a simple patient history. A wearable device can already show things like how many steps a patient is taking each day and their average heart rate, and at some point, they may also be able to measure disease markers or indicators like blood pressure, cholesterol or blood sugar. “Having information from these devices allows providers and patients to have a data-driven conversation, not one based on a one-time sample,” said Benden, who is a member of the Texas A&M Center for Remote Health Technologies and Systems. “Having objective data can also help with the natural tendency of patients telling their provider what they think they want to hear.”
For example, if a wearable device could accurately measure heart rate and blood pressure at every moment of the day, providers could keep this data as part of the person’s electronic health record. If someone’s blood pressure started to rise over time, the provider could consider prescribing a medication to bring it down to the healthy range and be confident that the rise was an actual trend, not a one-time high outlier.
Technology can help physicians make better decisions in other ways as well. Hongbin Wang, PhD, professor at the Texas A&M College of Medicine and co-director of the Texas A&M Biomedical Informatics Center, is working on a computer model of neurons to predict a decision—a medical diagnosis, for example—and illuminate any biases that might be present.
Wang’s work, and other applications of big data, may help with diagnosis by drawing together not only one person’s test results over time but also results from thousands or millions of other people. Data from many patients’ treatment outcomes may also help clinicians recommend the best treatment for each individual: the ultimate goal of precision medicine.
Although technology plays an important role in diagnosis and treatment, for Benden and other public health practitioners, it’s technology’s potential to aid in disease prevention that is most exciting. If exercise is one of the most effective methods of staving off diseases from cancer to heart disease to Alzheimer’s, the main challenge is motivating people to become active. Although fitness trackers were supposed to help, there’s little evidence that they make people more active over time. “We’re struggling to show that wearables are changing behaviors…what’s missing?” Benden asked. “I think we’re missing human connection.”
That human connection could be as simple as the provider receiving a notification about a shift in their patient’s trends, allowing the physician or nurse to follow up with a phone call to check in. Of course, as technology itself becomes more human-like, it may be able to motivate people on its own. “When we learn to use these devices in a way that responds to someone as a person and caters to their individual needs, it will be very powerful,” Benden said. “The technology will know you and be able to help you make healthy choices in whatever way works best for you personally.”
Someday technology may even allow patients to deal with less-complicated issues on their own—or possibly respond by itself. “Someday, the devices will be smart enough to know what’s happening to you and intervene when necessary,” Benden said, like a pacemaker that can help a heartbeat regularly while monitoring rhythms, and then if needed defibrillate automatically. “A lot of those corrections will be automatic, and people can continue about their days without ever knowing that a device just saved their lives.”
By coating tiny gel beads with lung-derived stem cells and then allowing them to self-assemble into the shapes of the air sacs found in human lungs, researchers at the Eli and Edythe Broad Center of Regenerative Medicine and Stem Cell Research at UCLA have succeeded in creating three-dimensional lung “organoids.” The laboratory-grown lung-like tissue can be used to study diseases including idiopathic pulmonary fibrosis, which has traditionally been difficult to study using conventional methods.
“While we haven’t built a fully functional lung, we’ve been able to take lung cells and place them in the correct geometrical spacing and pattern to mimic a human lung,” said Dr. Brigitte Gomperts, an associate professor of pediatric hematology/oncology and the study’s lead author.
Idiopathic pulmonary fibrosis is a chronic lung disease characterized by scarring of the lungs. The scarring makes the lungs thick and stiff, which over time results in progressively worsening shortness of breath and lack of oxygen to the brain and vital organs. After diagnosis, most people with the disease live about three to five years. Though researchers do not know what causes idiopathic pulmonary fibrosis in all cases, for a small percentage of people it runs in their families. Additionally, cigarette smoking and exposure to certain types of dust can increase the risk of developing the disease.
To study the effect of genetic mutations or drugs on lung cells, researchers have previously relied on two-dimensional cultures of the cells. But when they take cells from people with idiopathic pulmonary fibrosis and grow them on these flat cultures, the cells appear healthy. “Scientists have really not been able to model lung scarring in a dish,” said Gomperts, who is a member of the UCLA Broad Stem Cell Research Center. The inability to model idiopathic pulmonary fibrosis in the laboratory makes it difficult to study the biology of the disease and design possible treatments.
Gomperts and her colleagues started with stem cells created using cells from adult lungs. They used those cells to coat sticky hydrogel beads, and then they partitioned these beads into small wells, each only 7 millimeters across. Inside each well, the lung cells grew around the beads, which linked them and formed an evenly distributed three-dimensional pattern. To show that these tiny organoids mimicked the structure of actual lungs, the researchers compared the lab-grown tissues with real sections of human lung.
“The technique is very simple,” said Dan Wilkinson, a graduate student in the department of materials science and engineering and the paper’s first author. “We can make thousands of reproducible pieces of tissue that resemble lung and contain patient-specific cells.”
Moreover, when Wilkinson and Gomperts added certain molecular factors to the 3-D cultures, the lungs developed scars similar to those seen in the lungs of people who have idiopathic pulmonary fibrosis, something that could not be accomplished using two-dimensional cultures of these cells.
Using the new lung organoids, researchers will be able to study the biological underpinnings of lung diseases including idiopathic pulmonary fibrosis, and also test possible treatments for the diseases. To study an individual’s disease, or what drugs might work best in their case, clinicians could collect cells from the person, turn them into stem cells, coax those stem cells to differentiate into lung cells, then use those cells in 3-D cultures. Because it’s so easy to create many tiny organoids at once, researchers could screen the effect of many drugs. “This is the basis for precision medicine and personalized treatments,” Gomperts said.
Researchers are one step closer to understanding the genetic and biological basis of diseases like cancer, diabetes, Alzheimer’s and rheumatoid arthritis – and identifying new drug targets and therapies – thanks to work by three computational biology research teams from the University of Arizona Health Sciences, University of Pennsylvania and Vanderbilt University.
The researchers’ findings – a method demonstrating that independent DNA variants linked to a disease share similar biological properties – were published online in the April 27 edition of npj Genomic Medicine.
“The discovery of these shared properties offer the opportunity to broaden our understanding of the biological basis of disease and identify new therapeutic targets,” said Yves A. Lussier, MD, FACMI, lead and senior corresponding author of the study and UAHS associate vice president for health sciences and director of the UAHS Center for Biomedical Informatics and Biostatistics (CB2).
The researchers are striving to better understand the common genetic and biological backgrounds that make certain people susceptible to the same disease. They have developed a method to demonstrate how individual, disease-associated DNA variants share similar biological properties that provide a road map for disease origin.
Over the last ten years, genetics researchers have conducted large studies, called Genome Wide Association Studies (GWAS), which analyze DNA variants across thousands of human genomes to identify those that are more frequent in people with a disease. However, the impact of many of these disease-associated variants on the function and regulation of genes remains elusive, making clinical interpretation difficult.
A method to explore the biological impact of these variants and how they are linked to disease was developed through the collaboration of bioinformatics and systems biology researchers Dr. Lussier;Haiquan Li, PhD, research associate professor and director for translational bioinformatics, Department of Medicine, UA College of Medicine – Tucson; Ikbel Achour, PhD, director for precision health, CB2; Jason H. Moore, PhD, director, Institute for Biomedical Informatics, Perelman School of Medicine, University of Pennsylvania; and Joshua C. Denny, MD, MS, FACMI, associate professor of biomedical informatics and medicine, Vanderbilt University, along with their teams.
In their new paper, the researchers demonstrate that DNA risk variants can affect biological activities such as gene expression and cellular machinery, which together provide a more comprehensive picture of disease biology. When DNA risk variants for a given disease were analyzed in combination, similar biological activities were discovered, suggesting that distinct risk variants can affect the same or shared biological functions and thus cause the same disease. More detailed analyses of variants linked to bladder cancer, Alzheimer’s disease and rheumatoid arthritis showed that two variants can contribute to disease independently, but also interact genetically. Therefore, the precise combination of DNA variants of a patient may work to increase or decrease the relative risk of disease.