Use of artificial intelligence in clinical laboratory testing could improve the diagnosis of cancer worldwide
In a proof of concept study, scientists at Shanghai Jiao Tong University in China have developed a clinical laboratory test that utilizes artificial intelligence (AI) to diagnose three types of cancer from a single drop of dried blood. The paper-based test was able to identify patients with colorectal, gastric, and pancreatic cancers and distinguish between patients with and without cancer.
The team’s goal was to develop a way to diagnose cancer while the disease is still in the earlier stages, especially in rural areas.
“Over a billion people across the world experience a high rate of missed disease diagnosis, an issue that highlights the need for diagnostic tools showing increased accuracy and affordability. In addition, such tools could be used in ecologically fragile and energy-limited regions, pointing to the need for developing solutions that can maximize health gains under limited resources for enhanced sustainability,” the researchers wrote in an article published in the journal Nature Sustainability titled, “A Sustainable Approach to Universal Metabolic Cancer Diagnosis.”
The researchers determined that by using less than 0.05 millimeters of dried blood, their test could accurately and quickly identify if a patient had cancer between 82% to 100% of the time.
According to Chaoyuan Kuang, MD, PhD (above), an oncologist at Montefiore Health System and assistant professor at the Albert Einstein College of Medicine, unlike liquid blood, dried serum can be “collected, stored, and transported at much lower cost and with much simpler equipment,” Live Science reported. “This could help democratize the availability of cancer early detection testing across the world,” he added. A paper-based clinical laboratory test that can detect and distinguish one cancer type from another would be a boon to cancer diagnosis worldwide. (Photo copyright: Albert Einstein College of Medicine.)
Improving Cancer Screening in Rural Areas
An earlier study conducted in China in 2022 examined results from 1,570 cancer survivors from both urban and rural areas of China. That study showed that 84.1% of the patients were diagnosed with cancer only after developing symptoms and that urban patients were more likely to be diagnosed in the early stages of cancer. In addition, rural patients also had less screening and treatment options available to them.
The researchers in this latest Chinese study tested their AI model on blood donors with and without cancer and compared the results to traditional liquid-blood biopsy tests.
“Based on modeling they performed, they reported the new tool could reduce the estimated proportion of undiagnosed cases of pancreatic, gastric, and colorectal cancers by about 20% to 50% if it was used for population-level cancer screening in rural China,” Live Science reported.
The scientists used dried serum spots (DSS) and machine learning to perform the research. According to their Nature Sustainability paper, DSS can be challenging in cancer research because sensitive biomarkers in the samples are often degraded or have inadequate amount of blood for proper analysis. To circumvent these issues, the researchers used nanoparticle-enhanced laser desorption/ionization mass spectrometry (NPELDI MS) to increase reliability and sensitivity. Inorganic nanoparticles were applied to the samples to strengthen selectivity and refine metabolic compounds from the samples.
However, the study authors noted that “the adaptation of NPELDI MS to dried spot analysis has not been validated,” Interesting Engineering reported.
A ‘Great Start’
The machine learning algorithm the Chinese scientists created demonstrates that DSS samples can be used to preserve important biological markers and could be beneficial in the diagnosis of cancer.
Their research indicated an overall reduction rate of undiagnosed cancers in the range of 20.35% to 55.10%. The researchers estimated the implementation of their AI tool could reduce the proportion of specific undiagnosed cancer cases in rural China by:
84.30% to 29.20% for colorectal cancer,
77.57% to 57.22% for gastric cancer, and
34.56% to 9.30% for pancreatic cancer.
It’s a “great start,” Chaoyuan Kuang, MD, PhD, an oncologist at Montefiore Health System and assistant professor at the Albert Einstein College of Medicine told Live Science. “This cancer test won’t enter use for a long time,” he said. Nevertheless, the potential of the tool is “immense,” he added, but that “we are still years away from being able to offer this test to patients.
“With further development, it could theoretically be used for the early detection of other types of cancer or for other diseases, or to monitor the progression of disease in patients who have already been diagnosed,” he noted.
Further research and clinical trials are needed before this AI tool can be used in a clinical diagnostic setting. This study is another example of researchers looking for cancer biomarkers in specimen types that are not tissue and further supports the hope that machine learning may one day detect cancer in earlier stages, increase survival rates, and save healthcare costs.
One factor motivating this type of research in China is the fact that the nation has more than 36,000 hospitals and approximately 20,000 anatomic pathologists. Of this total, only a minority of these pathologists have been trained to the standards of North America and Northern Europe.
Like other nations, China’s demand for subspecialist pathology services outstrips its supply of such pathologists. This is the reason why researchers in that country want to develop diagnostic assays for cancer and other diseases that are faster, cheaper, and comparable to a human pathologist in accuracy.
Scientists turned to metabolomics to find cause of biological aging and release index of 25 metabolites that predict healthy and rapid agers
Researchers at the University of Pittsburg Medical Center and the University of Pittsburgh School of Medicine have identified biomarkers in human blood which appear to affect biological aging (aka, senescence). Since biological aging is connected to a person’s overall condition, further research and studies confirming UPMC’s findings will likely lead to a new panel of tests clinical laboratories can run to support physicians’ assessment of their patients’ health.
UPMC’s research “points to pathways and compounds that may underlie biological age, shedding light on why people age differently and suggesting novel targets for interventions that could slow aging and promote health span, the length of time a person is healthy,” according to a UPMC news release.
“We decided to look at metabolites because they’re very dynamic,” Aditi Gurkar, PhD, the study’s senior author, told the Pittsburgh Post-Gazette. Gurkar is Assistant Professor of Medicine, Division of Geriatric Medicine, Aging Institute at the University of Pittsburg. “They can change because of the diet, they can change because of exercise, they can change because of lifestyle changes like smoking,” she added.
The scientists identified 25 metabolites that “showed clear differences” in the metabolomes of both healthy and rapid agers. Based on those findings, the researchers developed the Healthy Aging Metabolic (HAM) Index, a panel of metabolites that predicted healthy agers regardless of gender or race.
“Age is more than just a number,” said Aditi Gurkar, PhD (above), Assistant Professor of Geriatric Medicine at University of Pittsburg School of Medicine and the study’s senior author in a news release. “Imagine two people aged 65: One rides a bike to work and goes skiing on the weekends and the other can’t climb a flight of stairs. They have the same chronological age, but very different biological ages. Why do these two people age differently? This question drives my research.” Gurkar’s research may one day lead to new clinical laboratory tests physicians will order when evaluating their patients’ health. (Photo copyright: University of Pittsburg.)
Clear Differences in Metabolites
According to the National Cancer Institute, a metabolite is a “substance made or used when the body breaks down food, drugs, or chemicals, or its own tissue (for example, fat or muscle tissue). This process, called metabolism, makes energy and the materials needed for growth, reproduction, and maintaining health. It also helps get rid of toxic substances.”
The UPMC researchers used metabolomics—the study of chemical process in the body that involves metabolites, other processes, and biproducts of cell metabolism—to create a “molecular fingerprint” of blood drawn from individuals in two separate study groups.
They included:
People over age 75 able to walk a flight of stairs or walk for 15 minutes without a break, and
People, age 65 to 75, who needed to rest during stair climbing and walk challenges.
The researchers found “clear differences” in the metabolomes of healthy agers as compared to rapid agers, suggesting that “metabolites in the blood could reflect biological age,” according to the UPMC news release.
“Other studies have looked at genetics to measure biological aging, but genes are very static. The genes you’re born with are the genes you die with,” said Gurkar in the news release.
Past studies on aging have explored other markers of biological age such as low grade-inflammation, muscle mass, and physical strength. But those markers fell short in “representing complexity of biological aging,” the UPMC study authors wrote in Aging Cell.
“One potential advantage of metabolomics over other ‘omic’ approaches is that metabolites are the final downstream products, and changes are closely related to the immediate (path) physiologic state of an individual,” they added.
The researchers used an artificial intelligence (AI) model that could identify “potential drivers of biological traits” and found three metabolites “that were most likely to promote healthy aging or drive rapid aging. In future research, they plan to delve into how these metabolites, and the molecular pathways that produce them, contribute to biological aging and explore interventions that could slow this process,” the new release noted.
“While it’s great that we can predict biological aging in older adults, what would be even more exciting is a blood test that, for example, can tell someone who’s 35 that they have a biological age more like a 45-year-old,” Gurkar said. “That person could then think about changing aspects of their lifestyle early—whether that’s improving their sleep, diet or exercise regime—to hopefully reverse their biological age.”
Looking Ahead
The UPMC scientists plan more studies to explore metabolites that promote healthy aging and rapid aging, and interventions to slow disease progression.
It’s possible that the blood-based HAM Index may one day become a diagnostic tool physicians and clinical laboratories use to aid monitoring of chronic diseases. As a commonly ordered blood test, it could help people find out biological age and make necessary lifestyle changes to improve their health and longevity.
With the incidence of chronic disease a major problem in the US and other developed countries, a useful diagnostic and monitoring tool like HAM could become a commonly ordered diagnostic procedure. In turn, that would allow clinical laboratories to track the same patient over many years, with the ability to use multi-year lab test data to flag patients whose biomarkers are changing in the wrong direction—thus enabling physicians to be proactive in treating their patients.
Study results from Switzerland come as clinical laboratory scientists seek new ways to tackle the problem of antimicrobial resistance in hospitals
Microbiologists and clinical laboratory scientists engaged in the fight against antibiotic-resistant (aka, antimicrobial resistant) bacteria will be interested in a recent study conducted at the University of Basel and University Hospital Basel in Switzerland. The epidemiologists involved in the study discovered that some of these so-called “superbugs” can remain in the body for as long as nine years continuing to infect the host and others.
The researchers wanted to see how two species of drug-resistant bacteria—K. pneumoniae and E. coli—changed over time in the body, according to a press release from the university. They analyzed samples of the bacteria collected from patients who were admitted to the hospital over a 10-year period, focusing on older individuals with pre-existing conditions. They found that K. pneumoniae persisted for up to 4.5 years (1,704 days) and E. coli persisted for up to nine years (3,376 days).
“These patients not only repeatedly become ill themselves, but they also act as a source of infection for other people—a reservoir for these pathogens,” said Lisandra Aguilar-Bultet, PhD, the study’s lead author, in the press release.
“This is crucial information for choosing a treatment,” explained Sarah Tschudin Sutter, MD, Head of the Division of Infectious Diseases and Hospital Epidemiology, and of the Division of Hospital Epidemiology, who specializes in hospital-acquired infections and drug-resistant pathogens. Sutter led the Basel University study.
“The issue is that when patients have infections with these drug-resistant bacteria, they can still carry that organism in or on their bodies even after treatment,” said epidemiologist Maroya Spalding Walters, MD (above), who leads the Antimicrobial Resistance Team in the Division of Healthcare Quality Promotion at the federal Centers for Disease Control and Prevention (CDC). “They don’t show any signs or symptoms of illness, but they can get infections again, and they can also transmit the bacteria to other people.” Clinical laboratories working with microbiologists on antibiotic resistance will want to follow the research conducted into these deadly pathogens. (Photo copyright: Centers for Disease Control and Prevention.)
COVID-19 Pandemic Increased Antibiotic Resistance
The Basel researchers looked at 76 K. pneumoniae isolates recovered from 19 patients and 284 E. coli isolates taken from 61 patients, all between 2008 and 2018. The study was limited to patients in which the bacterial strains were detected from at least two consecutive screenings on admission to the hospital.
“DNA analysis indicates that the bacteria initially adapt quite quickly to the conditions in the colonized parts of the body, but undergo few genetic changes thereafter,” the Basel University press release states.
The researchers also discovered that some of the samples, including those from different species, had identical mechanisms of drug resistance, suggesting that the bacteria transmitted mobile genetic elements such as plasmids to each other.
One limitation of the study, the authors acknowledged, was that they could not assess the patients’ exposure to antibiotics.
Meanwhile, recent data from the World Health Organization (WHO) suggests that the COVID-19 pandemic might have exacerbated the challenges of antibiotic resistance. Even though COVID-19 is a viral infection, WHO scientists found that high percentages of patients hospitalized with the disease between 2020 and 2023 received antibiotics.
“While only 8% of hospitalized patients with COVID-19 had bacterial co-infections requiring antibiotics, three out of four or some 75% of patients have been treated with antibiotics ‘just in case’ they help,” the WHO stated in a press release.
WHO uses an antibiotic categorization system known as AWaRe (Access, Watch, Reserve) to classify antibiotics based on risk of resistance. The most frequently prescribed antibiotics were in the “Watch” group, indicating that they are “more prone to be a target of antibiotic resistance and thus prioritized as targets of stewardship programs and monitoring.”
“When a patient requires antibiotics, the benefits often outweigh the risks associated with side effects or antibiotic resistance,” said Silvia Bertagnolio, MD, Unit Head in the Antimicrobial resistance (AMR) Division at the WHO in the press release. “However, when they are unnecessary, they offer no benefit while posing risks, and their use contributes to the emergence and spread of antimicrobial resistance.”
Citing research from the National Institutes of Health (NIH), NPR reported that in the US, hospital-acquired antibiotic-resistant infections increased 32% during the pandemic compared with data from just before the outbreak.
“While that number has dropped, it still hasn’t returned to pre-pandemic levels,” NPR noted.
The UPenn researchers have already developed an antimicrobial treatment derived from guava plants that has proved effective in mice, Vox reported. They’ve also trained an AI model to scan the proteomes of extinct organisms.
“The AI identified peptides from the woolly mammoth and the ancient sea cow, among other ancient animals, as promising candidates,” Vox noted. These, too, showed antimicrobial properties in tests on mice.
These findings can be used by clinical laboratories and microbiologists in their work with hospital infection control teams to better identify patients with antibiotic resistant strains of bacteria who, after discharge, may show up at the hospital months or years later.
Infection control teams and clinical laboratory managers may want to look at this new product designed to improve the diagnosis and treatment of sepsis
Accurate and fast diagnosis of sepsis for patients arriving in emergency departments is the goal of a new product that was just cleared by the federal Food and Drug Administration (FDA). It is also the newest example of how artificial intelligence (AI) continues to find its way into pathology and clinical laboratory medicine.
Sepsis is one of the deadliest killers in US hospitals. That is why there is interest in the recent action by the FDA to grant marketing authorization for an AI-powered sepsis detection software through the agency’s De Novo Classification Request. The DNCR “provides a marketing pathway to classify novel medical devices for which general controls alone, or general and special controls, provide reasonable assurance of safety and effectiveness for the intended use, but for which there is no legally marketed predicate device,” the FDA’s website states.
Unlike a single analyte assay that is run in a clinical laboratory, Prenosis’ AI/ML software uses 22 diagnostic and predictive parameters, along with ML algorithms, to analyze data and produce a clinically actionable answer on sepsis.
It is important for clinical laboratory managers and pathologists to recognize that this diagnostic approach to sepsis brings together a number of data points commonly found in a patient’s electronic health record (EHR), some of which the lab generated and others the lab did not generate.
“Sepsis is a serious and sometimes deadly complication. Technologies developed to help prevent this condition have the potential to provide a significant benefit to patients,” said Jeff Shuren, MD, JD, Director of the FDA’s Center for Devices and Radiological Health, in a statement. “The FDA’s authorization of the Prenosis Sepsis ImmunoScore software establishes specific premarket and post-market requirements for this device type.” Clinical laboratory EHRs contain some of the data points Prenosis’ diagnostic software uses. (Photo copyright: US Food and Drug Administration.)
How it Works
To assist doctors diagnose sepsis, the ImmunoScore software is first integrated into the patient’s hospital EHR. From there, it leverages 22 parameters including:
White blood cell count to produce a score that informs caregivers of the patient’s risk for sepsis within 24 hours, MedTech Dive reported.
Instead of requiring a doctor or nurse to look at each parameter separately, the SaMD tool uses AI “to evaluate all those markers at once”, CNBC noted. It then produces a risk score and four discrete risk stratification categories (low, medium, high, and very high) which correlate to “a patient’s risk of deterioration” represented by:
By sharing these details—a number from one to 100 for each of the 22 diagnostic and predictive parameters—Sepsis ImmunoScore helps doctors determine which will likely contribute most to the patient’s risk for developing sepsis, MedTech Dive reported.
“A lot of clinicians don’t trust AI products for multiple reasons. We are trying very hard to counter that skepticism by making a tool that was validated by the FDA first, and then the second piece is we’re not trying to replace the clinician,” Bobby Reddy Jr., PhD, Prenosis co-founder and CEO, told MedTech Dive.
Big Biobank and Blood Sample Data
Prenosis, which says its goal is the “enabling [of] precision medicine in acute care” developed Sepsis ImmunoScore using the company’s own biobank and a dataset of more than 100,000 blood samples from more than 25,000 patients.
AI algorithms drew on this biological/clinical dataset—the largest in the world for acute care patients suspected of having serious infections, according to Prenosis—to “elucidate patterns in rapid immune response.”
“It does not work without data, and the data started at Carle,” said critical care specialist Karen White, MD, PhD, Carle Foundation Hospital, St. Louis, MO, in the news release. “The project involved a large number of physicians, research staff, and internal medicine residents at Carle who helped recruit patients, collect data, and samples,” she said.
Opportunity for Clinical Laboratories
Sepsis is a life-threatening condition based on an “extreme response to an infection” that affects nearly 1.7 million adults in the US each year and is responsible for 350,000 deaths, according to US Centers for Disease Control and Prevention (CDC) data.
A non-invasive diagnostic tool like Sepsis ImmunoScore will be a boon to emergency physicians and the patients they treat. Now that the FDA has authorized the SaMD diagnostic tool to go to market, it may not be long before physicians can use the information it produces to save lives.
Clinical laboratory managers inspired by the development of Sepsis ImmunoScore may want to look for similar ways they can take certain lab test results and combine them with other data in an EHR to create intelligence that physicians can use to better treat their patients. The way forward in laboratory medicine will be combining lab test results with other relevant sets of data to create clinically actionable intelligence for physicians, patients, and payers.
Ten year collaboration between Google and Harvard may lead to a deeper understanding of the brain and new clinical laboratory diagnostics
With all our anatomic pathology and clinical laboratory science, we still do not know that much about the structure of the brain. But now, scientists at Harvard University and Google Research studying the emerging field of connectomics have published a highly detailed 3D reconstruction of human brain tissue that allows visualization of neurons and their connections at unprecedented nanoscale resolutions.
Further investigation of the nano-connections within the human brain could lead to novel insights about the role specific proteins and molecules play in the function of the brain. Though it will likely be years down the road, data derived from this study could be used to develop new clinical laboratory diagnostic tests.
The data to generate the model came from Google’s use of artificial intelligence (AI) algorithms to color-code Harvard’s electron microscope imaging of a cubic millimeter of neural tissue—equivalent to a half-grain of rice—that was surgically removed from an epilepsy patient.
“That tiny square contains 57,000 cells, 230 millimeters of blood vessels, and 150 million synapses, all amounting to 1,400 terabytes of data,” according to the Harvard Gazette, which described the project as “the largest-ever dataset of human neural connections.”
“A terabyte is, for most people, gigantic, yet a fragment of a human brain—just a minuscule, teeny-weeny little bit of human brain—is still thousands of terabytes,” said neuroscientist Jeff W. Lichtman, MD, PhD, Jeremy R. Knowles Professor of Molecular and Cellular Biology, whose Lichtman Lab at Harvard University collaborated on the project with researchers from Google. The two labs have been working together for nearly 10 years on this project, the Harvard Gazette reported.
Lichtman’s lab focuses on the emerging field of connectomics, defined “as understanding how individual neurons are connected to one another to form functional networks,” said neurobiologist Wei-Chung Allen Lee, PhD, Assistant Professor of Neurology, Harvard Medical School, in an interview with Harvard Medical News. “The goal is to create connectomes—or detailed structural maps of connectivity—where we can see every neuron and every connection.” Lee was not involved with the Harvard/Google Research study.
“The human brain uses no more power than a dim incandescent light bulb, yet it can accomplish feats still not possible with the largest artificial computing systems,” wrote Google Research scientist Viren Jain, PhD (above), in a blog post. “To understand how requires a level of understanding more profound than knowing what part of the brain is responsible for what function. The field of connectomics aims to achieve this by precisely mapping how each cell is connected to others.” Google’s 10-year collaboration with Harvard University may lead to new clinical laboratory diagnostics. (Photo copyright: Google Research.)
Study Data and Tools Freely Available
Along with the Science paper, the researchers publicly released the data along with analytic and visualization tools. The study noted that the dataset “is large and incompletely scrutinized,” so the scientists are inviting other researchers to assist in improving the model.
“The ability for other researchers to proofread and refine this human brain connectome is one of many ways that we see the release of this paper and the associated tools as not only the culmination of 10 years of work, but the beginning of something new,” wrote Google Research scientist Viren Jain, PhD, in a blog post that included links to the online resources.
One of those tools—Neuroglancer—allows any user with a web browser to view 3D models of neurons, axons, synapses, dendrites, blood vessels, and other objects. Users can rotate the models in xyz dimensions.
Users with the requisite knowledge and skills can proofread and correct the models by signing up for a CAVE (Connectome Annotation Versioning Engine) account.
Researchers Found Several Surprises
To perform their study, Lichtman’s team cut the neural tissue into 5,000 slices, each approximately 30 nanometers thick, Jain explained in the blog post. They then used a multibeam scanning electron microscope to capture high-resolution images, a process that took 326 days.
Jain’s team at Google used AI tools to build the model. They “stitched and aligned the image data, reconstructed the three dimensional structure of each cell, including its axons and dendrites, identified synaptic connections, and classified cell types,” he explained.
Jain pointed to “several surprises” that the reconstruction revealed. For example, he noted that “96.5% of contacts between axons and their target cells have just one synapse.” However, he added, “we found a class of rare but extremely powerful synaptic connections in which a pair of neurons may be connected by more than 50 individual synapses.”
In their Science paper, the researchers suggest that “these powerful connections are not the result of chance, but rather that these pairs had a reason to be more strongly connected than is typical,” Jain wrote in the blog post. “Further study of these connections could reveal their functional role in the brain.”
Mysterious Structures
Another anomaly was the presence of “axon whorls,” as Jain described them, “beautiful but mysterious structures in which an axon wraps itself into complicated knots.”
Because the sample came from an epilepsy patient, Jain noted that the whorls could be connected to the disease or therapies or could be found in all brains.
“Given the scale and complexity of the dataset, we expect that there are many other novel structures and characteristics yet to be discovered,” he wrote. “These findings are the tip of the iceberg of what we expect connectomics will tell us about human brains.”
The researchers have a larger goal to create a comprehensive high-resolution map of a mouse’s brain, Harvard Medical News noted. This would contain approximately 1,000 times the data found in the 1-cubic-millimeter human sample.
Dark Daily has been tracking the different fields of “omics” for years, as research teams announce new findings and coin new areas of science and medicine to which “omics” is appended. Connectomics fits that description.
Though the Harvard/Google research is not likely to lead to diagnostic assays or clinical laboratory tests any time soon, it is an example of how advances in technologies are enabling researchers to investigate smaller and smaller elements within the human body.
