At the center of the initiative is the K-MediST program, which will train professionals known as K-PRISM researchers—Korea-leading Physician-Scientists in Research, Innovation, Systems Medicine, and Manufacturing and Commercialization. The program is built around five core competencies: clinical problem-solving, convergent research, computation and artificial intelligence, interdisciplinary collaboration, and commercialization. Together, these capabilities are intended to address a persistent challenge in modern medicine: promising discoveries are often developed in isolation from clinical practice, making it difficult to validate technologies in real-world settings or move them efficiently into hospitals and the biomedical market.

Korea University and UNIST plan to tackle that challenge through three interconnected objectives. The institutions will create joint educational programs, establish collaborative laboratories for clinical validation and translational research, and develop an end-to-end commercialization platform to support the transition from laboratory discovery to practical medical technology. Such a structure could help shorten the path between a scientific insight and its application in diagnosis, treatment, medical devices, or healthcare systems by bringing clinicians, engineers, data scientists, and industry specialists into the same research environment from the beginning.

The partnership is also designed to respond to the growing technical complexity of healthcare. Artificial intelligence can identify patterns in medical images and clinical records, but algorithms require high-quality data, clinical expertise, and rigorous validation before they can be trusted in hospitals. Similarly, precision medicine depends on integrating biological information with patient histories, imaging, treatment outcomes, and other clinical variables. By combining UNIST’s capabilities in computation, engineering, and advanced analysis with Korea University’s strengths in clinical medicine and healthcare research, KUNIST aims to create a framework in which biomedical technologies can be evaluated under real clinical conditions rather than remaining confined to experimental settings.

The symposium opened with remarks from Tae Hoon Kim, Vice President for Research at Korea University Anam Hospital and principal investigator of the K-MediST project. Eul Sik Yoon, Executive Vice President for Medical Affairs and President of Korea University Medicine, said the collaboration could become a global model for linking medicine with emerging technologies. Sung Bom Pyun, Dean of Korea University College of Medicine, emphasized that the college had established its Center for Physician-Scientist Development in November 2025 and had built a training system spanning undergraduate and graduate education. He described the selection for the K-MediST Program as recognition of those efforts and as a step toward strengthening Korea University’s role in biomedical research and innovation.

The program’s scale is substantial. During the curriculum presentation, Kihoon Han, Chair of the Department of Biomedical Sciences at Korea University College of Medicine, announced plans to train 80 researchers between April 2026 and December 2030. The cohort will include 64 doctoral-level biomedical researchers and 16 physician-scientists pursuing MD-PhD training. By combining clinical education with research in engineering, computation, and biotechnology, the program seeks to produce specialists who can understand both the biological problem presented by a patient and the technical tools required to solve it.

A joint research center is planned for the Chung Mong-Koo Future Medical Center at Korea University’s Jeongneung Mediscience Park. The facility will unite UNIST’s computational infrastructure and high-performance analytical equipment with Korea University’s clinical research capacity. A Data Living Lab will provide researchers with access to real-time clinical data while enabling close interaction with physicians. This model could support the development of adaptive medical technologies, in which algorithms and devices are repeatedly tested against clinical needs, refined through patient data, and assessed for safety, accuracy, and practical usefulness.

The research agenda presented at the symposium reflects the breadth of the collaboration. Min Hyuk Lim of the UNIST Graduate School of Medical Science discussed medical artificial intelligence and data science, while Woo Young Jang of Korea University Anam Hospital outlined work in precision medicine and bioengineering. Hwang Kim of the UNIST Department of Design addressed digital healthcare and smart hospitals, areas in which user-centered design can determine whether a technically advanced system is actually adopted by patients and healthcare workers. Hyeon Soo Kim, Vice Dean for Academic Affairs at Korea University College of Medicine, presented plans involving medical robotics and extreme medicine, and Se Jun Oh described project management, progress monitoring, and annual evaluation for the K-MediST initiative.

Seungjae Baek, Dean of the UNIST Graduate School of Medical Science and Engineering, said the symposium offered a concrete foundation for future interdisciplinary research and expressed the institutions’ commitment to using advanced engineering technologies to address unmet clinical needs. Tae Hoon Kim concluded that Korea University’s international strengths in clinical medicine and healthcare, combined with UNIST’s scientific and engineering expertise, could establish a new benchmark for physician-scientist education in Korea. If the platform fulfills its ambitions, KUNIST will not simply train researchers in separate disciplines; it will create professionals able to move across the entire biomedical innovation chain, from identifying a clinical problem and analyzing biological or patient data to developing, validating, manufacturing, and commercializing a solution. That integrated model may prove crucial as healthcare becomes increasingly dependent on artificial intelligence, robotics, precision medicine, and large-scale clinical data.

Subject of Research: Physician-scientist education, biomedical research, medical artificial intelligence, precision medicine, bioengineering, digital healthcare, medical robotics, clinical data science, and biomedical commercialization.

Article Title: Korea University and UNIST Launch KUNIST to Train the Physician-Scientists of the Future

Keywords: KUNIST, K-MediST, Korea University, UNIST, physician-scientists, biomedical researchers, medical AI, data science, precision medicine, bioengineering, digital healthcare, smart hospitals, medical robotics, clinical research, biomedical innovation.

Hepatitis C virus represents a critical public health challenge, affecting millions of people globally. Traditional models often struggle to accommodate the intricate and dynamic nature of viral infections. The research by Sabir, Yessengaliyev, and Temirzhan introduces a cutting-edge radial basis function (RBF) neural network architecture that aims to improve predictions regarding HCV behavior. This model is not merely an attempt to refine existing methods but signifies a pivotal shift in how we approach viral modeling.

The first significant advantage of the RBF neural network lies in its ability to handle nonlinear relationships within data. Viruses like HCV exhibit rapid mutations, making them unpredictable and challenging to model accurately. By utilizing RBFs, which are well-suited for function approximation in high-dimensional spaces, the researchers have created a mechanism that can adapt to these fluctuations and yield more accurate predictions. This adaptability is crucial, especially given the viral genome’s propensity for rapid evolution.

In establishing the theoretical underpinnings of their research, the authors conducted extensive simulations that compared their RBF model’s performance against traditional linear and nonlinear models. The results were illuminating, revealing that the RBF neural structure significantly outperformed its predecessors. This performance leap is attributed to the model’s ability to interpolate complex data points and leverage local information more effectively than more conventional approaches.

Furthermore, the researchers applied their new model to real-world data sets related to HCV infection rates and treatment outcomes. The results indicate a striking correlation between their model’s predictions and observed infection dynamics. Such validation not only reinforces the model’s credibility but also its potential usefulness in public health epidemiology—providing a robust tool for policymakers and health officials.

The implications of this research extend beyond mere academic curiosity. As global health organizations strive to devise effective treatment plans, the incorporation of advanced computational models like the one presented by Sabir and colleagues could offer pivotal insights. Understanding the spread and mutation patterns of HCV can lead to more informed vaccinations, targeted therapies, and ultimately, better patient outcomes.

Moreover, this novel approach highlights the growing intersection of machine learning and virology. Researchers are increasingly recognizing that problems within biological systems can often be framed as computational challenges. The success of this RBF neural network model calls for a reevaluation of the tools we use in microbiology, hinting at a future where machine learning techniques are integral to all stages of viral research.

The findings from this research open the door to further exploration. Future studies could expand upon this model to tackle additional viral pathogens beyond HCV. By tweaking the RBF architecture and applying it to other viruses, researchers could uncover more about viral behavior, adaptive strategies, and the potential for cross-species transmissions. Each discovery could propel us closer to combating infectious diseases globally.

As we delve deeper into the era of artificial intelligence, it is essential to consider ethical implications that may arise from these advanced models. While the potential for improving health outcomes is vast, the accuracy and reliability of predictions must remain paramount. Ongoing evaluation and oversight will be crucial as we integrate such models into public health strategies and clinical applications.

The authors of this groundbreaking study are hopeful that their RBF neural network could also be adapted to assist in vaccine development. With the pressures of emerging viral strains constantly at our doorstep, the ability to model potential mutations and forecast their impact could play a crucial role in national health security. This innovative approach may thus serve as a blueprint for future interdisciplinary collaborations that fuse biology with computational sciences.

In conclusion, the comprehensive study undertaken by Sabir, Yessengaliyev, and Temirzhan marks a significant milestone in both the fields of virology and artificial intelligence. By pivoting towards a radial basis neural network, they have not only enhanced understanding of the hepatitis C virus but have also set a precedent for future research methodologies. Their work exemplifies the potential for technology to drive healthcare innovation, a necessity in an increasingly interconnected world facing multifaceted health challenges.

As this research awaits publication, the scientific community watches with anticipation, ready to engage with the insights it promises. The implications of such studies could pave the way for informed strategies, capable of tackling one of the most pressing health issues of our times, hepatitis C. The marriage of machine learning and virology stands as a beacon of hope for future healthcare advancements, embodying the spirit of innovation that could very well change the course of infectious disease management.

Subject of Research: Hepatitis C Virus Dynamics and Modeling

Article Title: Designing a novel radial basis neural structure for solving the dynamical hepatitis C virus model.

Article References:

Sabir, Z., Yessengaliyev, A., Temirzhan, A. et al. Designing a novel radial basis neural structure for solving the dynamical hepatitis C virus model.
Sci Rep (2025). https://doi.org/10.1038/s41598-025-29644-5

Image Credits: AI Generated

DOI: 10.1038/s41598-025-29644-5

Keywords: Hepatitis C virus, Radial Basis Function, Neural Networks, Viral Modeling, Artificial Intelligence, Infectious Disease Research.

SerpinB3, also recognized as squamous cell carcinoma antigen-1, has long been used in clinical diagnostics to indicate aggressive cancer growth and metastasis. For decades, elevated SerpinB3 levels in blood tests have served as ominous markers, signaling the presence of serious malignancies or inflammatory conditions where tissue barriers like skin and lungs are under duress. Despite extensive recognition in oncology, the normal physiological role of this serine protease inhibitor remained elusive until this recent study illuminated its fundamental part in skin regeneration processes.

The study, published in the Proceedings of the National Academy of Sciences, was spearheaded by ASU chemical engineers Jordan Yaron and Kaushal Rege, together with colleagues from the Biodesign Center for Biomaterials Innovation and Translation. Their research involved detailed genetic and molecular analyses of wounded skin tissues, revealing a dramatic surge in SerpinB3 expression within epithelial cells actively migrating into damaged areas. This upregulation indicates that SerpinB3 is not merely a pathological marker but an endogenous injury response element primed to accelerate tissue repair and restore barrier function.

Integral to wound closure, SerpinB3 facilitates the activation of keratinocytes—the predominant cell type in the epidermis responsible for forming new skin layers after injury. By modulating cellular adhesion properties, SerpinB3 enables keratinocytes to detach and migrate efficiently across the wound bed, a critical step in re-epithelialization. This mobility enhancement mirrors effects seen with growth factors such as Epidermal Growth Factor (EGF), yet SerpinB3’s mechanism is distinct, tied closely to its role as a serine protease inhibitor regulating proteolytic activity within the wound microenvironment.

Moreover, the presence of SerpinB3 influences the extracellular matrix remodeling that underpins effective healing. Wounds treated with agents boosting SerpinB3 expression showed markedly improved collagen fiber alignment, reinforcing tissue integrity and mechanical strength. Collagen, a primary structural protein in connective tissue, is essential for providing a scaffold that supports cell proliferation and differentiation during wound repair. Thus, SerpinB3’s involvement extends beyond cellular activation to orchestrating the biochemical milieu needed for skin restoration.

This duality of SerpinB3’s function—both in promoting tissue repair and facilitating cancerous invasion—exemplifies the complex biological balance between regeneration and disease progression. Cancer cells often hijack wound-healing pathways, exploiting proteins like SerpinB3 to enhance their invasiveness and metastasis. Understanding this interplay provides a foundation for developing targeted therapies that can either boost SerpinB3 activity to accelerate healing in patients with chronic wounds or inhibit it to neutralize tumor growth and spread in oncological settings.

Chronic wounds represent a significant healthcare burden, affecting approximately six million patients annually in the United States alone, with costs exceeding $20 billion. These non-healing wounds are especially prevalent in individuals with diabetes, advanced age, infections, or severe burns. By harnessing the natural biological role of SerpinB3, future treatments may more effectively stimulate regeneration where conventional methods have failed, offering hope for millions suffering from persistent ulcerations and pressure sores.

The innovative approach taken by the ASU team also stems from their broader research into bioactive nanomaterials designed for tissue repair. These materials have shown promise in enhancing endogenous repair signals, and the identification of SerpinB3 as a responsive factor within this context underscores its therapeutic potential. Nanomaterial-based wound dressings that amplify SerpinB3 expression could become a new standard in regenerative medicine, combining material science with molecular biology to optimize healing environments.

Further exploration of SerpinB3’s role may extend beyond the skin to other epithelial tissues such as the lungs, where chronic inflammatory diseases also cause significant morbidity. Its involvement in immune regulation and tissue homeostasis suggests that modulating SerpinB3 may prove beneficial in treating diverse pathologies including asthma and other inflammatory conditions, consolidating its position as a pivotal molecular regulator.

Jordan Yaron emphasizes the transformative nature of this discovery: “Our research has illuminated a natural injury response mechanism where SerpinB3 functions as a driver of epithelial regeneration. Recognizing this protein’s bifunctional identity—both as a cancer biomarker and as a healing facilitator—opens new avenues for therapeutic innovation.” His colleague, Kaushal Rege, who directs the Biodesign Center, notes the exciting prospects for translational research aiming to develop SerpinB3-based interventions tailored to specific medical challenges.

In conclusion, this revelation regarding SerpinB3 redefines our understanding of wound healing and cancer biology. It highlights the protein’s integral role within a finely tuned system that balances tissue repair and cellular proliferation. The finding presents compelling opportunities to manipulate this balance, promising improved clinical outcomes for patients with hard-to-heal wounds and aggressive cancers. As research advances, the full therapeutic horizon for SerpinB3 is poised to expand, potentially revolutionizing treatments across multiple medical disciplines.

Subject of Research: Animal tissue samples

Article Title: Squamous cell carcinoma antigen-1/SerpinB3 is an endogenous skin injury response element

News Publication Date: 23-Oct-2025

Web References: https://doi.org/10.1073/pnas.2415164122

References: Proceedings of the National Academy of Sciences, 10.1073/pnas.2415164122

Image Credits: Graphic by Jason Drees/ASU

Keywords: Tissue repair, Diseases and disorders, Cancer immunology, Carcinoma, Lung cancer, Metastasis, Cancer cells, Cancer genomics, Cancer research, Cancer treatments, Neoplasms, Oncology

Neonatal hypoglycemia remains one of the most common metabolic complications in the early postnatal period, posing significant risks to infant neurodevelopment when left unrecognized or untreated. Traditionally, early identification strategies rely heavily on maternal gestational diabetes mellitus (GDM) diagnosis based on OGTT performed during pregnancy. However, the absence of standardized screening in cases where mothers have incomplete or no OGTT data creates a challenging clinical scenario. This research aims to elucidate current practices and propose evidence-based guidelines to bridge this diagnostic void.

The interdisciplinary study conducted by Scholl, Ponnapakkam, Molina, and colleagues meticulously examined screening approaches across multiple healthcare centers, focusing exclusively on neonates born to mothers without sufficient glucose tolerance testing. Their work highlights a wide variability in screening timing, glucose threshold values, and follow-up protocols, underscoring the pressing need for uniform practice standards. Importantly, the authors employed rigorous retrospective data analysis complemented by prospective observational cohorts to ensure robust findings.

Central to their findings is the demonstration that infants born under these circumstances exhibit a similarly elevated risk of hypoglycemia as those born to mothers with diagnosed GDM. The metabolic derangements in neonates appear linked not only to overt maternal hyperglycemia but also to subtle disruptions in maternal glucose homeostasis not captured without comprehensive OGTT. This revelation challenges the prevailing notion that routine hypoglycemia screening can be safely omitted when maternal glucose status is unknown.

Moreover, the study documents the recurrent clinical dilemma faced by neonatal care teams: balancing the need to avoid unnecessary interventions against the imperative to prevent neuroglycopenic injuries due to missed hypoglycemia cases. The research suggests a stratified screening protocol incorporating initial risk assessment based on maternal history, fetal growth parameters, and early postnatal glucose monitoring to improve sensitivity without inundating NICUs with false positive cases.

Additionally, the authors delve into the pathophysiological mechanisms underpinning neonatal hypoglycemia. They contextualize how inadequate maternal OGTT data limits clinicians’ ability to anticipate neonatal endocrine responses. In pregnancies complicated by undiagnosed or misclassified glucose intolerance, fetal pancreatic islet cell hyperplasia and resultant hyperinsulinemia may predispose neonates to rapid postnatal glucose depletion. These insights emphasize the vital role of maternal metabolic assessment in guiding neonatal surveillance initiatives.

The study also critically evaluates the timing of initial glucose testing in the first hours after birth. While most guidelines advocate for glucose checks within the first 1-2 hours in at-risk infants, ambiguity surrounding infants born to mothers with undocumented glucose tolerance status results in heterogeneous clinical responses. The paper recommends early, frequent glucose monitoring coupled with a low threshold for intervention in this subset to mitigate neurological sequelae effectively.

Importantly, this research sheds light on the ethical and practical dimensions of neonatal screening policy—particularly in resource-limited settings. The authors argue for the development of a cost-effective, evidence-based framework that maximizes early detection rates without imposing undue burdens on healthcare resources. Their proposed algorithms aim to ensure equitable neonatal outcomes regardless of maternal testing completeness, a major step towards reducing healthcare disparities.

Cutting-edge analytical methodologies were integral to this work. The team harnessed advanced statistical modeling and machine learning techniques to identify subtle risk predictors for NH when maternal glucose tolerance data was missing. Such integration of computational tools with clinical insight exemplifies the evolving paradigm in perinatal epidemiology and highlights the potential for precision medicine approaches in neonatal care.

Furthermore, the implications of these findings extend beyond immediate neonatal management. By advocating for more rigorous maternal glucose assessment protocols during pregnancy and reinforcing neonatal screening vigilance, this study aligns with broader public health goals of reducing childhood morbidity linked to metabolic disorders. It suggests a dynamic interplay between maternal antenatal care and postnatal infant health that could influence future obstetric guidelines and neonatal screening policies globally.

The researchers also emphasize the importance of parental education and engagement in early NH screening practices. Given that mothers without comprehensive OGTT results may represent populations with limited access to prenatal care, targeted communication strategies are essential to empower families in recognizing signs of hypoglycemia and ensuring adherence to recommended screening schedules.

Clinical translation of these findings will necessitate multidisciplinary collaboration involving obstetricians, neonatologists, endocrinologists, and nursing staff. Establishing institutional protocols that account for maternal OGTT status and integrating electronic health record alerts can streamline screening workflows and enhance patient safety outcomes. The authors call for future prospective trials to validate their recommendations and further refine risk stratification models.

In conclusion, this trailblazing research by Scholl and colleagues sets a new benchmark in neonatal hypoglycemia screening by addressing a previously under-recognized high-risk group: infants born to mothers lacking adequate glucose tolerance testing. Their comprehensive, data-driven approach paves the way for standardized protocols that can prevent potential neurodevelopmental impairments stemming from undiagnosed neonatal hypoglycemia. As neonatal metabolic care continues to evolve, this study signifies a pivotal advance in safeguarding infant health through informed maternal-infant screening integration.

Subject of Research:
Neonatal hypoglycemia screening practices in infants born to mothers without adequate oral glucose tolerance testing.

Article Title:
Neonatal hypoglycemia screening practices in infants born to mothers without glucose tolerance testing.

Article References:
Scholl, J., Ponnapakkam, A., Molina, R. et al. Neonatal hypoglycemia screening practices in infants born to mothers without glucose tolerance testing. J Perinatol (2025). https://doi.org/10.1038/s41372-025-02455-2

Image Credits: AI Generated

DOI: https://doi.org/10.1038/s41372-025-02455-2

Gene-editing therapies have made headlines for their potential to address rare genetic diseases, yet such interventions frequently come with prohibitive costs and significant operational limitations. Conventional methods predominantly focus on disabling malfunctioning genes rather than directly correcting pathogenic mutations. Repair Drive emerges as a transformative alternative, not only repairing liver cells—hepatocytes—but enhancing their competitive advantage over unedited or inaccurately edited counterparts.

The implications of these findings are far-reaching. By employing the Repair Drive technique, the researchers documented an astounding rise in the rate of properly repaired hepatocytes, increasing success rates from a meager 1% to a remarkable 25% in murine liver models. This enhanced performance allows for greater cell division and thus more proficient liver regeneration—a vital aspect, given that the liver possesses inherent regenerative capabilities that exceed those of many other tissues.

At the heart of the Repair Drive methodology lies a synergistic approach utilizing small interfering RNA (siRNA) to temporarily suppress the FAH gene, essential for hepatocyte survival. By skillfully tuning this genetic switch, the team introduced a modified, siRNA-resistant version of the FAH gene along with a therapeutic gene into a select subset of hepatocytes, effectively allowing only these gene-edited cells to thrive and propagate. This innovative concept mirrors a head-start in a race, strategically positioning the gene-corrected cells to proliferate and restore liver function.

Leading the charge, Gang Bao, a prominent figure in bioengineering and a respected professor at Rice University, stated that this technical leap required not only refining existing techniques but also developing new methodologies to detect and quantify the off-target edits and various unintended modifications occurring at intended genomic sites. The complexities of achieving precision in targeted gene editing cannot be overstated, as researchers regularly grapple with issues like large deletions, unintended insertions, and even chromosomal irregularities.

Furthermore, Bao’s commitment to fostering collaborations with local Texas Medical Center partners underscores the essential nature of teamwork in revolutionary science. His leadership in initiatives such as the Baylor/Rice Genome Editing Testing Center, established in 2023, aims to facilitate engaged research and invigorate gene-editing therapy advancements nationwide, with foundational support from the National Institutes of Health.

Indeed, the Bao laboratory has been a trailblazer in the realm of gene editing, particularly in enhancing the accuracy, effectiveness, and safety of CRISPR/Cas9-based techniques. Notable endeavors have included work focused on sickle-cell disease, which is typically caused by a single-point mutation in the beta-globin gene. The lab’s current project integrates next-generation sequencing and bioinformatics to affirm precision in edits made via the Repair Drive protocol.

This commitment to broad-spectrum solutions has garnered recognition from peers, with William Lagor, a professor of integrative physiology at Baylor, emphasizing the inclusive nature of the research team that contributed to the initiative. Their unified goal is to create accessible treatments applicable to a wide array of genetic liver ailments, showcasing the intersection of diverse scientific talents in pursuit of common goals.

Marco De Giorgi, an assistant professor in Lagor’s lab and lead author on the study, received accolades from Bao for his dedication and vision in navigating complex biological and technical landscapes. This acknowledgment points to the collaborative spirit that underscores much of science’s success and highlights the critical role of research fellowship in advancing knowledge.

Associates such as So-Hyun (Julie) Park have likewise been instrumental in this endeavor, developing sequencing tools crucial for the successful execution of the project. Their partnership illustrates the confluence of various sub-disciplines within life sciences, which is often paramount to breakthroughs in complex fields such as genetics.

The extensive team involved in the research, comprising members from institutions such as BCM, Rice University, Texas Children’s Hospital, Texas Heart Institute, and Duke University, underscores the collective effort required for such ambitious scientific work. Their combined expertise brought varied perspectives to the project’s challenges, enriching the research process and enhancing the quality of outcomes.

Financial backing from prestigious organizations, including the National Institutes of Health and the American Heart Association, reflects the high value placed on this groundbreaking work by the broader scientific community. These institutions understand the significant impact that successful gene therapies could have on public health, urging continued support for research in innovative medical treatments.

The Repair Drive technology’s implications are immense, not only promising improved outcomes for patients with liver-related genetic disorders but also providing a framework that could expand the horizons of gene therapy as a whole. With existing U.S. and international patent applications pending, the potential for commercial partnerships and advancements in medical technology remains a key area of interest.

As the scientific community and the public await further developments following these exciting findings, one thing is clear: the future of gene therapy, particularly as it relates to regenerative medicine, holds transformative potential. With continued collaboration and innovation at the forefront of research efforts, the pursuit of effective treatments for genetic disorders may soon lead to groundbreaking solutions that change lives.

Subject of Research: Gene editing strategies for liver disorders
Article Title: In vivo expansion of gene-targeted hepatocytes through transient inhibition of an essential gene
News Publication Date: February 13, 2025
Web References: Rice University News
References: Science Translational Medicine
Image Credits: Photo by Gustavo Raskosky/Rice University

Keywords: Gene therapy, liver disorders, CRISPR technology, genetic editing, regenerative medicine.