Fujifilm acquired Inspirata’s Dynamyx digital pathology technology and business while GE Healthcare announced a partnership with Tribun Health in Europe
Clinical pathology laboratories, especially in the US, have been slow to adopt digital imaging systems. But recent industry deals suggest that the market may soon heat up, at least in the eyes of vendors. These collaborators may hope that, by integrating diagnostic data, the accuracy and productivity of anatomic pathologists will improve while also shortening the time to diagnosis.
In the press release, Fujifilm stated that 85% of US healthcare organizations use analog systems for pathology. That compares with 86% in Europe and 90% in Asia, the company stated.
“Acquiring Inspirata’s digital pathology business allows Fujifilm to be an even stronger healthcare partner—bridging a technological gap between pathology, radiology, and oncology to facilitate a more collaborative approach to care delivery across the enterprise,” said Fujifilm CEO and president Teiichi Goto in the press release.
The press release cited data from Signify Research, a healthcare technology marketing data firm that is predicting the global market for digital pathology systems would double from $320 million in 2021 to $640 million by 2025.
Fujifilm previously had a deal with Inspirata to sell the Dynamyx system exclusively in the UK, Italy, Spain, Portugal, Belgium, the Netherlands, and Luxembourg, an August press release noted.
“A $320 million global industry in 2021 projected to reach $640 million by 2025, the rising number of cancer cases and the demonstrated benefits of digital pathology are fueling significant demand and market growth in the hospital and pharmaceutical industries,” said Henry Izawa (above), president and CEO, Fujifilm Healthcare Americas Corporation, in a press release. “These evolving clinical needs fuel Fujifilm’s investment and innovation in the digital revolution, and we look forward to introducing Dynamyx and its host of unique features and benefits to our Synapse customers and prospects as we strive to enable more efficient medical diagnosis and high-quality care.” (Photo copyright: LinkedIn.)
In announcing their new collaboration, GE Healthcare and Tribun Health said the integration of their systems—Edison Datalogue and the Tribun Health suite—would foster collaboration between pathologists and clinicians by providing a consolidated location for imaging records. This capability is especially important in oncology, they said.
“The oncology care pathway is one of the most complex with multiple steps involving a variety of specialists, complex tools, frequent decisions, and large data sets,” said GE Healthcare CEO of Enterprise Digital Solutions Nalinikanth Gollagunta in a GE press release. “With this digital pathology collaboration, we continue our journey towards simplifying the oncology care pathway with improved data management, the digitization of pathology, and streamlined data access.”
Tribun Health, based in Paris, France, offers a digital pathology platform that incorporates a camera system, artificial intelligence (AI)-based analysis, remote collaboration, and storage management, plus integration with third-party automation apps.
GE Healthcare claims that Edison Datalogue has the largest share of the Vendor Neutral Archive (VNA) market. That term refers to image archiving systems that use standard formats and interfaces instead of proprietary formats. They are an alternative to the more widely used Picture Archiving and Communications Systems (PACS) used in medical imaging.
The collaboration between the companies “is probably a strategic move to position GE as an integrator of imaging data and digital pathology data in oncology,” said Robert Michel Editor-in-Chief of Dark Daily and its sister publication The Dark Report.
GE’s History with Dynamyx
This is not GE Healthcare’s first foray into digital pathology. In fact, the company had a major hand in launching the very Dynamyx system that Fujifilm recently acquired.
In “GE Healthcare Sells Omnyx to Inspirata,” The Dark Report interviewed Inspirata CEO Satish Sanan who at that time said the acquisition would allow his company to offer “a fully integrated, end-to-end digital pathology solution” in Canada and Europe. But GE Healthcare chose to end the partnership in 2016, citing regulatory uncertainty and variable global demand. Two years later, GE sold Omnyx to Inspirata.
GE Healthcare’s new collaboration with Tribun Health shows that the company “still recognizes the value of the pathology data in cancer diagnosis and wants to be in a position to manage that digital pathology data,” Michel said.
Fujifilm’s Plans
Fujifilm said it will incorporate Dynamyx into its Synapse Enterprise Imaging suite, which includes VNA, Radiology PACS, and Cardiology PACS. “Future releases of Dynamyx will also create opportunities for Fujifilm to support pharmaceutical and contract research organizations with toxicity testing data management for drug development,” the company stated in the press release.
With its recent moves into digital pathology, Fujifilm will be taking on major competitors including Philips, Danaher, and Roche, MedTech Dive reported.
Google designed the suite to ease radiologists’ workload and enable easy and secure sharing of critical medical imaging; technology may eventually be adapted to pathologists’ workflow
Clinical laboratory and pathology group leaders know that Google is doing extensive research and development in the field of cancer diagnostics. For several years, the Silicon Valley giant has been focused on digital imaging and the use of artificial intelligence (AI) algorithms and machine learning to detect cancer.
Now, Google Cloud has announced it is launching a new medical imaging suite for radiologists that is aimed at making healthcare data for the diagnosis and care of cancer patients more accessible. The new suite “promises to make medical imaging data more interoperable and useful by leveraging artificial intelligence,” according to MedCity News.
In a press release, medical technology company Hologic, and healthcare provider Hackensack Meridian Health in New Jersey, announced they were the first customers to use Google Cloud’s new suite of medical imaging products.
“Hackensack Meridian Health has begun using it to detect metastasis in prostate cancer patients earlier, and Hologic is using it to strengthen its diagnostic platform that screens women for cervical cancer,” MedCity News reported.
“Google pioneered the use of AI and computer vision in Google Photos, Google Image Search, and Google Lens, and now we’re making our imaging expertise, tools, and technologies available for healthcare and life sciences enterprises,” said Alissa Hsu Lynch (above), Global Lead of Google Cloud’s MedTech Strategy and Solutions, in a press release. “Our Medical Imaging Suite shows what’s possible when tech and healthcare companies come together.” Clinical laboratory companies may find Google’s Medical Imaging Suite worth investigating. (Photo copyright: Influencive.)
.
Easing the Burden on Radiologists
Clinical laboratory leaders and pathologists know that laboratory data drives most healthcare decision-making. And medical images make up 90% of all healthcare data, noted an article in Proceedings of the IEEE (Institute of Electrical and Electronics Engineers).
More importantly, medical images are growing in size and complexity. So, radiologists and medical researchers need a way to quickly interpret them and keep up with the increased workload, Google Cloud noted.
“The size and complexity of these images is huge, and, often, images stay sitting in data siloes across an organization,” said Alissa Hsu Lynch, Global Lead, MedTech Strategy and Solutions at Google, told MedCity News. “In order to make imaging data useful for AI, we have to address interoperability and standardization. This suite is designed to help healthcare organizations accelerate the development of AI so that they can enable faster, more accurate diagnosis and ease the burden for radiologists,” she added.
According to the press release, Google Cloud’s Medical Imaging Suite features include:
Imaging Storage: Easy and secure data exchange using the international DICOM (digital imaging and communications in medicine) standard for imaging. A fully managed, highly scalable, enterprise-grade development environment that includes automated DICOM de-identification. Seamless cloud data management via a cloud-native enterprise imaging PACS (picture archiving and communication system) in clinical use by radiologists.
Imaging Lab: AI-assisted annotation tools that help automate the highly manual and repetitive task of labeling medical images, and Google Cloud native integration with any DICOMweb viewer.
Imaging Datasets and Dashboards: Ability to view and search petabytes of imaging data to perform advanced analytics and create training datasets with zero operational overhead.
Imaging AI Pipelines: Accelerated development of AI pipelines to build scalable machine learning models, with 80% fewer lines of code required for custom modeling.
Imaging Deployment: Flexible options for cloud, on-prem (on-premises software), or edge deployment to allow organizations to meet diverse sovereignty, data security, and privacy requirements—while providing centralized management and policy enforcement with Google Distributed Cloud.
First Customers Deploy Suite
Hackensack Meridian Health hopes Google’s imaging suite will, eventually, enable the healthcare provider to predict factors affecting variance in prostate cancer outcomes.
“We are working toward building AI capabilities that will support image-based clinical diagnosis across a range of imaging and be an integral part of our clinical workflow,” said Sameer Sethi, Senior Vice President and Chief Data and Analytics Officer at Hackensack, in a news release.
The New Jersey healthcare network said in a statement that its work with Google Cloud includes use of AI and machine learning to enable notification of newborn congenital disorders and to predict sepsis risk in real-time.
Hologic, a medical technology company focused on women’s health, said its collaboration integrates Google Cloud AI with the company’s Genius Digital Diagnostics System.
“By complementing our expertise in diagnostics and AI with Google Cloud’s expertise in AI, we’re evolving our market-leading technologies to improve laboratory performance, healthcare provider decision making, and patient care,” said Michael Quick, Vice President of Research and Development and Innovation at Hologic, in the press release.
Hologic says its Genius Digital Diagnostics System combines AI with volumetric medical imaging to find pre-cancerous lesions and cancer cells. From a Pap test digital image, the system narrows “tens of thousands of cells down to an AI-generated gallery of the most diagnostically relevant,” according to the company website.
Hologic plans to work with Google Cloud on storage and “to improve diagnostic accuracy for those cancer images,” Hsu Lynch told MedCity News.
Medical image storage and sharing technologies like Google Cloud’s Medical Imaging Suite provide an opportunity for radiologists, researchers, and others to share critical image studies with anatomic pathologists and physicians providing care to cancer patients.
One key observation is that the primary function of this service that Google has begun to deploy is to aid in radiology workflow and productivity, and to improve the accuracy of cancer diagnoses by radiologists. Meanwhile, Google continues to employ pathologists within its medical imaging research and development teams.
Assuming that the first radiologists find the Google suite of tools effective in support of patient care, it may not be too long before Google moves to introduce an imaging suite of tools designed to aid the workflow of surgical pathologists as well.
University of Cincinnati researchers hypothesize that low levels of amyloid-beta protein, not amyloid plaques, are to blame
New research from the University of Cincinnati (UC) and Karolinska Institute in Sweden challenges the prevailing theory about the causes of Alzheimer’s disease, suggesting the possibility of new avenues for the development of effective clinical laboratory assays, as well as effective therapies for treating patients diagnosed with Alzheimer’s.
Scientists have long theorized that the disease is caused by a buildup of amyloid plaques in the brain. These plaques are hardened forms of the amyloid-beta protein, according to a UC news story.
“The paradox is that so many of us accrue plaques in our brains as we age, and yet so few of us with plaques go on to develop dementia,” said Alberto Espay, MD, one of the lead researchers of the study, in another UC news story. Espay is Professor of Neurology at the UC College of Medicine and Director and Endowed Chair of the Gardner Center for Parkinson’s Disease and Movement Disorders.
“Yet the plaques remain the center of our attention as it relates to biomarker development and therapeutic strategies,” he added.
“It’s only too logical, if you are detached from the biases that we’ve created for too long, that a neurodegenerative process is caused by something we lose, amyloid-beta, rather than something we gain, amyloid plaques,” said Alberto Espay, MD (above), in a University of Cincinnati news story. “Degeneration is a process of loss, and what we lose turns out to be much more important.” The UC study could lead to new clinical laboratory diagnostics, as well as treatments for Alzheimer’s and Parkinson’s diseases. (Photo copyright: University of Cincinnati.)
.
High Levels of Aβ42 Associated with Lower Dementia Risk
In their retrospective longitudinal study, the UC researchers looked at clinical assessments of individuals participating in the Dominantly Inherited Alzheimer Network (DIAN) cohort study. DIAN is an ongoing effort, sponsored by the Washington University School of Medicine in St. Louis, to identify biomarkers associated with Alzheimer’s among people who carry Alzheimer’s mutations.
The researchers found that study participants with high levels of a soluble amyloid-beta protein, Aβ42, were less likely to develop dementia than those with lower levels of the protein, regardless of the levels of amyloid plaques in their brains or the amount of tau protein—either as phosphorylated tau (p-tau) or total tau (t-tau)—in their cerebral spinal fluid. P-tau and t-tau are proteins that form “tau tangles” in the brain that are also associated with Alzheimer’s.
One limitation of the study was that the researchers were unable to include Aβ40, another amyloid-beta protein, in their analysis. But they noted that this “did not limit the testing of our hypothesis since Aβ40 exhibits lower fibrillogenicity and lesser depletion than Aβ42, and is therefore less relevant to the process of protein aggregation than Aβ42.” Fibrillogenicity, in this context, refers to the process by which the amyloid-beta protein hardens into plaque.
While the presence of plaques may be correlated with Alzheimer’s, “Espay and his colleagues hypothesized that plaques are simply a consequence of the levels of soluble amyloid-beta in the brain decreasing,” UC news stated. “These levels decrease because the normal protein, under situations of biological, metabolic, or infectious stress, transform into the abnormal amyloid plaques.”
The UC News story also noted that many attempts to develop therapeutics for Alzheimer’s have focused on reducing amyloid plaques, but “in some clinical trials that reduced the levels of soluble amyloid-beta, patients showed worsening in clinical outcomes.”
New Therapeutics for Multiple Neurodegenerative Diseases
Eisai, a Japanese pharmaceutical company, recently announced phase three clinical trial results of lecanemab, an experimental drug jointly developed by Eisai and Biogen, claiming that the experimental Alzheimer’s drug modestly reduced cognitive decline in early-stage patients, according to NBC News.
Espay noted that lecanemab “does something that most other anti-amyloid treatments don’t do in addition to reducing amyloid: it increases the levels of the soluble amyloid-beta.” That may slow the process of soluble proteins hardening into plaques.
Beyond their findings about Alzheimer’s, the researchers believe similar mechanisms could be at work in other neurodegenerative diseases such as Parkinson’s disease, where the soluble alpha-synuclein protein also hardens into deposits.
“We’re advocating that what may be more meaningful across all degenerative diseases is the loss of normal proteins rather than the measurable fraction of abnormal proteins,” Espay said. “The net effect is a loss not a gain of proteins as the brain continues to shrink as these diseases progress.”
Espay foresees two approaches to treating these diseases: Rescue medicine, perhaps based on increasing levels of important proteins, and precision medicine, which “entails going deeper to understand what is causing levels of soluble amyloid-beta to decrease in the first place, whether it is a virus, a toxin, a nanoparticle, or a biological or genetic process,” according to UC News. “If the root cause is addressed, the levels of the protein wouldn’t need to be boosted because there would be no transformation from soluble, normal proteins to amyloid plaques.”
Clinical Laboratory Impact
What does this mean for clinical laboratories engaged in treatment of both Alzheimer’s and Parkinson’s patients? A new understanding of the disease would create “the opportunity to identify new biomarkers and create new clinical laboratory tests that may help diagnose Alzheimer’s earlier in the disease progression, along with tests that help with the patient’s prognosis and monitoring his or her progression,” said Robert Michel, Editor-in-Chief of Dark Daily and its sister publication The Dark Report.
Given the incidence of Alzheimer’s disease in the population, any clinical laboratory test cleared by the FDA would be a frequently-ordered assay, Michel noted. It also would create the opportunity for pathologists and clinical laboratories to provide valuable interpretation about the test results to the ordering physicians.
Similar health monitoring devices have been popular with chronic disease patients and physicians who treat them; this technology may give clinical laboratories a new diagnostic tool
There is an ever-increasing number of companies working to develop lab testing technologies that would be used outside of the traditional clinical laboratory. One such example is Nutromics, an Australia-based medical technology company which recently announced it has raised US $14 million to fund its new lab-on-a-patch platform, according to a company press release.
Nutromics’ lab-on-a-patch device “uses DNA sensor technology to track multiple targets in the human body, including disease biomarkers and hard-to-dose drugs,” according to MobiHealthNews. Notably, Nutromics’ technology uses interstitial fluid as the sample source.
Nutromics raised $4 million last year to support a manufacturing facility and an initial human clinical trial of its “continuous molecular monitoring (CMM) platform technology that is able to track multiple targets in the human body via a single wearable sensor. The platform provides real-time, continuous molecular-level insights for remote patient monitoring and hospital-at-home systems,” MobiHealthNews reported.
“We are aiming to cause a paradigm shift in diagnostic healthcare by essentially developing a lab-on-a-patch. A lack of timely and continuous diagnostic insights can strongly impact outcomes when dealing with critical disease states. With this strategic industry and VC (venture capital) investment in us, we see more confidence in our technology and hope to accelerate our growth,” said entrepreneur and chemical engineer Peter Vranes (above), co-founder and CEO of Nutromics, in a press release. Clinical laboratory leaders have watched similar biometric monitoring devices come to fruition. (Photo copyright: Nutromics.)
.
How Nutromics’ Lab-on-a-Patch Works
“Our technology is, in fact, two technologies coming together—a marker and needle. What that does is give us access to fluid under your skin called interstitial fluid. If you’re going to measure something continuously, that’s a really good fluid [to measure],” Vranes told Outcomes Rocket.
Vranes calls the system’s aptamer-based sensor platform technology the “jewel in the crown.” An aptamer is a short sequence of artificial DNA or RNA that binds a specific target molecule. Nutromics’ aptamer sensor, Vranes said, enables targeting of analytes, unlike continuous glucose monitors (CGMs).
“[CGMs] are limited to metabolites—things that are already in the body like glucose and lactate. We’re not limited to those. We can do a whole range of different targets. And what that gives us is a ‘blue ocean’ opportunity to go in and solve problems in areas that other technologies just can’t solve,” Vranes said.
Nutromics plans to develop multiple aptamer-based sensors that measure a variety of analytes in interstitial fluid, Medtech Insight noted.
Nutromics’ wearable DNA sensor lab-on-a-patch technology (above) enables monitoring of multiple targets, including disease biomarkers and some medications, MobiHealthNews explained. The wearable patch contains microneedles that painlessly access interstitial fluid under the skin. Collected data is wirelessly transmitted to a software application and integrates with consumer health software and provider platforms, according to Nutromics. Medical laboratories could have a role in collecting this data and adding it other test results from patients using the wearable patch. (Photo copyright: Nutromics.)
Initial Launch Will Include Antibiotic Monitoring
Nutromics expects to initially launch therapeutic monitoring of vancomycin, a glycopeptide antibiotic medication used to treat various bacterial infections. The company says 60% of doses for this prescription antibiotic are not within therapeutic range.
“Interstitial fluid originates in the blood and then leaks out of capillaries to bring nutrients to cells in the body’s tissues. Because interstitial fluid is in direct communication with the cells, it should have information about the tissues themselves beyond what can be measured from testing the blood,” said Mark Prausnitz, PhD, Regents Professor and J. Erskine Love Jr. Chair, Georgia Tech School of Chemical and Biomolecular Engineering, in a 2020 news release announcing results of human trials of microneedle-based ISF sampling.
“We sampled interstitial fluid from 21 human participants and identified clinically relevant and sometimes distinct biomarkers in interstitial fluid when compared to companion plasma samples based on mass spectrometry analysis,” the scientists wrote.
Clinical laboratory leaders and pathologists will find it useful to monitor the development of diagnostics for use outside the lab. Nutromics is an example of a company developing wearable health technology that painlessly gathers data for lab tests to be conducted in point-of-care and near-patient settings.
Studies could lead to new prognostic biomarkers and clinical laboratory diagnostics for cancer
Might fungi be involved in human cancers? Two separately published studies have found fungal DNA in various cancers in the human body. However, the researchers are unclear on how the fungi got into the cancer cells and if it is affecting the cancers’ pathology. Nevertheless, these discoveries could lead to utilizing tumor-associated fungal DNA as clinical laboratory diagnostics or prognostic biomarkers in the fight against cancer.
“The finding that fungi are commonly present in human tumors should drive us to better explore their potential effects and re-examine almost everything we know about cancer through a ‘microbiome lens,’” said Ravid Straussman, MD, PhD (above), a principal investigator at Weizmann Institute of Science and one of the authors of the study in a UCSD press release. These findings could lead to new clinical laboratory diagnostics and prognostic biomarkers. (Photo copyright: Weizmann Institute of Science.)
.
Microbiome Key to Cancer Biology and Detection
To perform their research, the team examined 17,401 samples of patient tissues, blood, and plasma across 35 different types of cancers in four independent cohorts. They discovered fungal DNA and cells in low abundances in many human cancers.
“The existence of fungi in most human cancers is both a surprise and to be expected,” said biologist Rob Knight, PhD, founding Director of the Center for Microbiome Innovation and Professor of Pediatrics and Computer Science and Engineering at UC San Diego in a UCSD press release. “It is surprising because we don’t know how fungi could get into tumors throughout the body. But it is also expected because it fits the pattern of healthy microbiomes throughout the body, including the gut, mouth and skin, where bacteria and fungi interact as part of a complex community.”
The main highlights of this study include:
Fungi detected in the different cancer types were often intracellular.
Multiple fungal-bacterial-immune ecologies were detected across tumors.
Intratumoral fungi stratified clinical outcomes, including immunotherapy response.
Cell-free fungal DNA found in both healthy and cancer patients in early-stage disease.
Fungi found on the human body appear as either environmental fungi, such as yeasts and molds, and commensal fungi, which live either on or inside the body. Both are typically harmless to most healthy people and can provide some benefits, such as improving gut health, but they may also be a contributing factor in some disease.
The researchers found that there were notable parallels between specific fungi and certain factors, such as age, tumor subtypes, smoking status, immunotherapy responses, and survival measures.
“These findings validate the view that the microbiome in its entirety is a key piece of cancer biology and may present significant translational opportunities, not only in cancer detection, but also in other biotech applications related to drug development, cancer evolution, minimal residual disease, relapse, and companion diagnostics,” said Gregory Sepich-Poore, MD, PhD, one of the study’s authors and co-founder and chief analytics officer at biotechnology company Micronoma, in the UCSD press release.
New Clinical Laboratory Tests to Identify Fungal Species in Cancer
Researchers from Duke University and Cornell University uncovered compelling evidence of fungi in multiple cancer types and focused on a detected link between Candida and gastrointestinal cancers.
They found that “several Candida species were enriched in tumor samples and tumor-associated Candida DNA was predictive of decreased survival,” according to their paper.
Their analysis of multiple body sites revealed tumor-associated mycobiomes in fungal cells. The researchers found that fungal spores known as blastomyces were associated with tumor tissues in lung cancers, and that high rates of Candida were present in stomach and colon cancers.
The Duke/Cornell researchers hope their work can provide a framework to develop new tests that can distinguish fungal species in tumors and predict cancer progression and help medical professionals and patients chose the best treatment therapies.
“These findings open up a lot of exciting research directions, from the development of diagnostics and treatments to studies of the detailed biological mechanisms of fungal relationships to cancers,” said Iliyan Iliev, PhD, Associate Professor of Microbiology and Immunology in Medicine, Weill Cornell Medicine, and one of the authors of the study, in a Weill news release.
More research is needed to determine if fungal DNA plays a role in disease pathology or if its presence does not have any causal link.
“It’s plausible that some of these fungi are promoting tumor progression and metastasis, but even if they aren’t, they could be very valuable as prognostic indicators,” Iliev said.
The insights gleaned from these two studies will be of particular interest to microbiologists, clinical laboratory professionals, and anatomic pathologists. Additional research could answer questions about how and if fungi infect tumors and if such fungi is a factor that increases cancer risk and outcomes.
Proof-of-concept study ‘highlights that using AI to integrate different types of clinically informed data to predict disease outcomes is feasible’ researchers say
Artificial intelligence (AI) and machine learning are—in stepwise fashion—making progress in demonstrating value in the world of pathology diagnostics. But human anatomic pathologists are generally required for a prognosis. Now, in a proof-of-concept study, researchers at Brigham and Women’s Hospital in Boston have developed a method that uses AI models to integrate multiple types of data from disparate sources to accurately predict patient outcomes for 14 different types of cancer.
The process also uncovered “the predictive bases of features used to predict patient risk—a property that could be used to uncover new biomarkers,” according to Genetic Engineering and Biotechnology News (GEN).
Should these research findings become clinically viable, anatomic pathologists may gain powerful new AI tools specifically designed to help them predict what type of outcome a cancer patient can expect.
“Experts analyze many pieces of evidence to predict how well a patient may do. These early examinations become the basis of making decisions about enrolling in a clinical trial or specific treatment regimens,” said Faisal Mahmood, PhD (above) in a Brigham press release. “But that means that this multimodal prediction happens at the level of the expert. We’re trying to address the problem computationally,” he added. Should they be proven clinically-viable through additional studies, these findings could lead to useful tools that help anatomic pathologists and clinical laboratory scientists more accurately predict what type of outcomes cancer patient may experience. (Photo copyright: Harvard.)
AI-based Prognostics in Pathology and Clinical Laboratory Medicine
The team at Brigham constructed their AI model using The Cancer Genome Atlas (TCGA), a publicly available resource which contains data on many types of cancer. They then created a deep learning-based algorithm that examines information from different data sources.
Pathologists traditionally depend on several distinct sources of data, such as pathology images, genomic sequencing, and patient history to diagnose various cancers and help develop prognoses.
For their research, Mahmood and his colleagues trained and validated their AI algorithm on 6,592 H/E (hematoxylin and eosin) whole slide images (WSIs) from 5,720 cancer patients. Molecular profile features, which included mutation status, copy-number variation, and RNA sequencing expression, were also inputted into the model to measure and explain relative risk of cancer death.
The scientists “evaluated the model’s efficacy by feeding it data sets from 14 cancer types as well as patient histology and genomic data. Results demonstrated that the models yielded more accurate patient outcome predictions than those incorporating only single sources of information,” states a Brigham press release.
“This work sets the stage for larger healthcare AI studies that combine data from multiple sources,” said Faisal Mahmood, PhD, Associate Professor, Division of Computational Pathology, Brigham and Women’s Hospital; and Associate Member, Cancer Program, Broad Institute of MIT and Harvard, in the press release. “In a broader sense, our findings emphasize a need for building computational pathology prognostic models with much larger datasets and downstream clinical trials to establish utility.”
Future Prognostics Based on Multiple Data Sources
The Brigham researchers also generated a research tool they dubbed the Pathology-omics Research Platform for Integrative Survival Estimation (PORPOISE). This tool serves as an interactive platform that can yield prognostic markers detected by the algorithm for thousands of patients across various cancer types.
The researchers believe their algorithm reveals another role for AI technology in medical care, but that more research is needed before their model can be implemented clinically. Larger data sets will have to be examined and the researchers plan to use more types of patient information, such as radiology scans, family histories, and electronic medical records in future tests of their AI technology.
“Future work will focus on developing more focused prognostic models by curating larger multimodal datasets for individual disease models, adapting models to large independent multimodal test cohorts, and using multimodal deep learning for predicting response and resistance to treatment,” the Cancer Cell paper states.
“As research advances in sequencing technologies, such as single-cell RNA-seq, mass cytometry, and spatial transcriptomics, these technologies continue to mature and gain clinical penetrance, in combination with whole-slide imaging, and our approach to understanding molecular biology will become increasingly spatially resolved and multimodal,” the researchers concluded.
Anatomic pathologists may find the Brigham and Women’s Hospital research team’s findings intriguing. An AI tool that integrates data from disparate sources, analyzes that information, and provides useful insights, could one day help them provide more accurate cancer prognoses and improve the care of their patients.
