Artificial intelligence tools for radiology, clinical laboratory, and pathology diagnostics continue to advance and improve
Researchers in Germany have developed a fully automated, artificial intelligence (AI) tool that improves the diagnosis of prostate cancer. Developed by mediaire, a company that creates AI-based tools for radiologists, the software reduces clinical workloads and could be beneficial in counteracting issues associated with variability in magnetic resonance imaging (MRI) reporting. This is another example of AI’s growth in the clinical diagnostic industry, including clinical laboratory and pathology medicine.
The software, called mdprostate, has received the mandatory certification mark (CE or European Conformity) for products sold within the European Economic Area (EEA). It is now commercially available in those countries and was recently incorporated into the picture archiving and communications system (PACS) of some healthcare organizations and applied to a group of patients who had undergone a multiparametric prostate MRI (mpMRI).
The goal was to compare the overall performance of mdprostate against radiologists who executed the initial interpretations of the mpMRIs, according to Health Imaging.
“Mdprostate is intended to support radiologists by automating time-consuming processes and improving the objectivity of diagnosis through data quantification,” said Tonia Michaely, chief of staff at mediaire, in a news release.
“By providing objective assessments and standardizing lesion detection and classification, AI has the potential to augment radiologists’ performance throughout the PCa [prostate cancer] diagnostic pathway,” Nadine Bayerl, Dr. med., a radiologist with the Institute of Radiology at University Hospital Erlangen and corresponding author of the mediaire study, told Health Imaging. (Photo copyright: University Hospital Erlangen.)
Scoring Cancer Risk
To perform the comparison, a team of researchers applied the AI tool to 123 prostate MRI exams followed by systematic and targeted biopsies. The software was instructed to automatically segment the prostrate, calculate prostate volume, and classify lesions per the Prostate Imaging Reporting and Data System (PI-RADS).
PI-RADS, according to the America College of Radiology, is a reporting method that indicates how likely a lesion is to be clinically significant cancer on a score of one to five:
PI-RADS 1: very low (clinically significant cancer is highly unlikely to be present).
PI-RADS 2: low (clinically significant cancer is unlikely to be present).
PI-RADS 3: intermediate (the presence of clinically significant cancer is equivocal).
PI-RADS 4: high (clinically significant cancer is likely to be present).
PI-RADS 5: very high (clinically significant cancer is highly likely to be present).
For PI-RADS scores greater than two, mdprostate generated 100% sensitivity and dismissed all cancers for lesions that were below that threshold. For PI-RADS scores of four or higher, the AI tool yielded 85.5% sensitivity and specificity of 63.2% for clinically significant cancers.
Deep Learning in Diagnostic Pathway
“In practical terms, these results indicate that when a case falls below the PI-RADS ≥ 2 cutoff, clinicians can rule out malignancy with a high degree of confidence,” the authors explained in the European Journal of Radiology. “This capability is particularly valuable in clinical decision-making, as it allows for the safe avoidance of unnecessary biopsies or further invasive procedures in these patients.”
“Recent advances in deep learning algorithms, facilitated by larger labeled datasets, improved computing hardware, and refined training techniques, have led to several studies highlighting the diagnostic value of deep learning algorithms in prostate imaging,” radiologist Nadine Bayerl, Dr. med., Institute of Radiology at University Hospital Erlangen and corresponding author of the study, told Health Imaging.
The software “demonstrated high diagnostic performance in identifying and grading prostate lesions, with results comparable to those reported in meta-analyses of expert readers using PI-RADS,” the researchers noted in their published study.
“Its ability to standardize evaluations and potentially reduce variability underscores its potential as a valuable adjunct in the prostate cancer diagnostic pathway. The high accuracy of mdprostate, particularly in ruling out prostate cancers, highlights its clinical utility by reducing workload and enhancing patient outcomes,” they concluded.
AI in Clinical Laboratories and Pathology
Dark Daily has frequently covered AI’s expanding role in clinical laboratory testing and pathology diagnostics. At the recent Executive War College, a dozen sessions explored its growth in the industry. During one session, Sam Terese, CEO and president at Alverno Laboratories said, “AI is allowing us to drive our business. It is really resonating that we need to use AI in the future.”
Members who could not attend the 2025 Executive War College can order audio recordings of these valuable sessions by clicking here.
Lab professionals will learn more at the upcoming 30th annual edition of the event
Big changes and challenges are coming for the clinical laboratory anatomic pathology industry, and with them a slew of opportunities for lab and pathology practice leaders. At the upcoming 30th Annual Executive War College on Diagnostics, Pathology, and Clinical Laboratory Management, expert speakers and panelists will focus on the three most disruptive forces.
There will be more than 169 presenters at this year’s Executive War College. Those speakers include:
David Dexter, MD, clinical and laboratory pathology at M Health Fairview, and Sam Terese, president and CEO at Alverno Laboratories, who will present a strategic case study about the support labs can provide to parent hospitals when navigating new waters.
Paul Wilder, executive director of CommonWell Health Alliance, who will speak on the effort to improve the transferability and portability of patient and healthcare data in ways that improve the quality of care.
“Since the inception of The Dark Report in 1995 there has been continual change both within the US healthcare system and within the profession of laboratory medicine,” noted Robert L. Michel, Dark Daily’s editor-in-chief and creator of the Executive War College. “Now, three decades later, the following three items are imperatives for all labs: controlling costs; having adequate lab staff across all positions; and having enough capital to acquire and deploy new diagnostic technologies, along with the latest information technologies.”
“Most clinical laboratory managers would agree that many of the same operational pain points faced by labs in the 1990s exist today,” said Robert L. Michel (above), founder of the Executive War College. In an interview with Dark Daily, Michel broke down the nuances of this triad of forces and what participants in the Executive War College can expect. (Photo copyright: LabX.)
Forces at Work in Clinical Labs and Pathology Groups
Here’s a more detailed look at each of the forces that Michel noted.
Force 1: An acute shortage of experienced lab scientists
“When you look at the supply-demand for laboratory personnel in the United States today, it is recognized that demand exceeds supply, and that gap continues to widen,” Michel noted. “For example, in the case of anatomic pathologists, the increased number of case referrals grows faster than medical schools can train new pathologists. Currently, the ability of pathology laboratories large and small to hire and retain an adequate number of pathologists is a challenge.”
Executive War College attendees can expect panelists and speakers to highlight creative problem solving techniques to circumvent the challenges labor shortages cause.
Force 2: New applications of artificial intelligence
“Today every instrument vendor, every automation supplier, every software supplier, every service supplier is telling labs that they have artificial intelligence (AI) baked inside,” Michel observed. “It is important for lab managers to understand that a variety of technologies are used by different AI solutions.”
Clinical laboratory managers and pathologists interested in acquiring a deeper understanding of where to start with AI in their lab will find numerous sessions on artificial intelligence at this year’s Executive War College. “There will be a number of sessions this year where clinical labs discuss their success deploying various AI solutions,” Michel said.
Force 3: Financial stress across the entire US healthcare system
“It’s recognized that a significant number of US hospitals and integrated delivery networks (IDNs) are struggling to maintain adequate operating margins,” Michel noted. “This obviously impacts the clinical laboratories serving these hospitals. If the hospitals’ cash flows and operating profit margins are being squeezed, typically the administration comes to the lab team and says, ‘Your budget for next year will be x% less than this year.’
“There are many IDNs and hospital labs where budget cuts have happened for multiple years,” Michel continued. “As a consequence, labs in these hospitals must be nimble to maintain a high-quality menu of diagnostic tests. Several years of such budget cuts by the parent hospital can undermine the ability of the clinical lab team to offer competitive salary packages to attract and retain the clinical lab scientists, pathologists, and clinical chemists they need.”
Recognizing Opportunities in Today’s Lab Market
The good news is that—despite the negative forces acting upon the US healthcare system today—clinical laboratories, genetic testing companies, and anatomic pathology groups have a path forward.
“This path forward is informed by two longstanding precepts recognized by innovative managers. One precept is ‘Change creates new winners and losers.’ The other precept is ‘Change creates opportunity,’” Michel said. “Savvy lab leaders recognize the powerful truths in each precept.
“As healthcare has changed over the past four decades, nearly all the regional and national laboratories that were dominant in 1990, for example, don’t exist today!” he noted. “And yet, even as these lab organizations disappeared, new clinical lab organizations emerged that recognized healthcare’s changes and organized themselves to serve the changing needs of hospitals, office-based physicians, payers, and patients.”
All of these critical topics and more will be covered during the 30th Annual Executive War College on Diagnostics, Clinical Laboratory, and Pathology Management on April 29-30, 2025, at the Hyatt Regency in New Orleans. Signup today to bring your lab’s management team by registering at https://www.executivewarcollege.com.
New guidelines come on the heels of recommendations covering post-market modifications to AI products, including those incorporated into systems used by clinical laboratories
Artificial intelligence (AI) is booming in healthcare, and as the technology finds its way into more medical devices and clinical laboratory diagnostic test technologies the US Food and Drug Administration (FDA) has stepped up its efforts to provide regulatory guidance for developers of these products. This guidance will have an impact on the development of new lab test technology that uses AI going forward.
In December, the FDA issued finalized recommendations for submitting information about planned modifications to AI-enabled healthcare products. Then, in January, the federal agency issued draft guidance that covers product management and marketing submission more broadly. It is seeking public comments on the latter document through April 7.
“The FDA has authorized more than 1,000 AI-enabled devices through established premarket pathways,” said Troy Tazbaz, director of the Digital Health Center of Excellence at the FDA’s Center for Devices and Radiological Health, in a press release announcing the draft guidance.
This guidance “would be the first to provide total product life cycle recommendations for AI-enabled devices, tying together all design, development, maintenance and documentation recommendations, if and when finalized,” Healthcare IT News reported.
“Today’s draft guidance brings together relevant information for developers, shares learnings from authorized AI-enabled devices, and provides a first point-of-reference for specific recommendations that apply to these devices, from the earliest stages of development through the device’s entire life cycle,” said Troy Tazbaz (above), director of the Digital Health Center of Excellence at the FDA Center for Devices and Radiological Health, in a press release. The new guidance will likely affect the development of new clinical laboratory diagnostic technologies that use AI. (Photo copyright: LinkedIn.)
Engaging with FDA
One key takeaway from the guidance is that manufacturers “should engage with the FDA early to ensure that the testing to support the marketing submission for an AI-enabled device reflects the agency’s total product lifecycle, risk-based approach,” states an analysis from consulting firm Orrick, Herrington and Sutcliffe LLP.
Another key point is transparency, Orrick noted. For example, manufacturers should be prepared to offer details about the inputs and outputs of their AI models and demonstrate “how AI helps achieve a device’s intended use.”
Manufacturers should also take steps to avoid bias in data collection for these models. For example, they should gather evidence to determine “whether a device benefits all relevant demographic groups similarly to help ensure that such devices are safe and effective for their intended use,” Orrick said.
New Framework for AI in Drug Development
On the same day that FDA announced the device guidelines, the agency also proposed a framework for regulating use of AI models in developing drugs and biologics.
“AI can be used in various ways to produce data or information regarding the safety, effectiveness, or quality of a drug or biological product,” the federal agency stated in a press release. “For example, AI approaches can be used to predict patient outcomes, improve understanding of predictors of disease progression and process, and analyze large datasets.”
The press release noted that this is the first time the agency has proposed guidance on use of AI in drug development.
These include “bias and reliability problems due to variability in the quality, size, and representativeness of training datasets; the black-box nature of AI models in their development and decision-making; the difficulty of ascertaining the accuracy of a model’s output; and the dangers of data drift and a model’s performance changing over time or across environments. Any of these factors, in FDA’s thinking, could negatively impact the reliability and relevancy of the data sponsors provide FDA.”
The FDA also plans to participate in direct testing of AI-enabled healthcare tools. In October, the FDA and the Department of Veterans Affairs (VA) announced that they will launch “a joint health AI lab to evaluate promising emerging technologies,” according to Nextgov/FCW.
Elnahal said the facility will allow federal agencies and private entities “to test applications of AI in a virtual lab environment.” The goal is to ensure that the tools are safe and effective while adhering to “trustworthy AI principles,” he said.
“It’s essentially a place where you get rapid but effective evaluation—from FDA’s standpoint and from VA’s standpoint—on a potential new application of generative AI to, number one, make sure it works,” he told Nextgov/FCW.
He added that the lab will be set up with safeguards to ensure that the technologies can be tested safely.
“As long as they go through the right security protocols, we’d essentially be inviting parties to test their technology with a fenced off set of VA data that doesn’t have any risk of contagion into our actual live systems, but it’s still informative and simulated,” he told Nextgov/FCW.
There has been an explosion in the use of AI, machine learning, deep learning, and natural language processing in clinical laboratory diagnostic technologies. This is equally true of anatomic pathology, where AI-powered image analysis solutions are coming to market. That two federal agencies are motivated to establish guidelines on working relationships for evaluating the development and use of AI in healthcare settings tells you where the industry is headed.
Radiological method using AI algorithms to detect, locate, and identify cancer could negate the need for invasive, painful clinical laboratory testing of tissue biopsies
This will be of interest to histopathologists and radiologist technologists who are working to develop AI deep learning algorithms to read computed tomography scans (CT scans) to speed diagnosis and treatment of cancer patients.
“Researchers used the CT scans of 170 patients treated at The Royal Marsden with the two most common forms of retroperitoneal sarcoma (RPS)—leiomyosarcoma and liposarcoma—to create an AI algorithm, which was then tested on nearly 90 patients from centers across Europe and the US,” the news release notes.
The researchers then “used a technique called radiomics to analyze the CT scan data, which can extract information about the patient’s disease from medical images, including data which can’t be distinguished by the human eye,” the new release states.
The research team sought to make improvements with this type of cancer because these tumors have “a poor prognosis, upfront characterization of the tumor is difficult, and under-grading is common,” they wrote. The fact that AI reading of CT scans is a non-invasive procedure is major benefit, they added.
“This is the largest and most robust study to date that has successfully developed and tested an AI model aimed at improving the diagnosis and grading of retroperitoneal sarcoma using data from CT scans,” said the study’s lead oncology radiologist Christina Messiou, MD, (above), Consultant Radiologist at The Royal Marsden NHS Foundation Trust and Professor in Imaging for Personalized Oncology at The Institute of Cancer Research, London, in a news release. Invasive medical laboratory testing of cancer biopsies may eventually become a thing of the past if this research becomes clinically available for oncology diagnosis. (Photo copyright: The Royal Marsden.)
Study Details
RPS is a relatively difficult cancer to spot, let alone diagnose. It is a rare form of soft-tissue cancer “with approximately 8,600 new cases diagnosed annually in the United States—less than 1% of all newly diagnosed malignancies,” according to Brigham and Women’s Hospital.
In their published study, the UK researchers noted that, “Although more than 50 soft tissue sarcoma radiomics studies have been completed, few include retroperitoneal sarcomas, and the majority use single-center datasets without independent validation. The limited interpretation of the quantitative radiological phenotype in retroperitoneal sarcomas and its association with tumor biology is a missed opportunity.”
According to the ICR news release, “The [AI] model accurately graded the risk—or how aggressive a tumor is likely to be—[in] 82% of the tumors analyzed, while only 44% were correctly graded using a biopsy.”
Additionally, “The [AI] model also accurately predicted the disease type [in] 84% of the sarcomas tested—meaning it can effectively differentiate between leiomyosarcoma and liposarcoma—compared with radiologists who were not able to diagnose 35% of the cases,” the news release states.
“There is an urgent need to improve the diagnosis and treatment of patients with retroperitoneal sarcoma, who currently have poor outcomes,” said the study’s first author Amani Arthur, PhD, Clinical Research Fellow at The Institute of Cancer Research, London, and Registrar at The Royal Marsden NHS Foundation Trust, in the ICR news release.
“The disease is very rare—clinicians may only see one or two cases in their career—which means diagnosis can be slow. This type of sarcoma is also difficult to treat as it can grow to large sizes and, due to the tumor’s location in the abdomen, involve complex surgery,” she continued. “Through this early research, we’ve developed an innovative AI tool using imaging data that could help us more accurately and quickly identify the type and grade of retroperitoneal sarcomas than current methods. This could improve patient outcomes by helping to speed up diagnosis of the disease, and better tailor treatment by reliably identifying the risk of each patient’s disease.
“In the next phase of the study, we will test this model in clinic on patients with potential retroperitoneal sarcomas to see if it can accurately characterize their disease and measure the performance of the technology over time,” Arthur added.
Importance of Study Findings
Speed of detection is key to successful cancer diagnoses, noted Richard Davidson, Chief Executive of Sarcoma UK, a bone and soft tissue cancer charity.
“People are more likely to survive sarcoma if their cancer is diagnosed early—when treatments can be effective and before the sarcoma has spread to other parts of the body. One in six people with sarcoma cancer wait more than a year to receive an accurate diagnosis, so any research that helps patients receive better treatment, care, information and support is welcome,” he told The Guardian.
According to the World Health Organization, cancer kills about 10 million people worldwide every year. Acquisition and medical laboratory testing of tissue biopsies is both painful to patients and time consuming. Thus, a non-invasive method of diagnosing deadly cancers quickly, accurately, and early would be a boon to oncology practices worldwide and could save thousands of lives each year.
New artificial intelligence model agrees with interpretations of human medical technologists and microbiologists with extraordinary accuracy
Microbiology laboratories will be interested in news from Brescia University in Italy, where researchers reportedly have developed a deep learning model that can visually identify and analyze bacterial species in culture plates with a high level of agreement with interpretations made by medical technologists.
They initially trained and tested the system to digitally identify pathogens associated with urinary tract infections (UTIs). UTIs are the source for a large volume of clinical laboratory microbiological testing.
The system, known as DeepColony, uses hierarchical artificial intelligence technology. The researchers say hierarchical AI is better suited to complex decision-making than other approaches, such as generative AI.
In their Nature paper, the researchers explained that microbiologists use conventional methods to visually examine culture plates that contain bacterial colonies. The scientists hypothesize which species of bacteria are present, after which they test their hypothesis “by regrowing samples from each colony separately and then employing mass spectroscopy techniques,” to confirm their hypotheses.
However, DeepColony—which was designed for use with clinical laboratory automation systems—looks at high-resolution digital scans of cultured plates and attempts to identify the bacterial strains and analyze them in much the same way a microbiologist would. For example, it can identify species based on their appearance and determine which colonies are suitable for analysis, the researchers explained.
“Working on a large stream of clinical data, and a complete set of 32 pathogens, the proposed system is capable of effectively assisting plate interpretation with a surprising degree of accuracy in the widespread and demanding framework of urinary tract infections,” the study authors wrote. “Moreover, thanks to the rich species-related generated information, DeepColony can be used for developing trustworthy clinical decision support services in laboratory automation ecosystems from local to global scale.”
“Compared to the most common solutions based on single convolutional neural networks (CNN), multi-network architectures are attractive in our case because of their ability to fit into contexts where decision-making processes are stratified into a complex structure,” wrote the study’s lead author Alberto Signoroni, PhD (above), Associate Professor of Computer Science, University of Brescia, and his researcher team in their Nature paper. “The system must be designed to generate useful and easily interpretable information and to support expert decisions according to safety-by-design and human-in-the-loop policies, aiming at achieving cost-effectiveness and skill-empowerment respectively.” Microbiologists and clinical laboratory managers will want to follow the further development of this technology. (Photo copyright: University of Brescia.)
How Hierarchical AI Works
Writing in LinkedIn, patent attorney and self-described technology expert David Cain, JD, of Hauptman Ham, LLP, explained that hierarchical AI systems “are structured in layers, each with its own distinct role yet interconnected in a way that forms a cohesive whole. These systems are significant because they mirror the complexity of human decision-making processes, incorporating multiple levels of analysis and action. This multi-tiered approach allows for nuanced problem-solving and decision-making, akin to a seasoned explorer deftly navigating through a multifaceted terrain.”
DeepColony, the researchers wrote, consists of multiple convolutional neural networks (CNNs) that exchange information and cooperate with one another. The system is structured into five levels—labeled 0 through 4—each handling a different part of the analysis:
At level 0, the system determines the number of bacterial colonies and their locations on the plate.
At level 1, the system identifies “good colonies,” meaning those suitable for further identification and analysis.
At level 2, the system assigns each good colony to a bacterial species “based on visual appearance and growth characteristics,” the researchers wrote, referring to the determination as being “pathogen aware, similarity agnostic.”
The CNN used at this stage was trained by using images of 26,213 isolated colonies comprising 32 bacterial species, the researchers wrote in their paper. Most came from clinical laboratories, but some were obtained from the American Type Culture Collection (ATCC), a repository of biological materials and information resources available to researchers.
At level 3, the system attempts to improve accuracy by looking at the larger context of the plate. The goal here is to “determine if observed colonies are similar (pure culture) or different (mixed cultures),” the researchers wrote, describing this step as “similarity aware, pathogen agnostic.” This enables the system to recognize variants of the same strain, the researchers noted, and has the effect of reducing the number of strains identified by the system.
At this level, the system uses two “Siamese CNNs,” which were trained with a dataset of 200,000 image pairs.
Then, at level 4, the system “assesses the clinical significance of the entire plate,” the researchers added. Each plate is labeled as:
“Positive” (significant bacterial growth),
“No significant growth” (negative), or
“Contaminated,” meaning it has three or more “different colony morphologies without a particular pathogen that is prevalent over the others,” the researchers wrote.
If a plate is labeled as “positive,” it can be “further evaluated for possible downstream steps,” using MALDI-TOF mass spectrometry or tests to determine susceptibility to antimicrobial measures, the researchers stated.
“This decision-making process takes into account not only the identification results but also adheres to the specific laboratory guidelines to ensure a proper supportive interpretation in the context of use,” the researchers wrote.
Nearly 100% Agreement with Medical Technologists
To gauge DeepColony’s accuracy, the researchers tested it on a dataset of more than 5,000 urine cultures from a US laboratory. They then compared its analyses with those of human medical technologists who had analyzed the same samples.
Agreement was 99.2% for no-growth cultures, 95.6% for positive cultures, and 77.1% for contaminated or mixed growth cultures, the researchers wrote.
The lower agreement for contaminated cultures was due to “a deliberately precautionary behavior, which is related to ‘safety by design’ criteria,” the researchers noted.
Lead study author Alberto Signoroni, PhD, Associate Professor of Computer Science, University of Brescia, wrote in Nature that many of the plates identified by medical technologists as “contaminated” were labeled as “positive” by DeepColony. “We maximized true negatives while allowing for some false positives, so that DeepColony [can] focus on the most relevant or critical cases,” he said.
Will DeepColony replace medical technologists in clinical laboratories any time soon? Not likely. But the Brescia University study indicates the direction AI in healthcare is headed, with high accuracy and increasing speed. The day may not be far off when pathologists and microbiologists regularly employ AI algorithms to diagnose disease.
