Seoul National University Hospital

Seoul National University Hospital researchers have come up with an AI model that can both classify and explain bone densities in chest X-ray images. WHAT'S IT ABOUT The research team collected data from approximately 14,502 women who underwent chest X-ray (CXR) and bone density test (DXA) examinations at SNUH Health Promotion Center between 2004 and 2019. These were analysed using four different AI foundation models – two trained on general images and two trained on medical images. In each model, three validation approaches were applied: linear validation, partial fine-tuning, and low-rank adaptation, to assess predictive performance. Each model identifies whether a CXR shows normal bone density, osteopenia, or osteoporosis by assessing relevant features, such as the spine and ribs, and then comparing them with its learned patterns. An explainability system was also developed to confirm the highlighted area in the CXRs that forms the basis of the AI's judgment. It quantitatively verifies whether the AI makes judgments based on clinically significant bone structures. Findings published in Osteoporosis International noted that one of the models trained on general images (DINOv2) and validated through low-rank adaptation showed the highest predictive performance among models, with an area under the curve of 93%. WHY IT MATTERS The study's findings suggested AI's potential for opportunistic screening to detect osteoporosis – characterised by low bone mass and weakened bone structure – early without the need for a separate bone density test. Key to this, the research team noted, is verifying its basis of judgment. The research team specifically sought to address black box problems in existing AI models through their research "When applying a foundation model to medical images, high performance alone is not enough. A multidimensional evaluation system needs to be trusted in actual medical settings," said study first author Kim Jae-won, a researcher at the SNU Department of Medicine. "This study is significant in that it presents the criteria for such a system," he claimed. It also "provides guidance on how to select and utilise foundation models," added SNUH professor Park Sang-min, who led the study. MARKET SNAPSHOT A local medical AI company, Promedius , obtained regulatory approval from the Ministry of Food and Drug Safety for its osteoporosis diagnosis support AI software. Its solution also helps identify osteoporosis from a chest X-ray. Taiwan-based Acer Medical also received similar approval from the Indonesian Ministry of Health last year for its AI-assisted software, which analyses bone mineral density and predicts T-scores in X-ray images. The National University Health System in Singapore has also developed an AI that flags in real-time cases of hypercalcemia , which can lead to osteoporosis.

Seoul National University Hospital and internet conglomerate Naver have created an AI model that can calculate a person's biological age and predict their health risks using their health checkup data. HOW IT WORKS A team of researchers from the SNUH Department of Endocrinology and Metabolism and Naver Digital Healthcare Lab gathered data from approximately 151,281 people who had their checkup at SNUH Gangnam Center in 2003-2020. Data included body measurements, blood and urine tests, pulmonary function tests, diseases, and death information. Patients were classified into three groups: normal, pre-disease, and disease, based on their blood sugar, blood pressure, and cholesterol levels. The team trained a transformer-based deep learning model using this dataset. Besides calculating biological age, the model was also trained to assess the statistical correlation between current health status and future survival rates. Separate gender-specific models were also trained to account for physiological differences between men and women. According to a media release, the AI estimates an individual's biological age by integrating and analysing various health indicators, such as blood pressure, blood sugar, lung function, and cholesterol. It then calculates the difference between the biological age and the actual or chronological age. The AI also checks whether a person's present health indicators match groups in the training dataset, and then provides its predictions. FINDINGS Citing findings published in the Journal of Medical Internet Research, researchers said the AI model "clearly distinguished" between normal, pre-disease, and disease groups. "The normal group showed a lower biological age than the chronological age, while the disease group showed a higher biological age, demonstrating a clear difference according to health status," SNUH noted in a press release. "The gap [between biological and chronological ages] widened as blood sugar, blood pressure, and lipid levels decreased, and in the presence of cardiovascular disease or cancer," it added. Meanwhile, by getting the difference between biological and chronological age, the researchers were able to stratify individuals into healthy, reference, and unhealthy groups and conduct a Kaplan-Meier survival analysis. The analysis, they said, showed that the larger the difference between biological and chronological age, the higher the actual death risk, given that the survival rate in the unhealthy group across both men and women cohorts was significantly lower than that in the healthy group. WHY IT MATTERS The study claims to be the first to develop a transformer-based AI model for calculating biological age – or the age that accounts for a person's overall health status – that also learns disease prevalence and mortality. According to SNUH, existing biological age models primarily rely on data from healthy individuals, "making them difficult to apply to patients with chronic diseases and failing to reflect mortality risk." These existing models, like the Klemera and Doubal method and chronological age cluster, also "could not consistently distinguish" among patient cohorts based on biological and chronological age, present health status, and future disease survival rates. The SNUH-Naver study suggests the potential of using the AI model's output to assist doctors in developing personalised health risk management and disease prevention strategies. THE LARGER TREND Research announced early in June in Singapore also developed an algorithm-based biological age model that predicts mortality and health outcomes using only clinical data. The model, created by a team of researchers from the National University of Singapore Yong Loo Lin School, was built on two earlier models and trained using a public medical dataset from the United States. Another biological age model in Singapore uses retinal images and can also predict a person's 10-year disease and mortality risk.

YONGIN, South Korea, Nov. 3, 2025 /PRNewswire/ -- GC Biopharma, a leading global pharmaceutical company based in South Korea, announced today that the Phase 2 clinical trial results for its anthrax vaccine 'BARYTHRAX inj. (GC1109)' have been published in the international journal Vaccine. BARYTHRAX is the world's first recombinant anthrax vaccine, jointly developed by GC Biopharma and the Korea Disease Control and Prevention Agency (KDCA). This randomized, double-blind, placebo-controlled Phase 2 clinical trial, approved by the Ministry of Food and Drug Safety (MFDS) was conducted to evaluate the safety and immunogenicity of the vaccine. A total of 240 healthy adult subjects were recruited across five hospitals in Korea: Seoul National University Hospital, The Catholic University of Korea St. Vincent's Hospital, Korea University Ansan Hospital, Soon Chun Hyang University Hospital Seoul, and Hallym University Sacred Heart Hospital. After administration of either BARYTHRAX or a placebo, the clinical research team assessed the participants for antibody production and any adverse reactions. The Phase 2 clinical trial demonstrated that the vaccine elicited antibodies that neutralized the anthrax toxin at a level exceeding the pre-defined protective threshold. This finding indicates robust immunogenicity and suggests that BARYTHRAX can provide clinically meaningful protection against anthrax infection. In addition, safety assessments showed that the adverse events in the BARYTHRAX group, such injection site pain, myalgia, fatigue, and malaise, were mostly mild and temporary. No cases of acute adverse reactions or deaths were reported, confirming a favorable safety profile for the vaccine. GC Biopharma emphasized that, BARYTHRAX has the significant advantage of being more safe than traditional methods that use live attenuated bacteria and allowing for rapid mass production when needed. Dr. Nam Joong Kim, the principal investigator of the clinical trial and a professor at Seoul National University Hospital, stated, "This clinical research exemplifies successful academia-industry collaboration and represents a significant milestone in the development of safe and effective vaccines." He added, "We anticipate that the successful development of Korea's first recombinant anthrax vaccine will serve as a stepping stone for advancing research on infectious diseases and boosting vaccine localization." "We are thrilled to have our study result published in a globally renowned academic journal Vaccine, once again recognizing its clinical significance and value," said Dr. Chang Kyung Kang of Seoul National University Hospital, the lead author of the publication. "The clinical trial has successfully demonstrated the efficacy and safety of BARYTHRAX," commented Jae-Uk Jeong, Head of R&D at GC Biopharma. "This domestically developed anthrax vaccine is poised to make a significant contribution to strengthening our national pandemic preparedness." [Reference] Barythrax Inj. This vaccine is the world's first recombinant protein-based anthrax vaccine and Korea's 39th New Drug and 3rd New Drug Vaccine. The vaccine is produced using recombinant technology to express and purify solely the Protective Antigen (PA) protein from Bacillus anthracis, the key component of the anthrax toxin. Anthrax Anthrax is a zoonotic disease that can infect both humans and animals. It is classified as a potential bioterrorism agent as the aerosolized anthrax spores can cause fatal respiratory infections. About GC Biopharma GC Biopharma (formerly known as Green Cross Corporation) is a biopharmaceutical company headquartered in Yong-in, South Korea. The company has over half a century of experience in the development and manufacturing of plasma derivatives and vaccines, and is expanding its global presence with successful US market entry of Alyglo™(intravenous immunoglobulin G) in 2024. In line with its mission to meet the demands of future healthcare, GC Biopharma continues to drive innovation by leveraging its core R&D capabilities in engineering of proteins, mRNAs, and lipid nanoparticle (LNP) drug delivery platform to develop therapeutics for the field of rare disease as well as I&I (Immunology & Inflammation). To learn more about the company, visit This press release may contain biopharmaceuticals in forward-looking statements, which express the current beliefs and expectations of GC Biopharma's management. Such statements do not represent any guarantee by GC Biopharma or its management of future performance and involve known and unknown risks, uncertainties, and other factors. GC Biopharma undertakes no obligation to update or revise any forward-looking statement contained in this press release or any other forward-looking statements it may make, except as required by law or stock exchange rule. GC Biopharma Contacts (Media) Sohee Kim [email protected] Yelin Jun [email protected] Yoonjae Na [email protected] SOURCE GC Biopharma 21% more press release views with Request a Demo