International research team that developed swarm learning believe it could ‘significantly promote and accelerate collaboration and information exchange in research, especially in the field of medicine’
“Swarm Learning” is a technology that enables cross-site analysis of population health data while maintaining patient privacy protocols to generate improvements in precision medicine. That’s the goal described by an international team of scientists who used this approach to develop artificial intelligence (AI) algorithms that seek out and identify lung disease, blood cancer, and COVID-19 data stored in disparate databases.
Since 80% of patient records feature clinical laboratory test results, there’s no doubt this protected health information (PHI) would be curated by the swarm learning algorithms.
In their study they wrote, “Fast and reliable detection of patients with severe and heterogeneous illnesses is a major goal of precision medicine. … However, there is an increasing divide between what is technically possible and what is allowed, because of privacy legislation. Here, to facilitate the integration of any medical data from any data owner worldwide without violating privacy laws, we introduce Swarm Learning—a decentralized machine-learning approach that unites edge computing, blockchain-based peer-to-peer networking, and coordination while maintaining confidentiality without the need for a central coordinator, thereby going beyond federated learning.”
What is Swarm Learning?
Swarm Learning is a way to collaborate and share medical research toward a goal of advancing precision medicine, the researchers stated.
The technology blends AI with blockchain-based peer-to-peer networking to create information exchange across a network, the DZNE news release explained. The machine learning algorithms are “trained” to detect data patterns “and recognize the learned patterns in other data as well,” the news release noted.
“Medical research data are a treasure. They can play a decisive role in developing personalized therapies that are tailored to each individual more precisely than conventional treatments,” said Joachim Schultze, MD (above), Director, Systems Medicine at DZNE and Professor, Life and Medical Sciences Institute at the University of Bonn, in the news release. “It’s critical for science to be able to use such data as comprehensively and from as many sources as possible,” he added. This, of course, would include clinical laboratory test results data. (Photo copyright: University of Bonn.)
Since, as Dark Daily has reported many times, clinical laboratory test data comprises as much as 80% of patients’ medical records, such a treasure trove of information will most likely include medical laboratory test data as well as reports on patient diagnoses, demographics, and medical history. Swarm learning incorporating laboratory test results may inform medical researchers in their population health analyses.
“The key is that all participants can learn from each other without the need of sharing confidential information,” said Eng Lim Goh, PhD, Senior Vice President and Chief Technology Officer for AI at Hewlett Packard Enterprise (HPE), which developed base technology for swarm learning, according to the news release.
An HPE blog post notes that “Using swarm learning, the hospital can combine its data with that of hospitals serving different demographics in other regions and then use a private blockchain to learn from a global average, or parameter, of results—without sharing actual patient information.
“Under this model,” the blog continues, “‘each hospital is able to predict, with accuracy and with reduced bias, as though [it has] collected all the patient data globally in one place and learned from it,’ Goh says.”
Swarm Learning Applied in Study
The researchers studied four infectious and non-infectious diseases:
They used 16,400 transcriptomes from 127 clinical studies and assessed 95,000 X-ray images.
Data for transcriptomes were distributed over three to 32 blockchain nodes and across three nodes for X-rays.
The researchers “fed their algorithms with subsets of the respective data set” (such as those coming from people with disease versus healthy individuals), the news release noted.
Findings included:
90% algorithm accuracy in reporting on healthy people versus those diagnosed with diseases for transcriptomes.
76% to 86% algorithm accuracy in reporting of X-ray data.
Methodology worked best for leukemia.
Accuracy also was “very high” for tuberculosis and COVID-19.
X-ray data accuracy rate was lower, researchers said, due to less available data or image quality.
“Our study thus proves that swarm learning can be successfully applied to very different data. In principle, this applies to any type of information for which pattern recognition by means of artificial intelligence is useful. Be it genome data, X-ray images, data from brain imaging, or other complex data,” Schultze said in the DZNE news release.
The scientists say hospitals as well as research institutions may join or form swarms. So, hospital-based medical laboratory leaders and pathology groups may have an opportunity to contribute to swarm learning. According to Schultze, sharing information can go a long way toward “making the wealth of experience in medicine more accessible worldwide.”
Scientists worldwide engaged in research to develop a biomarker for dementia are predicting success, though some say additional research will be needed
Could a blood test for Alzheimer’s disease soon be on clinical laboratory test menus nationwide? Perhaps so. A recent Associated Press (AP) article that was picked up by NBC News and other healthcare publications reported that experimental test results presented during the Alzheimer’s Association International Conference (AAIC) in July suggest the Holy Grail of dementia tests—one where the specimen can be collected in a doctor’s office during a routine screening exam—may be close at hand.
The AP story noted that “half a dozen research groups gave new results on various experimental tests, including one that seems 88% accurate at indicating Alzheimer’s risk.” And Richard Hodes, MD, Director of the National Institute on Aging, told AP, “In the past year, we’ve seen a dramatic acceleration in progress [on Alzheimer’s tests]. This has happened at a pace that is far faster than any of us would have expected.”
This could be a boon for medical laboratories seeking way to contribute more value to patient care. Especially among Alzheimer’s patients, who account for as many as 70% of all dementia cases.
Plasma Biomarker for Predicting Alzheimer’s
One of the experimental blood tests presented at the AAIC involved a 2018 study into “the potential clinical utility of plasma biomarkers in predicting brain amyloid-β burden at an individual level. These plasma biomarkers also have cost-benefit and scalability advantages over current techniques, potentially enabling broader clinical access and efficient population screening,” the researchers stated an article they published in Nature.
AP also reported that Japanese scientists at the AAIC
presented results of a validation test conducted on 201 people who had either
Alzheimer’s, other types of dementia, or little or no symptoms. They found that
the test “correctly identified 92% of people who had Alzheimer’s and correctly
ruled out 85% who did not have it, for an overall accuracy of 88%.”
Akinori Nakamura, MD, PhD, of the National Center for
Geriatrics and Gerontology in Obu, Japan, was a member of the research team and
first author of the research paper. He told the AP that the test results “closely
matched those from the top tests used now—three types of brain scans and a
mental assessment exam.”
Eric McDade, DO (above), Associate Professor of Neurology at Washington University in St. Louis, told Neurology Today, “The results reported here provide a relatively high level of confidence given that this is a relatively well characterized population with an amyloid PET scan to provide confirmation of a significant level of amyloid plaque burden in the brain.” Could this level of physician confidence lead to a clinical laboratory test based on the plasma biomarker? (Photo copyright: Washington University.)
Koichi Tanaka is a Japanese engineer who won the Nobel prize winner for chemistry. He heads the Koichi Tanaka Research Lab at Shimadzu Corp. (OTCMKTS:SHMZF) in Kyoto, Japan, and was on the team that developed the Amyloid beta biomarker test that was presented at AAIC. He told Bloomberg, “Our finding overturned the common belief that it wouldn’t be possible to estimate amyloid accumulation in the brain from blood. We’re now being chased by others, and the competition is intensifying.”
But Tanaka cautions that the test needs further study before
it is ready for clinical use, and that for now “it belongs in the hands of drug
developers and research laboratories,” Bloomberg reported.
Other Studies into Developing an Alzheimer’s Biomarker
Alzheimer’s is usually diagnosed after symptoms appear, such
as memory loss. To arrive at their diagnoses, doctors often rely on medical
history, brain imaging (MRI, CT), PET, and measurement of amyloid in spinal
fluid.
An article published on Alzforum, a website and news service dedicated to the research and treatment for Alzheimer’s and other related disorders, noted a study by King’s College London researchers who, using mass spectrometry, “found a panel of biomarkers that predicted with almost 90% accuracy whether cognitively normal people had a positive amyloid scan.”
Nicholas Ashton, PhD, neuroscientist and Wallenberg Postdoctoral Fellow at University of Gothenburg in Sweden, and first author of the King’s College study, explained that “Amyloid-burden and neurofilament light polypeptide (NFL) peptides were important in predicting Alzheimer’s, but alone they weren’t as predictable as when we combined them with novel proteins related to amyloid PET.”
The researchers published their study earlier this year in Science Advances. “Using an unbiased mass spectrometry approach, we have found and replicated with high accuracy, specificity, and sensitivity a plasma protein classifier reflecting amyloid-beta burden in a cognitively unimpaired cohort,” the researchers wrote.
Meanwhile, researchers at Washington University School of Medicine St. Louis, along with the German Center for Neurodegenerative Diseases, a member of the Helmholtz Association, stated in a news release that a blood test they developed works by detecting leaks of NFL before the onset of symptoms. When the protein is found in cerebrospinal fluid, it could be a sign that Alzheimer’s may develop, as well as point to other neurodegenerative conditions such as multiple sclerosis, brain injury, or stroke, the researchers stated.
“This is something that would be easy to incorporate into a screening test in a neurology clinic,” Brian Gordon, PhD, Assistant Professor of Radiology at Washington University’s Mallinckrodt Institute of Radiology, and an author of the study, stated in the news release.
These parallel studies into screening for Alzheimer’s by
researchers worldwide are intriguing. The favorable results suggest that
someday there may be a screen for Alzheimer’s using a clinical laboratory blood
test.
With Alzheimer’s affecting nearly six million Americans of all ages, such an assay would enable clinical laboratories to help many people.
