-
Development of the Next-Generation Transfer Learning Technology Solving High-Dimensional Data Analysis Challenges
Professor Eun Ryung Lee (Department of Statistics), as a first author, has developed a new statistical methodology to overcome the limitations of high-dimensional analysis caused by data scarcity. In collaboration with Professor Seyoung Park (Yonsei University) and Professor Hongyu Zhao (Yale University), Prof. Lee successfully implemented a 'Transfer Learning Algorithm' that maximizes learning performance by selectively utilizing useful information, based on the insight that the contrast between target data and external source data exhibits a 'Low-rank' structure. This achievement paves the way for dramatically improving prediction accuracy in fields such as rare disease research and precision medicine, where analysis has been difficult due to small sample sizes, by effectively integrating external big data. ■ Innovative Algorithm Design Overcoming Limitations of Existing Transfer Learning This study focused on resolving the predictive uncertainty of 'Small Data', which persists even in the big data era, and the side effects of existing transfer learning. In high-dimensional regression problems like genomic analysis, accurate model estimation is difficult because the number of variables reaches tens of thousands while the target samples of interest are very few. To complement this, transfer learning utilizing external data has been attempted, but 'Negative Transfer' problems, where prediction performance degrades due to the indiscriminate use of data irrelevant to the target, have frequently occurred. To solve these problems, the research team proposed a two-step estimation method that effectively controls the structural difference between the target model and the source model within a 'Low-Rank Regression' framework. In particular, the 'Forward Source Detection (FSD)' technique devised by the team sequentially detects only those information sources among numerous external datasets that practically help target analysis. This amplifies common signals between data and blocks unnecessary noise, enabling precise estimation without bias even in high-dimensional environments. ■ Proven Superior Prediction Performance and Theoretical Optimality Theoretical verification proved that the newly developed transfer learning methodology has a much faster statistical convergence rate than using target data alone and achieves optimal efficiency from a Minimax perspective. Its superiority was also confirmed in actual data application. The research team conducted an experiment predicting anticancer drug responses of specific lung cancer mutations (KRAS-mutant NSCLC), which had only 28 samples, using Cancer Cell Line Encyclopedia (CCLE) data. As a result, the proposed algorithm recorded significantly higher prediction accuracy compared to existing pooled analysis methods or simple marginal screening methods by effectively selecting and integrating data from other cancer types with similar genetic characteristics to lung cancer. ■ Applicability to Various Fields The 'Forward Source Detection Transfer Learning (FSD-Trans-NR)' technology of this study is designed to operate stably even in high-dimensional environments where the data dimension is much larger than the sample size, and can be flexibly applied to complex data situations where low-rank structures and sparse structures are combined. These characteristics are expected to be widely utilized for predictive modeling in various fields, such as financial risk analysis and new material development, where data acquisition is difficult and costly, as well as drug response prediction in the biomedical field. This research was supported by the National Research Foundation of Korea (NRF) and the U.S. National Institutes of Health (NIH). This research outcome was published online in October 2025 in the Journal of the American Statistical Association (JASA), the world's most prestigious journal in the field of statistics. ※Title: Transfer Learning Under Large-Scale Low-Rank Regression Models ※Journal: Journal of the American Statistical Association (JASA) ※DOI: https://doi.org/10.1080/01621459.2025.2555057
- No. 350
- 2026-01-06
- 1134
-
Professor Yohan Kim at SKKU Develops “Periportal Liver Assembloids” That Fully Recapitulate Human Liver Tissue In Vitro
Professor Yohan Kim of the Department of MetaBioHealth at Sungkyunkwan University (President Yoo Ji-Beom), in collaboration with the Max Planck Institute of Molecular Cell Biology and Genetics (MPI-CBG) in Germany, has successfully developed human periportal liver assembloids that precisely reproduce the periportal region of the human liver outside the body. This achievement was recognized for its scientific significance and published on December 17, 2025 in Nature, the world's most prestigious journal in life sciences. The liver-often referred to as the body's "chemical factory"-is responsible for essential functions such as metabolism, detoxification, and bile production. Until now, researchers have relied on organoids-miniature organ-like structures grown from stem cells-to study liver disease in the laboratory. However, conventional liver organoids have been limited in their ability to reproduce the complex interactions among diverse liver cell types, making it difficult to reflect the liver's highly sophisticated architecture and function in the human body. To overcome these limitations, Professor Kim utilized patient-derived liver tissue obtained during surgery. The researchers established a technology that enables the direct expansion of mature human hepatocytes in vitro and successfully generated hepatocyte organoids. These organoids formed functional bile canaliculi, retained long-term metabolic activity, and demonstrated drug detoxification and energy metabolism comparable to the human liver. Taking the approach further, the team combined hepatocyte organoids with bile duct organoids and periportal fibroblasts derived from the same patient. By assembling these components in a highly controlled three-dimensional architecture-much like building with LEGO blocks-they created periportal liver assembloids. The term assembloid refers to a next-generation engineered tissue constructed by assembling multiple cell types or organoids into a unified functional structure. The resulting assembloids faithfully reproduced the periportal region of the human liver, where hepatocytes, bile ducts, and blood vessels converge and active molecular exchange occurs. Transcriptomic analysis revealed that the assembloids carried out complex liver-specific functions such as gluconeogenesis and urea metabolism. Notably, the team also demonstrated hepatic zonation, in which cells adopt different functional identities depending on their spatial location-one of the most defining features of the native liver. This breakthrough has profound implications for reducing animal experimentation and accelerating patient-specific therapies. By artificially increasing fibroblast populations within the assembloids, the team generated a liver fibrosis model, in which pathological features such as collagen deposition and cell death observed in cirrhotic patients could be faithfully reproduced and studied in vitro. Professor Kim stated, "This is the world's first demonstration that multiple patient-derived liver cell types can be assembled into a single functional tissue that recapitulates the structural and pathological complexity of the human liver in the laboratory. Going forward, this platform will enable the development of new treatments for liver fibrosis, biliary diseases, and liver cancer, and will allow patient-specific drug testing as a precision medicine tool." This study was supported by the Max Planck Society, the German Federal Ministry of Education and Research (BMBF), and the European Research Council (ERC). ※ Title: Human assembloids recapitulate periportal liver tissue in vitro ※ Journal: Nature (Published on 17th of December, 2025) ※ DOI: https://www.nature.com/articles/s41586-025-09884-1 ※ Pure: https://pure.skku.edu/en/persons/yohan-kim/ ▲Schematic of human hepatocyte-derived organoid generation and periportal liver assembloid
- No. 349
- 2025-12-30
- 3040
-
Development of a Biomechanobiology-Based Cardiac Regeneration Bioplatform for Myocardial Infarction Treatment and Tendon
Professor Geun Hyung Kim’s research team (first author: Dr. Wonjin Kim) in the School of Medicine has successfully developed a biomimetic gradient tissue construct that continuously recapitulates the tendon–tendon-to-bone interface (TBI)–bone architecture using 3D bioprinting technology, in order to address the clinical challenge of regenerating the tendon-to-bone interface in rotator cuff tears. The team fabricated two types of tissue-specific bioinks based on decellularized extracellular matrices derived from bone and tendon tissues. To precisely reproduce the native microenvironment, hydroxyapatite was incorporated into the bone region, while physical and biochemical cues that induce cellular alignment were applied to the tendon region. In addition, by employing a core–shell nozzle–based gradient bioprinting process, the researchers successfully established a construct exhibiting seamless biological and mechanical continuity from tendon to TBI to bone within a single printing step. The resulting 3D gradient construct provided precise tissue-specific signals that guided human adipose-derived stem cells (hASCs) toward tendon, fibrocartilage, and bone lineages, and notably enhanced fibrocartilage formation within the TBI zone. In vitro assessments further demonstrated that the gradient architecture significantly increased TBI-related gene expression, cytoskeletal organization, and mechanical strength compared with conventional single-tissue models. Collaborative in vivo studies with Professor Sang Cheol Lee’s team in the Department of Rehabilitation Medicine at Yonsei University confirmed that implantation of the gradient construct in a rabbit rotator cuff tear model resulted in robust and continuous regeneration of tendon, TBI, and bone tissues. Professor Kim emphasized, “The tissue-specific bioinks and gradient bioprinting platform introduced in this study represent an important technological breakthrough that overcomes the long-standing challenges associated with the complex structural and mechanical discontinuities of the tendon–TBI–bone interface, offering a promising therapeutic strategy for difficult-to-treat tissue defects such as rotator cuff tears.” The research team further extended this gradient tissue regeneration concept to develop a next-generation cardiac and skeletal muscle regenerative bioplatform that actively harnesses mechanotransduction. Myocardial infarction (MI), in particular, is a representative disease that causes irreversible damage to cardiac tissue, and existing cell-based therapies or bioprinting-based approaches have faced inherent limitations, including restricted cellular responsiveness and insufficient paracrine effects. To overcome these challenges, the team fabricated a bioprinted cellular patch composed of cardiomyocytes, cardiac fibroblasts, and endothelial progenitor cells embedded in a collagen matrix incorporating aligned gold nanowires (AuNWs). Through systematic optimization of AuNW concentration, bioprinting process parameters, and the mixing ratios of the three cell types, stable fabrication of a 3D cardiac patch containing aligned AuNWs was achieved. In vitro analyses demonstrated enhanced cellular alignment, activation of integrin-mediated signaling, increased focal adhesion kinase (FAK) formation, and robust secretion of diverse paracrine factors. These synergistic effects effectively promoted the formation of vascularized cardiac tissue during the culture of the bioprinted cardiac patch. Furthermore, implantation of the 3D cardiac patch into an animal model of myocardial infarction resulted in significantly enhanced angiogenesis and myocardial regeneration, with clear evidence of functional cardiac recovery driven by paracrine mechanisms. In parallel, the research team designed a magnetorheological bioink and established a magnetic field–based mechanobiology bioprinting platform that enables real-time magnetic stimulation during the printing process. In addition, they developed a fabrication process for highly aligned, mechanically reinforced, cell-laden collagen filaments using a catalyst-free collagen peptide bonding technology. These technologies are regarded as next-generation regenerative strategies capable of simultaneously addressing the long-standing challenges of low mechanical stability and insufficient tissue maturation in cardiac and skeletal muscle regeneration. This research was supported by the Ministry of Science and ICT, the National Research Foundation of Korea, and the Korea Disease Control and Prevention Agency. In recognition of its scientific excellence, the outcomes of this work were published in leading international journals, including *Bioactive Materials (IF = 20; development of a tendon–TBI–bone gradient regeneration platform), **Chemical Engineering Journal (IF = 13; bioprinted cardiac regeneration patch), ***Bioactive Materials (development of magnetic field–based bioprinting technology), and ****Advanced Science (IF = 14.3; fabrication of high-strength aligned collagen filament technology). Title: 3D bioprinted multi-layered cell constructs with gradient core-shell interface for tendon-to-bone tissue regeneration. Bioactive Materials 43 (2025) 471–490 Journal: Bioactive Materials DOI: https://doi.org/10.1016/j.bioactmat.2024.10.002 Pure: https://pure.skku.edu/en/persons/geunhyung-kim/ **Bioprinting of cardiac patches with gold-nanowires and tri-culture system for the treatment of myocardial infarction, Chemical Engineering Journal 526 (2025) 171562 ***In situ magnetic-field-assisted bioprinting process using magnetorheological bioink to obtain engineered muscle constructs, Bioactive Materials 45 (2025) 417–433 ****Catalyst-Free Collagen Filament Crosslinking for Engineering Anisotropic and Mechanically Robust Tissue Scaffolds, Adv. Sci. (2025) e14319 Figure. Schematic of the Gradient 3D Bioprinting Process for Tendon–TBI–Bone Composite Tissue Fabrication and the Corresponding In vitro/In vivo Outcomes
- No. 348
- 2025-12-22
- 9002
-
Sungkyunkwan University Introduces Economics into Biofoundry Operations, Accelerating Innovation in Biomanufacturing
On the 3rd of December, SKKU's Biofoundry Research Center, led by Professor Han Min Woo of the Department of Food Science and Biotechnology, has announced that they have established a new biofoundry system that maximizes the efficiency of laboratory automation by incorporating economic principles, thereby accelerating the development of technologies for producing high-value biobased materials. This study has been widely recognized for going beyond simple robot-assisted automation by introducing an evaluation model that quantitatively analyzes cost and time efficiency, thereby opening a new paradigm in the field of biomanufacturing. The research team integrated the Experiment Price Index (EPI), which enables an at-a-glance assessment of experimental economics, with the concept of Robot-Assisted Modules (RAMs) that can be assembled like Lego blocks. While conventional biofoundries have primarily focused on executing experiments using robotic automation, the team mathematically calculated the cost and time of each process to design optimal automation pathways that eliminate unnecessary steps and maximize performance. Using this system, the research team established automated workflows for five core bioprocesses, including gene assembly and microbial genome editing, and applied them to the production of compounds with high industrial value. As a result, the team successfully achieved the rapid development of microbial strains capable of producing cannabigerolic acid (CBGA), a key component of medical cannabinoids, as well as the functional amino acid L-tryptophan. Notably, the system demonstrated markedly superior speed and accuracy compared with manual experimentation, thereby validating its potential for commercialization. In addition, the research team presented a techno-economic analysis of biofoundry operations under large-scale project scenarios. The analysis provides concrete data showing that, with an annual utilization rate of 50–75%, the initial capital investment can be recovered within approximately five years, offering practical evidence to support investment decision-making by both industry and government. This achievement is expected to serve as a foundation for future advancements toward a fully autonomous “self-driving laboratory,” in which artificial intelligence (AI) independently designs and executes experiments based on accumulated operational data. This achievement is expected to serve as a foundation for future advancements toward a fully autonomous “self-driving laboratory,” in which artificial intelligence (AI) independently designs and executes experiments based on accumulated operational data. This achievement is expected to serve as a foundation for future advancements toward a fully autonomous “self-driving laboratory,” in which artificial intelligence (AI) independently designs and executes experiments based on accumulated operational data. Meanwhile, the research findings were published in the December 1 online edition of Trends in Biotechnology, a leading international journal in the field of biotechnology. This work was supported by the National Research Foundation of Korea, the Ministry of Science and ICT, and other supporting organizations, including Daesang Corporation. ※ Title: Techno-economic assessment-guided biofoundry for microbial strain development ※ Journal: Trends in Biotechnology (Published by Cell Press; Impact Factor 14.9; top 1.4%) ※ DOI: https://doi.org/10.1016/j.tibtech.2025.11.002 ※ Authors: Han Min Woo, Professor (Corresponding Author, SKKU), Yu Been Heo, Ph.D. (First Author, SKKU) ※ Pure: https://pure.skku.edu/en/persons/han-min-woo/ ※ Video Overview: https://youtu.be/5Yle6oRfBl0 ▲ Automated workflow optimization development by the Experiment Price Index and Techno-economic assessment of biofoundry
- No. 347
- 2025-12-16
- 5870
-
Detecting Unknown Defects in Semiconductor Manufacturing Processes Using Artificial Intelligence
Wafer Bin Map (WBM) is a map that visualizes and shows the defective quality/defect status and defective type information of individual semiconductor chips obtained through electrical test results in the semiconductor manufacturing process. As modern semiconductor manufacturing processes become increasingly miniaturized at the nanoscale, accurately detecting defect patterns appearing in the WBM and rapidly identifying their causes are critical challenges for improving semiconductor yield and quality control. While deep learning technology has enabled attempts at automating defect classification, existing supervised learning-based methodologies have limitations in that they only function for predefined defect types. This leads to problems where new types of “unknown defect patterns” arising from product diversification or process miniaturization are either undetected or misclassified under existing definitions. Furthermore, training the model to recognize new patterns incurs significant inefficiency due to the substantial costs of data labeling and model retraining time. To address this, this study developed an integrated defect detection framework based on active learning that maintains high classification performance for known defect patterns while effectively identifying unknown defect patterns and continuously learning. The developed system consists primarily of two stages: unknown defect detection and classification/learning. First, an anomaly detector based on One-Class Support Vector Machine (SVM) preliminarily determines whether the input WBM is a previously learned known defect pattern or a new type of unknown pattern. If identified as an existing pattern, the classification model precisely classifies the specific defect type. (See Figure 1) Conversely, data classified as unknown patterns are clustered into groups with similar characteristics using the DBSCAN algorithm. This clustered data enables efficient labeling with minimal intervention from process engineers. Through active learning techniques, the classifier updates new pattern information in real-time. This process allows the model to adapt to the constantly changing process environment, maintaining and improving its performance autonomously. (See Figure 2) Experimental results using the WM-811K dataset demonstrated that the developed model maintained high classification accuracy for known defects while effectively filtering out unknown patterns. Furthermore, ‘Eye Defect Patterns,’ which is not present in WM-811K but present in the actual mass production line in Samsung Electronics, was successfully detected and learned these unknown patterns, proving the model's applicability and utility in real industrial settings. This research is significant in that it presents a methodology demonstrating how artificial intelligence models can practically contribute to building intelligent defect management systems for semiconductor processes. The research findings were published in Expert Systems with Applications, which is famous in industrial engineering discipline. Figure 1. Multi-step detection process for unknown defect patterns Figure 2. Process of updating the classifier to an unknown pattern ※ Title: A framework for detecting unknown defect patterns on wafer bin maps using active learning ※ Journal: Expert Systems with Applications ※ DOI: https://www.sciencedirect.com/science/article/pii/S0957417424022450 ※ Pure: https://pure.skku.edu/en/persons/donghee-lee/
- No. 346
- 2025-12-12
- 2067
-
The Vivid Realism of AI Deepfakes Undermines Our Judgement
This study is an international collaboration between Professor Jiyoung Lee of the Department of Media and Communication at Sungkyunkwan University and Professor Michael Hameleers of the University of Amsterdam. It provides empirical evidence on how AI-generated health deepfakes amplify misbeliefs and how individual characteristics shape these effects. The research was published in Media Psychology, a leading international journal in the media and psychology fields. The researchers examined three central questions: Do AI-generated health deepfakes exert stronger effects than text-based misinformation? Do individuals exposed to deepfakes show greater or reduced intentions to verify information accuracy? How do personal factors—such as interest in health issues and accuracy motivation—shape these influence processes? A key contribution of the study lies in showing that deepfakes can mimic the authority of medical experts by reproducing visual and linguistic cues that resemble professional communication. In doing so, deepfakes function not merely as vivid, persuasive videos but as technological mechanisms that simulate authority and make their messages appear more credible and expert-like. The findings show that deepfakes produced the largest increase in health-related misperceptions among all misinformation types tested. Participants exposed to deepfakes reported substantially higher levels of misbelief than those exposed to text-based misinformation. This effect appears to stem from the combination of deepfakes’ vivid realism and their ability to imitate authoritative styles—through expert-like tone, wording, and visual presentation—making false health claims particularly convincing. Although exposure to deepfakes did not directly alter accuracy-checking intentions, the pattern varied across individuals. For text-based misinformation, those with greater interest in health issues showed stronger intentions to verify accuracy. However, this pattern disappeared in the deepfake condition. Even participants highly interested in health topics did not show higher intentions to fact-check, suggesting that the realistic and authority-mimicking nature of deepfakes may inhibit critical evaluation even among highly involved individuals. Another notable finding concerns accuracy motivation. Conventional wisdom suggests that individuals who prioritize accuracy should be better at recognizing misinformation. Yet the study found the opposite: those with stronger accuracy motivation exhibited greater increases in misperceptions when exposed to deepfakes. This paradox may reflect a psychological tendency toward illusory accuracy perception—the belief that one’s judgment is correct—particularly when the misinformation is delivered through a video that appears authoritative. The deepfake’s expert-like cues may have reinforced this misplaced sense of confidence. Overall, the study provides empirical evidence of how AI-driven, multimodal misinformation affects cognition, highlighting previously understudied risks in the domain of health communication. By demonstrating how deepfakes technologically reproduce authority cues to enhance perceived credibility and persuasion, the research offers important theoretical insights into the mechanisms through which misinformation exerts its influence. It also clarifies how factors such as health-issue interest and accuracy motivation operate—sometimes counterintuitively—when individuals process deepfake content. The findings carry significant societal implications. Deepfakes can function not only as distorted health information but as serious public health risks. Moreover, individuals who normally value accuracy and engage in careful information processing may still be vulnerable to deepfakes that convincingly imitate expert authority. These insights underscore the need for strengthened AI literacy education, improved digital risk communication strategies, and more robust institutional response systems to mitigate the threats posed by synthetic media in health contexts. Article Information Title: Effects of Health-related Deepfakes on Misperceptions: Moderating Effects of Issue Relevance and Accuracy Motivation Authors: Jiyoung Lee & Michael Hameleers Journal: Media Psychology DOI: https://doi.org/10.1080/15213269.2024.2401539 Pure: https://pure.skku.edu/en/persons/jiyoung-lee/ (Deepfake stimuli)
- No. 345
- 2025-12-08
- 2774
-
“Apply and Deliver” A Pain-Free Patch that Enables Deep Dermal Delivery of Extracellular Vesicles
A joint research team led by Professors Jae Hyung Park and Changhyun Pang from the School of Chemical Engineering (lead authors: Minwoo Song, Minji Ha, and Dr. Sol Shin) has developed a core technology that enables a next-generation transdermal drug delivery system. By hierarchically integrating octopus-sucker–inspired suction cups with short microneedles, the team successfully created a highly adhesive dual-amplified transdermal patch capable of delivering large biomolecular therapeutics such as extracellular vesicles (EVs) deep into the skin uniformly, efficiently, and without pain. This innovation overcomes long-standing limitations of injection-based therapy and opens up new possibilities for applications ranging from aesthetic and anti-aging treatments to future smart healthcare platforms. The research addressed fundamental drawbacks of conventional microneedle patches, particularly pain, skin irritation, and poor adhesion. The stratum corneum, the outermost layer of the skin, possesses a dense and protective structure that blocks the penetration of external substances, making it difficult for biological agents such as exosomes and proteins to reach the dermis via topical formulations like creams or ointments. Previous microneedle systems attempted to bypass this barrier by utilizing long needles over 600 μm, but such lengths often induced discomfort, irritation, and detachment on curved or moist skin surfaces. To overcome these challenges, the research team designed a bioinspired suction-cup microstructure that generates stable negative pressure upon simple attachment, without requiring external devices or power. By combining this suction mechanism with short microneedles under 300 μm, the researchers constructed a novel dual-amplification architecture. Once applied to the skin, the patch naturally forms microscale negative pressure within each suction chamber, enhancing conformal adhesion. Simultaneously, this negative pressure induces nanoscale opening and temporary deformation of the stratum corneum, thereby amplifying microneedle insertion efficiency and enabling the effective delivery of large biomolecules, including exosomes, into the dermal layer. Animal studies demonstrated the superior therapeutic performance and safety of the newly developed exosome transdermal patch. Compared to conventional microneedle patches, the system increased drug delivery depth by approximately 2.6-fold (up to 290 μm). It also stimulated collagen production, reduced reactive oxygen species (ROS), and improved the microenvironment of aging skin. Importantly, the patch maintained strong, stable adhesion even on curved or moist skin surfaces, ensuring sustained drug delivery during long-term application. The dual-amplified patch exhibits excellent biocompatibility and irritation-free delivery, making it suitable for stable daily use. Beyond exosomes, the platform can effectively deliver a wide range of biological therapeutics, including proteins and nucleic acids. Therefore, its applicability is expected to expand into diverse fields such as cosmetic dermatology, anti-aging therapy, regenerative applications, and wearable smart healthcare systems. This work was supported by National Research Foundation of Korea (NRF) grants funded by the Korean government (MSIT) (No. RS-2023-00256265, RS-2024-00352352, RS-2024-00405818) and by the Korean Fund for Regenerative Medicine (KFRM) grant funded by the Korea government (the Ministry of Science and ICT, the Ministry of Health & Welfare). (No. 25A0102L1). The authors gratefully acknowledge the support from the Market-led K-sensor technology program (RS-2022-00154781, Development of large-area wafer-level flexible/stretch- able hybrid sensor platform technology for form factor-free highly integrated convergence sensor), funded By the Ministry of Trade, Industry & Energy (MOTIE, Korea). The study was published online on July 23, 2025, in the international journal Nano-Micro Letters (IF 36.3, JCR top 1%) and was selected as the Cover Paper. *Paper Title: A Hierarchical Short Microneedle-Cupping Dual-Amplified Patch Enables Accelerated, Uniform, Pain-Free Transdermal Delivery of Extracellular Vesicles *Journal: Nano-Micro Letters *Paper Link: https://doi.org/10.1007/s40820-025-01853-7 *Pure: https://pure.skku.edu/en/persons/jae-hyung-park/ (Prof. Jae Hyung Park) https://pure.skku.edu/en/persons/changhyun-pang/(Prof. Changhyun Pang)
- No. 344
- 2025-12-03
- 2229
-
Turning Food Waste into Future Resources
Prof. Am Jang’s team (Sustainable WAter Treatment, SWAT lab.) in the Department of Civil and Environmental Engineering (first author: Dr. Hongrae Im) has successfully developed a next-generation membrane-based recovery platform capable of converting fatty acids contained in food waste into high-value precursors for energy and chemical materials. Conventional membrane processes have long suffered from structural limitations, such as membrane fouling and wetting, which lead to reduced selectivity and decreased recovery efficiency. To overcome these challenges, the research team introduced a supported liquid membrane contactor (SLMC) system that integrates superhydrophobic surface modification with selective organic extractant impregnation. This approach enabled highly selective and efficient recovery of target fatty acids even from the complex matrix of food-derived waste streams. A key achievement of this study is the pioneering application of optical coherence tomography (OCT) for real-time monitoring of fouling and wetting transitions occurring on membrane surfaces and within membrane pores. By visualizing and quantifying these microscale structural changes, the team successfully identified the primary causes of performance deterioration and established an operational framework capable of actively controlling process conditions. Using OCT-based wetting and fouling signals, the researchers proposed an “intelligent membrane operation strategy” that accurately determines optimal cleaning and replacement intervals, thereby significantly enhancing long-term operational stability and fatty-acid recovery efficiency. Furthermore, the researchers systematically elucidated the material transport mechanisms within the SLMC system by investigating the synergistic effects between superhydrophobic membrane surface modification and the chemical properties of selective extractants. Through this approach, they developed an integrated solution capable of simultaneously addressing three major technical bottlenecks: membrane fouling and wetting suppression, enhancement of selective mass transfer, and low-energy/high-efficiency fatty acid recovery. Owing to its robust performance, the developed technology shows strong potential for expansion into a universal circular-resource platform applicable not only to food waste but also to desalination brines, spent battery leachates, and other complex multicomponent aqueous systems. This research provides significant academic and industrial value by presenting a sustainable resource, upcycling strategy that transforms low-value waste streams into high-value chemical materials, thereby transcending the traditional paradigm of waste treatment. The work was supported by a National Research Foundation of Korea (NRF) grant funded by the Korean Government (MSIT), and its excellence and novelty were recognized through publication in top-tier international journals in the environmental and energy fields, including Water Research (IF 12.4, JCR top 1.0%) and Chemical Engineering Journal (IF 13.2, JCR top 3.0%). Title: Efficiently enhanced short-chain fatty acids (SCFAs) recovery from food waste condensate: Real-time wettability monitoring with supported liquid membrane contactor Authors: Hongrae Im (First author, Sungkyunkwan University), Duc Anh Nguyen (Sungkyunkwan University), Dong-gun Jun (Sungkyunkwan University), Sojeong Jang (Sungkyunkwan University), Am Jang (Corresponding author, Sungkyunkwan University) Journal: Water Research DOI: https://doi.org/10.1016/j.watres.2025.123093 Title: Enhanced anti-wetting and anti-fouling performance of superhydrophobic supported liquid membrane contactors for selective short-chain fatty acid recovery from food waste leachate Authors: Hongrae Im (First author, Sungkyunkwan University), Semi Lee (Sungkyunkwan University), Duc Anh Nguyen (Sungkyunkwan University), Duc Viet Nguyen (Ghent University), Di Wu (Ghent University), Am Jang (Corresponding author, Sungkyunkwan University) Journal: Chemical Engineering Journal DOI: https://doi.org/10.1016/j.cej.2025.168433 Figure 1. Schematic of short-chain fatty acid recovery from food waste Figure 2. Mechanism, recovery efficiency, and mass transfer behavior of fatty acid from food waste streams Figure 3. Physicochemical properties of modified membranes Figure 4. Real-time monitoring of the physical properties of the membrane surface
- No. 343
- 2025-11-28
- 5283
-
Sungkyunkwan University research team develops a novel Water-Energy Nexus technology using metal-organic framework mate
Sungkyunkwan University (President Ji Beom Yoo) announced that a research team led by Professor Ho Seok Park in the School of Chemical Engineering, in collaboration with Samsung Research at Samsung Electronics, has developed a novel Water-Energy Nexus* technology that can efficiently treat saline water and wastewater generated from high-value-added industrial processes—such as semiconductor, secondary battery, and display manufacturing—while enabling their use for energy storage and resource recovery. *Water-Energy Nexus: Water is essential for energy production, and energy is required for water management. This concept focuses on optimizing the interdependence between these two resources to address water scarcity and climate change and to enable sustainable resource management. Conventional wastewater treatment technologies typically rely on physical adsorption and desorption of ions on electrode surfaces, resulting in high energy consumption and low ion selectivity. In this study, the researchers utilized the distinctive ion-storage mechanism of metal-organic frameworks (MOFs)*—which were awarded this year’s Nobel Prize in Chemistry—to develop an energy-efficient electrochemical device that can selectively remove or store cations and anions simultaneously without an ion-exchange membrane. *Metal-organic frameworks (MOFs): Porous crystalline materials composed of metal ions or clusters connected by organic linkers. MOFs offer a large internal surface area and high structural stability, enabling their use in diverse applications such as gas storage, catalysis, and sensing. MOFs are three-dimensional porous materials constructed from metal ions and organic ligands, and they are used in a wide range of fields, including gas storage, catalysis, and sensing. By taking advantage of these structural characteristics, Professor Park’s team developed an electrode technology that can selectively and simultaneously remove and store divalent cations such as Ca²⁺ and Mg²⁺, as well as anions such as Br⁻, I⁻, and Cl⁻, while operating stably with low energy consumption of about 76 Wh kg⁻¹. Professor Park said, “This study goes beyond simply purifying saline water and wastewater, as it simultaneously demonstrates their potential to be converted into renewable energy and valuable resources,” adding, “We expect this technology to find broader applications in various fields, including secondary batteries, desalination, and resource recovery.” This research was supported by the Global Leader Research and the Future Promising Fusion Technology Pioneer programs of the National Research Foundation of Korea (NRF), funded by the Korean Government (MSIT), and was carried out in collaboration with Samsung Research at Samsung Electronics. The results were published on October 28, 2025 in Joule (impact factor 35.4; top 1.4%), a leading international journal in energy research. ※ Paper Title: Divalent and Halide Dual Ion Storage of A Redox-Active Symmetric Cell for Efficient Wastewater-Energy Nexus ※ Journal: Joule ※ Authors: Corresponding author – Professor Ho Seok Park (Sungkyunkwan University); First authors – Dr. Gun Jang and PhD candidate Sang Baek Kim (Sungkyunkwan University) ※ Paper Link: https://www.cell.com/joule/fulltext/S2542-4351(25)00357-5?rss=yes ※ (Pure): https://pure.skku.edu/en/persons/hoseok-park/ ▲Conceptual illustration of a novel Water-Energy Nexus technology based on an energy-efficient, highly selective ion-recovery and storage device using MOF-derived electrode materials ▲(From left) Professor Ho Seok Park, Dr. Gun Jang, and PhD candidate Sang Baek Kim (Sungkyunkwan University)
- No. 342
- 2025-11-25
- 1611
-
Discovery of Novel Lung Cancer Biomarkers and Therapeutic Targets to Overcome Drug Resistance
Lung cancer is the leading cause of cancer-related mortality worldwide. In particular, non–small cell lung cancer (NSCLC) is often diagnosed at an advanced stage due to the lack of early symptoms. Although targeted therapies such as EGFR-TKIs have significantly improved clinical outcomes, the rapid emergence of therapeutic resistance remains a major barrier to long-term survival. Consequently, identifying novel biomarkers that determine tumor progression and treatment responsiveness, as well as discovering therapeutic targets capable of overcoming resistance, has become a crucial strategy for transforming the lung cancer treatment paradigm. Cancer precision medicine is an approach that integrates genomic, transcriptomic, and proteomic information to design personalized therapeutic strategies tailored to each patient’s molecular profile. This research, grounded in precision-medicine–based analyses, identifies previously unrecognized signaling networks driving NSCLC progression and highlights actionable targets with strong potential for future therapeutic development. 1. PYCR1–EGFR–TLR Signaling Axis: A Newly Identified Mechanism Driving Lung Cancer Progression In the study “PYCR1 drives lung cancer progression through functional interactions with EGFR and TLR signaling pathways” (Experimental & Molecular Medicine, 2025, IF 12.9), the researchers uncovered a novel molecular mechanism in which PYCR1, a key enzyme in proline metabolism, functionally interacts with EGFR and TLR signaling to promote lung cancer growth and metastasis. This research is the first to demonstrate that PYCR1—traditionally viewed only as a metabolic enzyme—acts as a central regulatory node within the lung cancer signaling network. The findings highlight PYCR1 as a promising strategic target for developing lung cancer–specific therapeutics. (See Fig. 1.). 2. USP21–EGFR–Lyn Signaling Axis: A Therapeutic Target for Overcoming Resistance The second study, “USP21–EGFR–Lyn axis drives NSCLC progression and therapeutic potential of USP21 inhibition” (Biomarker Research, 2025, IF 11.5), elucidates a mechanism in which the deubiquitinase USP21 simultaneously activates EGFR and Lyn kinase signaling to drive NSCLC progression. This work establishes USP21 as a key regulator of lung cancer progression and demonstrates its value as a biomarker and therapeutic target capable of overcoming resistance to EGFR-targeted therapy. The clinical significance of USP21 inhibition is particularly notable in the context of combination treatment strategies designed to counteract EGFR inhibitor resistance. (See Fig. 2.) These studies were led by Ha-Jeong Lee, an Integrated B.S.–M.S. program applicant, in collaboration with graduate researchers Ji-Young Kim, Ji-Hye Shin, and Ye-Eun Kang from the Laboratory of Molecular Immunology (PI: Professor Kiyoung Lee, School of Medicine). Ha-Jeong’s achievements exemplify the fact that “undergraduate researchers can also produce world-class scientific outcomes.” Her work represents an outstanding model of SKKU’s research-oriented education and demonstrates the impact of a supportive research environment combined with rigorous scientific training. ※ Article title: PYCR1 drives lung cancer progression through functional interactions with EGFR and TLR signaling pathways. ※ Journal name: Experimental & Molecular Medicine. ※ Article link: https://www.nature.com/articles/s12276-025-01577-z ※ Article title: USP21-EGFR-Lyn axis drives NSCLC progression and therapeutic potential of USP21 inhibition. ※ Journal name: Biomarker research. ※ Article link: https://pmc.ncbi.nlm.nih.gov/articles/PMC12239452/
- No. 341
- 2025-11-21
- 1660
-
The Backfire Effect of Upside-Down Logos on Consumer Responses to Brands
Brand logos refer to a set of symbolic elements that provide distinctiveness and communicate a company’s identity and values, helping consumers recognize and trust the brand. Recently, several global companies have begun experimenting with an unconventional twist—flipping their logos upside down—to signal creativity and differentiation. Adidas drew attention with an inverted-logo advertising campaign, and other brands such as Nike, New Era, and Supreme have applied similar designs to limited-edition products to create novel brand experiences. However, it still remains unclear whether inverted logos are perceived as bold and innovative or are seen as undermining brand consistency and creating consumer confusion. Professor Tae Hyun Baek from SKKU’s Department of Media and Communication, in collaboration with research teams from the University of Massachusetts (USA), Peking University (China), and the London School of Economics and Political Science (UK), examined how logo orientation influences consumer responses to brands. Across four experimental studies, consumers consistently preferred products featuring standard, upright logos over inverted, upside-down versions. In a consequential choice experiment using a Comme des Garçons T-shirt (Study 1A), 74.7% of participants selected the standard logo. A follow-up study using a baseball cap (Study 1B) revealed an even stronger pattern, with 80.8% choosing the standard logo. Notably, these preferences were not explained by demographic characteristics or by consumers’ need for uniqueness, suggesting that the effect is both robust and broadly generalizable. Study 2 extended the findings of Studies 1A and 1B to the context of social media advertising by identifying the psychological mechanisms underlying consumer responses to inverted logos. The results showed that the effect of an inverted logo on purchase intention was sequentially mediated by perceived unexpectedness and perceived rebelliousness. Consumers viewed inverted logos as unfamiliar and unconventional design choices that deliberately violated expected visual norms, which in turn led to more negative reactions. Study 3 revealed that political ideology moderated the negative effect of inverted logos. Whereas liberal consumers were largely indifferent to logo orientation, conservative consumers exhibited significantly more negative attitudes toward brands featuring inverted logos. It is suggested that inverted logos conflicted with conservatives’ preference for orderly, conventional design elements. Thus, political ideology shaped how consumers interpreted visual disruptions in branding, indicating that unconventional design strategies such as inverted logos could be ineffective—or even counterproductive—for certain segments. Taken together, these findings offer practical insights into the growing trend of global brands leveraging inverted logos. Such unconventional logo designs should be deployed selectively, particularly when they align with the political values of the target audience and with a brand’s rebellious, edgy, or innovative personality. The research was recently published in the Journal of Retailing and Consumer Services, a top-tier SSCI Q1 journal ranked within the top 1.7% in the Business category (2024 Impact Factor: 13.1). Baek, T. H., Yim, M. Y-C., Park, J., & Cho, A. (2026). Disruptive but costly: How upside-down logos backfire in consumer responses to brands. Journal of Retailing and Consumer Services, 88, 104500. https://doi.org/10.1016/j.jretconser.2025.104500 ※ Title: Disruptive but costly: How upside-down logos backfire in consumer responses to brands. ※ Journal: Journal of Retailing and Consumer Services ※ Link: https://doi.org/10.1016/j.jretconser.2025.104500 ※ Portal(Pure): https://pure.skku.edu/en/persons/tae-hyun-baek/
- No. 340
- 2025-11-18
- 1857
-
Innovating User Experience in the Era of Generative AI
This study is the result of a collaborative effort by Professor Jang Hyun Kim of the School of Global Convergence and members of the Data Science & Social Analytics Lab (DSSAL), including Dongyan Nan (Ph.D. graduate, now Assistant Professor at Macau University of Science and Technology), Seungjong Sun (Ph.D. candidate), Shunan Zhang (Ph.D. graduate, now Fellow at Huaqiao University in China), and Xiangying Zhao (Ph.D. graduate). The purpose of this research is to gain an in-depth understanding of user behavior toward Generative Artificial Intelligence. Focusing on ChatGPT as a representative example, the study identifies key factors that influence users’ continued usage intention and recommendation intention. Theoretically, it proposes a new integrated model that extends the Expectation Confirmation Model (ECM) by incorporating Information System Success Theory (ISST), privacy concerns, and perceived innovativeness. This approach addresses the limitations of prior studies, which largely focused on initial usage intention, and instead highlights both cognitive and emotional determinants of post-adoption behavior—providing meaningful academic contributions. A total of 252 Korean ChatGPT users participated in an online survey, and the results were analyzed using structural equation modeling. The findings show that the proposed integrated model effectively explains users’ continued use and recommendation behaviors. Information Quality and System Quality emerged as core variables that enhance both types of behavioral intentions by strengthening confirmation, perceived usefulness, and satisfaction. Perceived innovativeness also had a positive effect on user satisfaction, demonstrating that users form more favorable experiences when they view ChatGPT as a creative and cutting-edge technology. Conversely, privacy concerns negatively affected satisfaction, although the impact was relatively small—suggesting that users may be willing to accept certain privacy risks in exchange for convenience and utility. Based on these findings, the study offers practical implications for promoting the adoption of generative AI services. Service providers can enhance user engagement by improving model accuracy and stability to reduce information bias, designing user-friendly interfaces, and effectively promoting the creativity and innovativeness of AI technologies. Professor Kim noted, “By comprehensively analyzing the determinants of continued use and recommendation of generative AI, this study offers new insights into user experience–based AI adoption research. We plan to further advance the model by expanding our research to include voice- and image-based generative AI in the future.” His research team continues to explore the intersection of AI and user experience (UX) and has published numerous SSCI/SCIE-indexed papers in related fields. ※ Title: Analyzing behavioral intentions toward Generative Artificial Intelligence: the case of ChatGPT ※ Journal: Universal Access in the Information Society ※ Link: https://doi.org/10.1007/s10209-024-01116-z ※ Portal(Pure): https://pure.skku.edu/en/persons/janghyun-kim/ Dongyan Nan(Ph.D. graduate, now Assistant Professor at Macau University of Science and Technology), Shunan Zhang(Ph.D. graduate, now Fellow at Huaqiao University in China), Xiangying Zhao(Ph.D. graduate)
- No. 339
- 2025-11-13
- 1951
