Radically different
We’re bringing the whole team to DeviceTalks Boston 2026 on May 27 and 28 at the Thomas Michael Menino Convention and Exhibition Center (formerly known as the Boston Convention and Exhibition Center). We’ve got a great lineup of keynotes, interviews and panels of experts from medical device OEMs and their partners. (Go here for the full…
The post DeviceTalks Boston 2026 show preview: Speakers, exhibitors and more appeared first on Medical Design and Outsourcing.
Fighting diabetes with next-gen sensors and drug delivery devices We’ve come a long way from fingersticks for diabetes patients. Though the diabetes patient population continues to grow, I’ve never been more optimistic about medtech’s ability to take on this global epidemic. Medical devices have traditionally played a more passive role, allowing patients and physicians…
The post May 2026 issue: Fighting diabetes with next-gen sensors and drug delivery devices appeared first on Medical Design and Outsourcing.
Dr. Baoyi Huang, Southern Medical University Upper gastrointestinal abnormalities are one of the more common medical conditions, but there is a huge unmet need in the diagnosis due to a shortage of endoscopists, endoscopy equipment and venues to cater for the entire population. Our study looks into an innovative solution — an AI-integrated capsule gastroscopy…
The post Top takeaways from a new study of an AI-integrated capsule gastroscopy (ACG) system appeared first on Medical Design and Outsourcing.
After a tidal wave of blowback that culminated in a lawsuit, a nonprofit health system has reversed course in its plan to replace its Oregon emergency physicians with a national chain.
PeaceHealth’s announcement Wednesday didn’t disclose what prompted its change of heart, but those familiar with the situation say it’s because the health system’s plan was poised for defeat in a legal challenge. When PeaceHealth said in February it was cutting ties with Eugene Emergency Physicians, the local group that had staffed its Oregon hospitals for 35 years, the news drew tremendous pushback from doctors, nurses, lawmakers, mayors, and emergency medicine groups.
Then, on March 20, the Eugene emergency physicians sued, arguing that PeaceHealth’s plan to use the Atlanta-based staffing chain ApolloMD violated a new Oregon law prohibiting managed service organizations from directly owning medical practices or interfering with clinical decisions. The case has had four hearings, in which the judge was “quite clear” that the scheme violated the law, Senate Bill 951, said Hayden Rooke-Ley, an attorney who represented the doctors and a senior fellow for health care with the American Economic Liberties Project.
A research team led by scientists at the Princess Máxima Center for Pediatric Oncology, the University of Michigan Medical School, EMBL European Bioinformatics Institute, and the Institute for Systems Biology, has uncovered more than 1,700 new proteins that could have implications for human diseases, including cancer.
Mostly very small, these proteins have been discovered in what’s known as the “dark proteome,” which covers gene products from previously overlooked sections of DNA. These proteins have unusual properties, motivating scientists to coin a new concept, peptideins, to help understand their potentially unique biology. Research co-lead Sebastiaan van Heesch, PhD, a group leader at the Princess Máxima Center, commented, “We know that the current overview of recognized proteins doesn’t capture the full picture. With this study, we show that thousands of overlooked genetic sequences contribute to the dark proteome by producing a new class of protein-like molecules, microproteins, that had been missed before now. But for most of them, we don’t yet know what they do.”
Research co-lead and co-corresponding author Robert Moritz, PhD, professor and head of proteomics at the Institute for Systems Biology, further noted, “Biology has long relied on a relatively small cast of well-characterized proteins to explain the regulatory logic of the cell, but peptideins suggest that beneath that familiar layer lies an entire untapped layer of molecular actors whose functional roles in gene regulation, signaling, and cytopersistence, many we are only beginning to imagine. Given their smaller size and the diversity of cellular contexts in which they appear, I believe peptideins may prove to be among the most versatile and consequential regulatory molecules we have yet encountered in human biology. This is not the end of a search—it is the opening of a vast and fertile new territory for the entire scientific community to explore and exploit, and I look forward to seeing what the broader scientific community uncovers as these molecules, and many more that are yet to be confirmed, are brought into the light.”
Research co-lead John Prensner, MD, pediatric neurooncologist at the University of Michigan Medical School, together with Van Heesch and Moritz, are co-senior and co-corresponding authors of the researchers’ published paper in Nature titled “Expanding the human proteome with microproteins and peptideins.” The team is sharing its discoveries with scientists worldwide in an open-source format to stimulate further research.
Van Heesch added, “With growing interest in industry and academia, peptideins are at the center of multiple drug development initiatives. Similarly, we see them increasingly turning up as important players in diseases, including childhood cancers. We hope to inspire a new wave of research into peptideins and to unlock new insights and drug targets across human biology, particularly for the development of cellular immunotherapies and cancer vaccines.”
The study is the work of the TransCODE Consortium, an international collaboration of more than 60 researchers at over 30 institutions worldwide, co-led by the Princess Máxima Center for Pediatric Oncology in the Netherlands, the University of Michigan Medical School, the EMBL European Bioinformatics Institute in Hinxton, and the Institute for Systems Biology in Seattle.
Genes in DNA provide the recipe for cells to produce peptides. Historically, peptides have been called proteins if they are long enough and have existing evidence for a biological role, such as the appearance of the same protein across species in evolution. “Protein-coding genes are the bedrock of biomedical investigations, including the overwhelming majority of drug development programs,” the authors wrote. A large, curated international database of proteins contains some 19,500 entities.
But increasingly, scientists believe the traditional definition of a protein needs to be broadened. “Whether the human genome encodes substantially more than the approximately 19,500 canonical protein-coding genes has sparked a spirited debate in recent years,” the scientist continued. “Therefore, any wholesale addition of protein-coding genes creates ripple effects across human bioscience.”
Through their newly reported study the team looked at more than 7,200 previously understudied sections of the DNA called non-canonical open reading frames (ncORFs). They found that some 25% of these sections—more than 1,700—generated detectable protein-like molecules. These proteins, smaller than traditional proteins, are referred to as “microproteins.”
Generating their results involved looking at 3.7 billion individual bits of raw data that may support known and previously unknown proteins—drawing upon 95,520 experiments. “We show that about 25% of a set of 7,264 ncORFs gives rise to detectable peptides in a large-scale analysis of 95,520 proteomics experiments,” they wrote. The process took around 20,000 hours for computers to complete, working non-stop. They found 1,785 microproteins, a number that at first glance would increase the protein databases by nearly 10%.
![Predicted binding between a non-canonical open reading frame (blue) and traditional protein (yellow). [Leron Kok/Princess Máxima Center for pediatric oncology]](https://www.genengnews.com/wp-content/uploads/2026/05/Low-Res_Protein-structure-copyright-Leron-Kok-Princess-Maxima-Center-300x201.jpg)
Moritz further explained, “By deploying our battle-hardened Trans Proteomic Pipeline across nearly 100,000 mass spectrometry experiments encompassing 3.7 billion spectra—derived from the world’s collective publicly available mass spectrometry data, with the results housed within PeptideAtlas at ISB for the scientific community to view and share—we were able to confirm, with high confidence, the existence of more than 1,700 of these newly identified peptideins that would otherwise have largely remained invisible to science.”
But most of these 1,785 microproteins didn’t resemble the other 19,500 traditional proteins. For example, they were very small: 65% were fewer than 50 amino acids in length, compared to less than 1% of the 19,500 previously catalogued. Looking more closely at the microproteins the investigators saw that only a few—perhaps a dozen—resembled the traditional proteins. The team then spent more than a year trying to make sense out of the remaining bulk.
Working with protein experts from across the globe in the TransCODE consortium, the scientists coined a new biological concept, which they coined peptidein. For decades, the research community has had a binary view of the relationship between human DNA and human proteins. A given piece of DNA either does or does not produce a protein. In their new study, the scientists propose a third choice, which is that DNA could make a protein, a peptidein, or neither.
The team defined a peptidein as existing in cells as a protein-like molecule, meaning that it is made of amino acids, as are proteins. But the role of a peptidein is ambiguous. Perhaps it has a function in normal human biology, or perhaps not; this is the key distinction with traditional proteins, where all are believed to have a function in normal human biology even if the details of that function are not fully known yet. “To advance these ncORFs in biological inquiry, we invoke the emerging umbrella term of peptidein, which we define as an ORF with experimentally confirmed RNA translation and protein synthesis, but for which the data are currently insufficient to claim conventional protein-coding gene status,” the investigators stated in their report.
Importantly, this definition of peptidein leaves the door open for it to become a ‘protein’ in the future—that is, if scientists gather more evidence on it. To start exploring this idea, the team searched for peptideins without which cells cannot survive. These so-called pan-essential peptideins can be important candidate drug targets in cancer and other diseases.
Using large-scale CRISPR gene editing, the scientists found six peptideins that looked promising. For example, one of these was a peptidein produced from OLMALINC, a genetic sequence previously thought not to produce proteins. When the researchers switched this gene off, 85% of more than 485 cancer cell lines showed impaired survival. The researchers confirmed that this effect comes from the peptidein itself, not the RNA molecule it sits on, and found that it plays a role in cell division and DNA damage response. “Our work here highlights c10riboseqorf92 (in the OLMALINC transcript),” they commented. “… while we do not yet have sufficient evidence that this ncORF encodes a bona fide protein, its CRISPR-based phenotypes in the context of cancer cells are intriguing.”
Many of the newly detected peptideins are presented on cell surfaces for recognition by the immune system, making them potential targets for cancer immunotherapy. A number of such molecules presented to the immune system are already under development as drug targets, and there is growing interest from both academia and industry in exploiting this new class of cancer antigens. Peptideins could also shed light on genetic diseases that conventional gene analysis has been unable to explain, simply because genetic diagnostics were unaware that these molecules were encoded by the human genome.
Members of the consortium had previously uncovered an essential role for a microprotein, ASNSD1-uORF, in children with a high-risk form of the brain cancer, medulloblastoma. Scientists at the Princess Máxima Center are now carrying out further research to determine its role in additional pediatric cancers with the activated MYC oncogene, such as neuroblastoma.
van Heesch commented, “It felt really special to discuss and decide what to do with this new class of molecules, as we had gathered enough early evidence to suspect that they might be widespread across cell types and tissues. By classifying these molecules of unknown functionality as peptideins, we’ve given them a formal place in reference databases so the wider community can study them.”
In their paper the researchers concluded, “The extent of the undiscovered proteome is one of the central questions in human biomedicine. This work reflects the multi-consortium collaboration between the TransCODE Consortium, the HUPO-HPP/PeptideAtlas project, the HIPP immunopeptidomics project and the GENCODE gene annotation group to coalesce a generalizable approach towards understanding which ncORFs can be understood as encoding proteins … Through our efforts, we bring microproteins and alternative protein molecules into reference gene annotation by defining them as either a protein-coding gene or a peptidein, a new concept referring to confirmed protein molecules of indeterminate consequence.”
Prensner added, “We’re just beginning to see what this ‘dark proteome’ has to offer. It’s like the trailer to a movie. We see the outline of a game-changing view of human biology. We’re incredibly excited that the coming years will open new doors to help solve and treat human diseases such as cancer.”
Moritz further stated, “Our collaborative work represents a culmination of decades of investment from federal funding agencies in building the computational and data infrastructure needed to interrogate the proteome at truly unprecedented scale at the Institute for Systems Biology … What excites me most is not simply that these molecules exist, but what their existence implies.”
The researchers are making we make all ncORFs, peptides and spectra publicly available through PeptideAtlas.
The post Microproteins and Peptideins Expand Boundaries of the Human Proteome appeared first on GEN – Genetic Engineering and Biotechnology News.
Background: The application of artificial intelligence (AI) is increasingly valuable as a tool and assistant in many areas of clinical and academic medicine. Generative AI (GenAI) creates new content used by large language models, which can generate language that strongly resembles or even improves on that of humans. Learners and educators in many areas of education are using GenAI for essays and assessments, raising issues regarding learning and assessment. GenAI is also raising new concerns in health professions education (HPE), an area of health professions training that sometimes has different aims and assessment methods compared to its clinical counterparts. HPE needs to assess levels of knowledge and understanding of pedagogy, and the use of GenAI presents challenges to its current assessments, which are predominantly written. Objective: The study aimed to investigate educators’ and learners’ perspectives on the opportunities and challenges presented by GenAI in postgraduate HPE assessments. It particularly focused on perspectives of how GenAI may influence the future of assessment and essay-based assessments in HPE. Methods: Informed by a constructivist paradigm, a qualitative approach was adopted, undertaking 8 semistructured interviews conducted via Microsoft Teams. Purposive sampling ensured a mixture of educators and learners in current HPE courses from a range of health care professions. Data were thematically analyzed. Results: There was no difference between educator and learner perspectives. Four themes were identified: AI is here, students are at a disservice if we do not embrace it; AI as an opportunity to rethink HPE assessments; AI is a “gray area”; and AI is fallible. Conclusions: The findings present AI as an external catalyst, highlighting the current internal desire for assessment change within HPE. It offers opportunities for creative, authentic assessments that reflect real-life academic and clinical practice, aiming to develop competent future HPE educators and keep courses relevant. These findings contribute to the debate around the future potential and development of AI in HPE assessments.
<img src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/0c77d2e8765c4b20533fdb19cba1beac" />
The White House’s new strategy for addressing the nation’s drug crisis calls for a number of consensus public health measures: the overdose-reversal medication naloxone, medication-assisted treatment, and test strips used to detect fentanyl or other drug supply adulterants.
But the May 4 document appears to run counter to many of the Trump administration’s latest drug policy actions. In particular, it comes just days after the administration issued new restrictions on using federal dollars to distribute test strips and warned against the use of medication-assisted treatment unless accompanied by other services, like counseling.
Pancreatic ductal adenocarcinoma (PDAC) remains one of the deadliest malignancies, with five-year survival rates below 15% and more than 85% of patients diagnosed only after the disease has metastasized. The absence of reliable early detection strategies is a primary barrier to improving outcomes. Conventional imaging, including standard abdominal CT scans, typically fails to identify PDAC during its preclinical, “visually occult” stage, when curative intervention is still possible.
To address this detection gap, a team of researchers at Mayo Clinic, led by radiologist and nuclear medicine specialist Ajit Goenka, MD, has developed and validated a radiomics-based artificial intelligence model called REDMOD (Radiomics-based Early Detection Model), which can detect subtle imaging signatures of PDAC before tumors are visible. By analyzing quantitative texture and structural features embedded within routine CT scans, REDMOD identifies early biological changes associated with carcinogenesis. In a multi-institutional validation study reflecting real-world clinical conditions, the model detected 73% of prediagnostic cancers at a median lead time of approximately 16 months—nearly doubling the sensitivity of radiologists manually reviewing the same scans. Notably, detection rates were even higher more than two years prior to diagnosis, pointing toward REDMOD’s potential for make much earlier interventions possible.
REDMOD’s automated pipeline integrates advanced radiomic feature engineering, including wavelet-based analysis, and an ensemble classification approach trained to handle the low-prevalence nature of early detection. Its longitudinal stability and consistent performance across diverse imaging systems could help spur its eventual clinical adoption.
Importantly, REDMOD is designed to operate on CT scans already acquired in routine care, particularly in high-risk populations such as individuals with new-onset diabetes. This raises the possibility of embedding AI-driven risk assessment directly into existing clinical workflows, enabling opportunistic screening without additional imaging burden. If validated prospectively, such as in the ongoing AI-PACED trial, REDMOD could shift the paradigm from late-stage diagnosis to proactive detection, potentially increasing the proportion of patients eligible for curative treatment and improving survival in this otherwise lethal disease.
Inside Precision Medicine recently interviewed Goenka to provide an in-depth view of the development of REDMOD, its detection capabilities, and its potential for providing early signals of the development of PDAC.
IPM: Can you walk through how REDMOD was developed, from the initial concept to a fully automated system, and what key technical breakthroughs enabled it to detect pancreatic cancer before tumors are visible?
Goenka: The origin of REDMOD traces back to a question we asked several years ago: if pancreatic cancer is almost always lethal because we find it too late, is there information already sitting in routine computed tomography (CT) scans that we are failing to extract? We published a proof-of-concept in Gastroenterology in 2022 showing that radiomic features from the pancreas could distinguish prediagnostic CTs from controls with high accuracy. But that first-generation model had real limitations. It relied on manual pancreas segmentation, which is labor-intensive and introduces variability. It was tested at a 1:1 case-to-control ratio, which does not reflect the rarity of pancreatic cancer in any realistic screening scenario. And it used a standard classifier without mechanisms to handle severe class imbalance.
REDMOD was built to systematically address each of those barriers. The first breakthrough was automating the front end of the pipeline. We developed and validated a fully automated volumetric pancreas segmentation model based on the three-dimensional (3D) nnU-Net architecture, published separately, which removes the human bottleneck entirely. That made the system scalable; you can run it on thousands of scans without a radiologist drawing a single contour.
The second breakthrough was in feature engineering. We extracted 968 quantitative radiomic features from each segmented pancreas, then applied multi-scale image filtering using wavelet transforms and Laplacian-of-Gaussian (LoG) filters. The wavelet decomposition breaks the image into eight directional sub-bands at different spatial frequencies, allowing the model to detect textural patterns at scales that the human eye cannot resolve. We then used the Minimum Redundancy Maximum Relevance (mRMR) algorithm to distill those 968 features down to 40 that carried the most predictive information. What emerged was striking: 90% of the selected features were filter-derived, meaning the signal lives in the texture of the tissue, not in anything visible on the standard grayscale image.
The third breakthrough was the ensemble classifier. Rather than relying on a single algorithm, REDMOD combines logistic regression, random forest, and extreme gradient boosting (XGBoost) through a soft-voting mechanism. Each algorithm processes the same 40 features; their probabilistic outputs are averaged to produce the final classification. This architecture achieved the highest sensitivity among all configurations we tested, 73%, which matters enormously in a disease where missing a case is effectively a death sentence. The entire system was trained using Synthetic Minority Over-sampling Technique (SMOTE) to handle the class imbalance inherent in early detection, and validated on an independent test set with a roughly 7:1 control-to-case ratio that approximates real-world prevalence in high-risk cohorts.
The fourth breakthrough, and one that distinguishes REDMOD from models that produce a simple binary output, is the pliability of the operating threshold. REDMOD generates a continuous probability score from zero to one. We used the Youden Index to define a statistically optimized default threshold (0.41), but this threshold can be adjusted to match different clinical objectives without retraining the model. In a non-invasive triage setting, the threshold can be lowered to maximize sensitivity, catching as many cancers as possible even at the cost of more false positives. When the clinical pathway moves toward invasive procedures such as biopsy, the threshold can be raised to prioritize specificity and precision, reducing the risk of subjecting healthy patients to unnecessary procedures. This tunability means that a single trained model can serve multiple roles across the clinical cascade, from initial risk stratification through confirmatory workup.
IPM: The model relies heavily on radiomic features, particularly wavelet-filtered textures. What do these features capture biologically, and why are they better suited to detecting early pancreatic cancer than conventional imaging markers?
Goenka: Conventional imaging markers for pancreatic cancer, such as a visible mass, ductal dilation, or vascular involvement, are late manifestations. By the time you see them, the disease has typically been present for years. What we needed was a way to detect the biological processes that precede mass formation.
Radiomic texture features quantify the spatial relationships between voxels, which are the three-dimensional equivalent of pixels. They measure how intensity values co-occur, how they cluster, and how uniform or heterogeneous the tissue appears at different scales. Specifically, features derived from the Gray-Level Co-occurrence Matrix (GLCM) measure local patterns of intensity variation; Gray-Level Size Zone Matrix (GLSZM) features capture the distribution of connected regions of similar intensity; and Gray-Level Dependence Matrix (GLDM) features quantify how dependent each voxel’s value is on its neighbors. These are mathematical descriptions of tissue microarchitecture.
The wavelet filtering is what makes this work in the prediagnostic setting. A wavelet transform decomposes the image into sub-bands that isolate different spatial frequencies and directions. This allows the model to detect textural disruptions across multiple scales: fine-grained changes that might reflect early stromal remodeling or desmoplastic reaction, and coarser patterns that could correspond to alterations in parenchymal organization. When we performed ablation studies, models built from filtered features alone matched the full REDMOD performance (area under the receiver operating characteristic curve [AUC] of 0.82), while models restricted to unfiltered features dropped to 0.74. That 8-point difference was statistically significant and tells us that the prediagnostic signal is fundamentally a multi-scale textural phenomenon.
Biologically, this aligns with what we know about early pancreatic carcinogenesis. Before a mass forms, the tumor microenvironment undergoes extracellular matrix remodeling, fibrotic changes, and shifts in cellular density that alter tissue texture at microscopic scales. These changes are invisible to a radiologist reading the scan on a monitor, but they leave a quantitative fingerprint in the image data. That fingerprint is what REDMOD reads.
IPM: How did you assemble the training dataset, and why was it important to simulate a low-prevalence, real-world screening environment?
Goenka: Assembling the dataset was one of the most labor-intensive aspects of this work, because prediagnostic CT scans are inherently rare. These are scans obtained for unrelated clinical reasons in patients who were later diagnosed with pancreatic cancer, but at the time of the scan, the pancreas appeared entirely normal on radiology review. We identified 219 such patients across the Mayo Clinic enterprise, with scans obtained three to 36 months before histopathologic diagnosis. Each was verified by expert radiologists to confirm the absence of any discernible pancreatic abnormality.
The control cohort comprised 1,243 patients whose CT scans showed a normal pancreas and who remained cancer-free for at least three years of follow-up. That three-year washout period was essential; without it, you risk contaminating the control group with patients who had undetected cancer at the time of their scan.
We then split the full cohort into 969 training cases and 493 test cases, with the test set held completely independent. The resulting control-to-case ratio of approximately 7:1 was a deliberate design choice. Most artificial intelligence (AI) studies in this space have used balanced 1:1 ratios, which inflate performance metrics and do not reflect the reality of early detection. In any high-risk cohort you would screen clinically, for example patients with new-onset diabetes and elevated Enriching New-Onset Diabetes for Pancreatic Cancer (ENDPAC) scores, pancreatic cancer prevalence is roughly 3-4%. If you train and test your model at 1:1, you get numbers that look strong in a paper but collapse when deployed in a real population. We wanted REDMOD’s reported performance to approximate what a clinician would actually experience.
IPM: You validated the model across multiple institutions, imaging systems, and external datasets. What were the biggest challenges in ensuring consistent performance across such heterogeneous data?
Goenka: The central challenge is that CT scans are not standardized. Different hospitals use different scanners from different manufacturers, different acquisition protocols, different reconstruction algorithms, and different contrast timing. All of these affect the pixel-level values that radiomic features depend on. A model that works well on data from one scanner can fail on data from another.
We addressed this at multiple levels. First, our prediagnostic cohort was inherently heterogeneous. 71% of the prediagnostic CTs in the test set were acquired at external institutions, not at Mayo Clinic. These scans came from a range of scanners (Siemens, GE, Toshiba, Philips) and clinical settings. Second, we validated specificity on two independent external cohorts: a multi-institutional dataset drawn from the Mayo Clinic enterprise across multiple campuses, and the National Institutes of Health Pancreas CT (NIH-PCT) dataset, which is a publicly available benchmark that uses entirely different acquisition parameters. REDMOD achieved 87.5% specificity on the NIH-PCT dataset, data the model had never encountered and that was acquired under conditions completely outside our control.
Third, we performed a longitudinal test-retest analysis. For patients with serial CT scans, we assessed whether REDMOD produced consistent predictions across time points. The concordance rate was 90-92%, meaning the model’s output was stable despite natural variations in patient hydration, contrast timing, and physiologic state between scans. That kind of temporal stability is essential for any tool used in a surveillance context, where you need to trust that a change in the model’s output reflects a real biological change, not scanner noise.
IPM: How do you see REDMOD being integrated into existing clinical workflows, for example in evaluating incidental CT scans or screening high-risk groups like patients with new-onset diabetes?
Goenka: The population where this has the most immediate clinical relevance is individuals with glycemically-defined new-onset diabetes (gNOD) and an ENDPAC score of three or higher. This is a well-characterized high-risk group with a 3-4% short-term risk of developing pancreatic cancer, roughly 20 times the general population rate. Many of these patients already receive CT scans for other clinical indications. The question is not whether to scan them; the question is whether we are extracting all the information those scans already contain. We were not. REDMOD changes that.
The workflow we envision is not a population-wide screening program. It is a targeted, risk-stratified approach. An electronic medical record (EMR)-based algorithm identifies patients who meet gNOD and ENDPAC criteria. When those patients undergo a CT scan, either for clinical reasons or as part of a structured surveillance protocol, REDMOD runs in the background, analyzes the pancreas automatically, and generates a risk score. If the score exceeds a defined threshold, it triggers a clinical pathway: the referring physician is notified, and the patient enters a structured workup that could include enhanced imaging, molecular imaging with fibroblast activation protein (FAP)-targeted positron emission tomography (PET) radiotracers, or closer follow-up.
REDMOD does not replace the radiologist. The radiologist reads the scan according to standard practice and generates their clinical report independently. REDMOD operates as a parallel, complementary layer, a second opinion from a system that reads data the human eye cannot access. The physician integrates both sources of information to make clinical decisions.
This is precisely the model we are testing in the AI-PACED (Artificial Intelligence for Pancreatic Cancer Early Detection) prospective clinical trial at Mayo Clinic. In this trial, all CT scans are interpreted by non-study radiologists who are blinded to the study objectives, and their reports enter the patient’s medical record as part of routine clinical care. Independently, the AI analysis is performed on de-identified data on secure research servers. A strict firewall separates the two: AI-generated outputs are not integrated into the EMR, are not communicated to the clinical team, and are not used to guide diagnosis or treatment. This dual-layered design ensures that participants receive the benefit of structured clinical surveillance while allowing a blinded, independent evaluation of the AI’s performance.
IPM: With the AI-PACED prospective trial underway, what are the key questions you still need to answer about clinical utility, false positives, and patient outcomes before this technology can become part of standard care?
Goenka: There are several questions that retrospective data alone cannot answer, and AI-PACED is designed to address them.
The first is lead-time advantage. We know REDMOD detects prediagnostic signal at a median of 475 days before clinical diagnosis in retrospective data. The question is whether that lead time translates into an actual shift in diagnostic timing in a prospective setting, that is, whether patients in a structured AI-augmented surveillance protocol receive their diagnosis earlier, and at a more resectable stage, compared to patients receiving symptom-driven standard care. The trial’s primary endpoint is the time-to-diagnosis from gNOD onset, compared between the interventional and observational cohorts using Kaplan-Meier survival analysis and Cox proportional hazards modeling.
The second is false positives. In the retrospective validation, REDMOD had an 81% specificity, which means approximately 19% of healthy patients received a positive flag. In a low-prevalence screening population, even a modest false positive rate generates a meaningful number of patients who undergo additional workup for a cancer they do not have. AI-PACED will quantify the downstream diagnostic burden, including additional imaging studies, biopsies, and the psychological impact, so we can make an honest assessment of the risk-benefit tradeoff. It is worth noting that REDMOD’s precision of 36.2% at its default operating point already exceeds the 3% precision threshold recommended by the United Kingdom’s National Institute for Health and Care Excellence (NICE) at the first step of cancer referral, and established screening programs for lung and breast cancer accept similar tradeoffs at their initial triage steps.
The third is adherence. This is a surveillance protocol in asymptomatic people. They feel fine. Asking them to return for serial CT scans and blood draws over 12 months requires trust, and that trust has to be earned through transparency about what we know and what we do not know. AI-PACED will measure recruitment yield from EMR-identified high-risk individuals, retention rates across the imaging and biobanking protocol, and the practical challenges of integrating AI into existing radiology workflows without disrupting standard care.
The fourth, and perhaps most important for the long term, is whether earlier detection actually changes outcomes. Stage shift, moving a patient from stage IV to stage I or II, is necessary but not sufficient. We need evidence that patients diagnosed through AI-augmented surveillance live longer, have access to curative surgical resection, and experience better quality of life. That is the bar this technology must clear, and it is the bar we intend to hold ourselves to.
The ongoing phase of AI-PACED is a feasibility study. It is designed to generate the operational, logistical, and preliminary clinical data needed to justify and design a fully powered, multi-institutional trial. In addition, we are running in silico clinical trials and cost-effectiveness analyses. We are building the evidence base one layer at a time, because the stakes, for patients and for the credibility of AI in clinical medicine, are too high to cut corners.
The post Mayo Clinic’s REDMOD AI Doubles Early Detection Sensitivity in Pancreatic Cancer appeared first on Inside Precision Medicine.