CHI
-
KAIST & CMU Unveils Amuse, a Songwriting AI-Collaborator to Help Create Music
Wouldn't it be great if music creators had someone to brainstorm with, help them when they're stuck, and explore different musical directions together? Researchers of KAIST and Carnegie Mellon University (CMU) have developed AI technology similar to a fellow songwriter who helps create music.
KAIST (President Kwang-Hyung Lee) has developed an AI-based music creation support system, Amuse, by a research team led by Professor Sung-Ju Lee of the School of Electrical Engineering in collaboration with CMU. The research was presented at the ACM Conference on Human Factors in Computing Systems (CHI), one of the world’s top conferences in human-computer interaction, held in Yokohama, Japan from April 26 to May 1. It received the Best Paper Award, given to only the top 1% of all submissions.
< (From left) Professor Chris Donahue of Carnegie Mellon University, Ph.D. Student Yewon Kim and Professor Sung-Ju Lee of the School of Electrical Engineering >
The system developed by Professor Sung-Ju Lee’s research team, Amuse, is an AI-based system that converts various forms of inspiration such as text, images, and audio into harmonic structures (chord progressions) to support composition.
For example, if a user inputs a phrase, image, or sound clip such as “memories of a warm summer beach”, Amuse automatically generates and suggests chord progressions that match the inspiration.
Unlike existing generative AI, Amuse is differentiated in that it respects the user's creative flow and naturally induces creative exploration through an interactive method that allows flexible integration and modification of AI suggestions.
The core technology of the Amuse system is a generation method that blends two approaches: a large language model creates music code based on the user's prompt and inspiration, while another AI model, trained on real music data, filters out awkward or unnatural results using rejection sampling.
< Figure 1. Amuse system configuration. After extracting music keywords from user input, a large language model-based code progression is generated and refined through rejection sampling (left). Code extraction from audio input is also possible (right). The bottom is an example visualizing the chord structure of the generated code. >
The research team conducted a user study targeting actual musicians and evaluated that Amuse has high potential as a creative companion, or a Co-Creative AI, a concept in which people and AI collaborate, rather than having a generative AI simply put together a song.
The paper, in which a Ph.D. student Yewon Kim and Professor Sung-Ju Lee of KAIST School of Electrical and Electronic Engineering and Carnegie Mellon University Professor Chris Donahue participated, demonstrated the potential of creative AI system design in both academia and industry. ※ Paper title: Amuse: Human-AI Collaborative Songwriting with Multimodal Inspirations DOI: https://doi.org/10.1145/3706598.3713818
※ Research demo video: https://youtu.be/udilkRSnftI?si=FNXccC9EjxHOCrm1
※ Research homepage: https://nmsl.kaist.ac.kr/projects/amuse/
Professor Sung-Ju Lee said, “Recent generative AI technology has raised concerns in that it directly imitates copyrighted content, thereby violating the copyright of the creator, or generating results one-way regardless of the creator’s intention. Accordingly, the research team was aware of this trend, paid attention to what the creator actually needs, and focused on designing an AI system centered on the creator.”
He continued, “Amuse is an attempt to explore the possibility of collaboration with AI while maintaining the initiative of the creator, and is expected to be a starting point for suggesting a more creator-friendly direction in the development of music creation tools and generative AI systems in the future.”
This research was conducted with the support of the National Research Foundation of Korea with funding from the government (Ministry of Science and ICT). (RS-2024-00337007)
2025.05.07 View 613 -
KAIST provides a comprehensive resource on microbial cell factories for sustainable chemical production
In silico analysis of five industrial microorganisms identifies optimal strains and metabolic engineering strategies for producing 235 valuable chemicals
Climate change and the depletion of fossil fuels have raised the global need for sustainable chemical production. In response to these environmental challenges, microbial cell factories are gaining attention as eco-friendly platforms for producing chemicals using renewable resources, while metabolic engineering technologies to enhance these cell factories are becoming crucial tools for maximizing production efficiency. However, difficulties in selecting suitable microbial strains and optimizing complex metabolic pathways continue to pose significant obstacles to practical industrial applications.
KAIST (President Kwang-Hyung Lee) announced on 27th of March that Distinguished Professor Sang Yup Lee’s research team in the Department of Chemical and Biomolecular Engineering comprehensively evaluated the production capabilities of various industrial microbial cell factories using in silico simulations and, based on these findings, identified the most suitable microbial strains for producing specific chemicals as well as optimal metabolic engineering strategies.
Previously, researchers attempted to determine the best strains and efficient metabolic engineering strategies among numerous microbial candidates through extensive biological experiments and meticulous verification processes. However, this approach required substantial time and costs. Recently, the introduction of genome-scale metabolic models (GEMs), which reconstruct the metabolic networks within an organism based on its entire genome information, has enabled systematic analysis of metabolic fluxes via computer simulations. This development offers a new way to overcome limitations of conventional experimental approaches, revolutionizing both strain selection and metabolic pathway design.
Accordingly, Professor Lee’s team at the Department of Chemical and Biomolecular Engineering, KAIST, evaluated the production capabilities of five representative industrial microorganisms—Escherichia coli, Saccharomyces cerevisiae, Bacillus subtilis, Corynebacterium glutamicum, and Pseudomonas putida—for 235 bio-based chemicals. Using GEMs, the researchers calculated both the maximum theoretical yields and the maximum achievable yields under industrial conditions for each chemical, thereby establishing criteria to identify the most suitable strains for each target compound.
< Figure 1. Outline of the strategy for improving microbial cell factories using a genome-scale metabolic model (GEM) >
The team specifically proposed strategies such as introducing heterologous enzyme reactions derived from other organisms and exchanging cofactors used by microbes to expand metabolic pathways. These strategies were shown to increase yields beyond the innate metabolic capacities of the microorganisms, resulting in higher production of industrially important chemicals such as mevalonic acid, propanol, fatty acids, and isoprenoids.
Moreover, by applying a computational approach to analyze metabolic fluxes in silico, the researchers suggested strategies for improving microbial strains to maximize the production of various chemicals. They quantitatively identified the relationships between specific enzyme reactions and target chemical production, as well as the relationships between enzymes and metabolites, determining which enzyme reactions should be up- or down-regulated. Through this, the team presented strategies not only to achieve high theoretical yields but also to maximize actual production capacities.
< Figure 2. Comparison of production routes and maximum yields of useful chemicals using representative industrial microorganisms >
Dr. Gi Bae Kim, the first author of this paper from the KAIST BioProcess Engineering Research Center, explained, “By introducing metabolic pathways derived from other organisms and exchanging cofactors, it is possible to design new microbial cell factories that surpass existing limitations. The strategies presented in this study will play a pivotal role in making microbial-based production processes more economical and efficient.” In addition, Distinguished Professor Sang Yup Lee noted, “This research serves as a key resource in the field of systems metabolic engineering, reducing difficulties in strain selection and pathway design, and enabling more efficient development of microbial cell factories. We expect it to greatly contribute to the future development of technologies for producing various eco-friendly chemicals, such as biofuels, bioplastics, and functional food materials.”
This research was conducted with the support from the Development of platform technologies of microbial cell factories for the next-generation biorefineries project and Development of advanced synthetic biology source technologies for leading the biomanufacturing industry project (Project Leader: Distinguished Professor Sang Yup Lee, KAIST) from National Research Foundation supported by the Korean Ministry of Science and ICT.
2025.03.27 View 1536 -
KAIST Develops Eco-Friendly, Nylon-Like Plastic Using Microorganisms
Poly(ester amide) amide is a next-generation material that combines the advantages of PET (polyester) and nylon (polyamide), two widely used plastics. However, it could only be produced from fossil fuels, which posed environmental concerns. Using microorganisms, KAIST researchers have successfully developed a new bio-based plastic to replace conventional plastic.
KAIST (represented by President Kwang Hyung Lee) announced on the 20th of March that a research team led by Distinguished Professor Sang Yup Lee from the Department of Chemical and Biomolecular Engineering has developed microbial strains through systems metabolic engineering to produce various eco-friendly, bio-based poly(ester amide)s. The team collaborated with researchers from the Korea Research Institute of Chemical Technology (KRICT, President Young-Kook Lee) to analyze and confirm the properties of the resulting plastic.
Professor Sang Yup Lee’s research team designed new metabolic pathways that do not naturally exist in microorganisms, and developed a platform microbial strain capable of producing nine different types of poly(ester amide)s, including poly(3-hydroxybutyrate-ran-3-aminopropionate) and poly(3-hydroxybutyrate-ran-4-aminobutyrate).
Using glucose derived from abundant biomass sources such as waste wood and weeds, the team successfully produced poly(ester amide)s in an eco-friendly manner. The researchers also confirmed the potential for industrial-scale production by demonstrating high production efficiency (54.57 g/L) using fed-batch fermentation of the engineered strain.
In collaboration with researchers Haemin Jeong and Jihoon Shin from KRICT, the KAIST team analyzed the properties of the bio-based plastic and found that it exhibited characteristics similar to high-density polyethylene (HDPE). This means the new plastic is not only eco-friendly but also strong and durable enough to replace conventional plastics.
The engineered strains and strategies developed in this study are expected to be useful not only for producing various poly(ester amide)s but also for constructing metabolic pathways for the biosynthesis of other types of polymers.
Professor Sang Yup Lee stated, “This study is the first to demonstrate the possibility of producing poly(ester amide)s (plastics) through a renewable bio-based chemical process rather than relying on the petroleum-based chemical industry. We plan to further enhance the production yield and efficiency through continued research.”
The study was published online on March 17 in the international journal Nature Chemical Biology.
·Title: Biosynthesis of poly(ester amide)s in engineered Escherichia coli
·DOI: 10.1038/s41589-025-01842-2
·Authors: A total of seven authors including Tong Un Chae (KAIST, first author), So Young Choi (KAIST, second author), Da-Hee Ahn (KAIST, third author), Woo Dae Jang (KAIST, fourth author), Haemin Jeong (KRICT, fifth author), Jihoon Shin (KRICT, sixth author), and Sang Yup Lee (KAIST, corresponding author).
This research was supported by the Ministry of Science and ICT (MSIT) under the Eco-Friendly Chemical Technology Development Project as part of the "Next-Generation Biorefinery Technology Development to Lead the Bio-Chemical Industry" initiative (project led by Distinguished Professor Sang Yup Lee at KAIST).
2025.03.24 View 2405 -
A Way for Smartwatches to Detect Depression Risks Devised by KAIST and U of Michigan Researchers
- A international joint research team of KAIST and the University of Michigan developed a digital biomarker for predicting symptoms of depression based on data collected by smartwatches
- It has the potential to be used as a medical technology to replace the economically burdensome fMRI measurement test
- It is expected to expand the scope of digital health data analysis
The CORONA virus pandemic also brought about a pandemic of mental illness. Approximately one billion people worldwide suffer from various psychiatric conditions. Korea is one of more serious cases, with approximately 1.8 million patients exhibiting depression and anxiety disorders, and the total number of patients with clinical mental diseases has increased by 37% in five years to approximately 4.65 million. A joint research team from Korea and the US has developed a technology that uses biometric data collected through wearable devices to predict tomorrow's mood and, further, to predict the possibility of developing symptoms of depression.
< Figure 1. Schematic diagram of the research results. Based on the biometric data collected by a smartwatch, a mathematical algorithm that solves the inverse problem to estimate the brain's circadian phase and sleep stages has been developed. This algorithm can estimate the degrees of circadian disruption, and these estimates can be used as the digital biomarkers to predict depression risks. >
KAIST (President Kwang Hyung Lee) announced on the 15th of January that the research team under Professor Dae Wook Kim from the Department of Brain and Cognitive Sciences and the team under Professor Daniel B. Forger from the Department of Mathematics at the University of Michigan in the United States have developed a technology to predict symptoms of depression such as sleep disorders, depression, loss of appetite, overeating, and decreased concentration in shift workers from the activity and heart rate data collected from smartwatches.
According to WHO, a promising new treatment direction for mental illness focuses on the sleep and circadian timekeeping system located in the hypothalamus of the brain, which directly affect impulsivity, emotional responses, decision-making, and overall mood.
However, in order to measure endogenous circadian rhythms and sleep states, blood or saliva must be drawn every 30 minutes throughout the night to measure changes in the concentration of the melatonin hormone in our bodies and polysomnography (PSG) must be performed. As such treatments requires hospitalization and most psychiatric patients only visit for outpatient treatment, there has been no significant progress in developing treatment methods that take these two factors into account. In addition, the cost of the PSG test, which is approximately $1000, leaves mental health treatment considering sleep and circadian rhythms out of reach for the socially disadvantaged.
The solution to overcome these problems is to employ wearable devices for the easier collection of biometric data such as heart rate, body temperature, and activity level in real time without spatial constraints. However, current wearable devices have the limitation of providing only indirect information on biomarkers required by medical staff, such as the phase of the circadian clock.
The joint research team developed a filtering technology that accurately estimates the phase of the circadian clock, which changes daily, such as heart rate and activity time series data collected from a smartwatch. This is an implementation of a digital twin that precisely describes the circadian rhythm in the brain, and it can be used to estimate circadian rhythm disruption.
< Figure 2. The suprachiasmatic nucleus located in the hypothalamus of the brain is the central biological clock that regulates the 24-hour physiological rhythm and plays a key role in maintaining the body’s circadian rhythm. If the phase of this biological clock is disrupted, it affects various parts of the brain, which can cause psychiatric conditions such as depression. >
The possibility of using the digital twin of this circadian clock to predict the symptoms of depression was verified through collaboration with the research team of Professor Srijan Sen of the Michigan Neuroscience Institute and Professor Amy Bohnert of the Department of Psychiatry of the University of Michigan.
The collaborative research team conducted a large-scale prospective cohort study involving approximately 800 shift workers and showed that the circadian rhythm disruption digital biomarker estimated through the technology can predict tomorrow's mood as well as six symptoms, including sleep problems, appetite changes, decreased concentration, and suicidal thoughts, which are representative symptoms of depression.
< Figure 3. The circadian rhythm of hormones such as melatonin regulates various physiological functions and behaviors such as heart rate and activity level. These physiological and behavioral signals can be measured in daily life through wearable devices. In order to estimate the body’s circadian rhythm inversely based on the measured biometric signals, a mathematical algorithm is needed. This algorithm plays a key role in accurately identifying the characteristics of circadian rhythms by extracting hidden physiological patterns from biosignals. >
Professor Dae Wook Kim said, "It is very meaningful to be able to conduct research that provides a clue for ways to apply wearable biometric data using mathematics that have not previously been utilized for actual disease management." He added, "We expect that this research will be able to present continuous and non-invasive mental health monitoring technology. This is expected to present a new paradigm for mental health care. By resolving some of the major problems socially disadvantaged people may face in current treatment practices, they may be able to take more active steps when experiencing symptoms of depression, such as seeking counsel before things get out of hand."
< Figure 4. A mathematical algorithm was devised to circumvent the problems of estimating the phase of the brain's biological clock and sleep stages inversely from the biodata collected by a smartwatch. This algorithm can estimate the degree of daily circadian rhythm disruption, and this estimate can be used as a digital biomarker to predict depression symptoms. >
The results of this study, in which Professor Dae Wook Kim of the Department of Brain and Cognitive Sciences at KAIST participated as the joint first author and corresponding author, were published in the online version of the international academic journal npj Digital Medicine on December 5, 2024. (Paper title: The real-world association between digital markers of circadian disruption and mental health risks) DOI: 10.1038/s41746-024-01348-6
This study was conducted with the support of the KAIST's Research Support Program for New Faculty Members, the US National Science Foundation, the US National Institutes of Health, and the US Army Research Institute MURI Program.
2025.01.20 View 4322 -
KAIST Professor Uichin Lee Receives Distinguished Paper Award from ACM
< Photo. Professor Uichin Lee (left) receiving the award >
KAIST (President Kwang Hyung Lee) announced on the 25th of October that Professor Uichin Lee’s research team from the School of Computing received the Distinguished Paper Award at the International Joint Conference on Pervasive and Ubiquitous Computing and International Symposium on Wearable Computing (Ubicomp / ISWC) hosted by the Association for Computing Machinery (ACM) in Melbourne, Australia on October 8.
The ACM Ubiquitous Computing Conference is the most prestigious international conference where leading universities and global companies from around the world present the latest research results on ubiquitous computing and wearable technologies in the field of human-computer interaction (HCI).
The main conference program is composed of invited papers published in the Proceedings of the ACM (PACM) on Interactive, Mobile, Wearable and Ubiquitous Technologies (IMWUT), which covers the latest research in the field of ubiquitous and wearable computing.
The Distinguished Paper Award Selection Committee selected eight papers among 205 papers published in Vol. 7 of the ACM Proceedings (PACM IMWUT) that made outstanding and exemplary contributions to the research community. The committee consists of 16 prominent experts who are current and former members of the journal's editorial board which made the selection after a rigorous review of all papers for a period that stretched over a month.
< Figure 1. BeActive mobile app to promote physical activity to form active lifestyle habits >
The research that won the Distinguished Paper Award was conducted by Dr. Junyoung Park, a graduate of the KAIST Graduate School of Data Science, as the 1st author, and was titled “Understanding Disengagement in Just-in-Time Mobile Health Interventions”
Professor Uichin Lee’s research team explored user engagement of ‘Just-in-Time Mobile Health Interventions’ that actively provide interventions in opportune situations by utilizing sensor data collected from health management apps, based on the premise that these apps are aptly in use to ensure effectiveness.
< Figure 2. Traditional user-requested digital behavior change intervention (DBCI) delivery (Pull) vs. Automatic transmission (Push) for Just-in-Time (JIT) mobile DBCI using smartphone sensing technologies >
The research team conducted a systematic analysis of user disengagement or the decline in user engagement in digital behavior change interventions. They developed the BeActive system, an app that promotes physical activities designed to help forming active lifestyle habits, and systematically analyzed the effects of users’ self-control ability and boredom-proneness on compliance with behavioral interventions over time.
The results of an 8-week field trial revealed that even if just-in-time interventions are provided according to the user’s situation, it is impossible to avoid a decline in participation. However, for users with high self-control and low boredom tendency, the compliance with just-in-time interventions delivered through the app was significantly higher than that of users in other groups.
In particular, users with high boredom proneness easily got tired of the repeated push interventions, and their compliance with the app decreased more quickly than in other groups.
< Figure 3. Just-in-time Mobile Health Intervention: a demonstrative case of the BeActive system: When a user is identified to be sitting for more than 50 mins, an automatic push notification is sent to recommend a short active break to complete for reward points. >
Professor Uichin Lee explained, “As the first study on user engagement in digital therapeutics and wellness services utilizing mobile just-in-time health interventions, this research provides a foundation for exploring ways to empower user engagement.” He further added, “By leveraging large language models (LLMs) and comprehensive context-aware technologies, it will be possible to develop user-centered AI technologies that can significantly boost engagement."
< Figure 4. A conceptual illustration of user engagement in digital health apps. Engagement in digital health apps consists of (1) engagement in using digital health apps and (2) engagement in behavioral interventions provided by digital health apps, i.e., compliance with behavioral interventions. Repeated adherences to behavioral interventions recommended by digital health apps can help achieve the distal health goals. >
This study was conducted with the support of the 2021 Biomedical Technology Development Program and the 2022 Basic Research and Development Program of the National Research Foundation of Korea funded by the Ministry of Science and ICT.
< Figure 5. A conceptual illustration of user disengagement and engagement of digital behavior change intervention (DBCI) apps. In general, user engagement of digital health intervention apps consists of two components: engagement in digital health apps and engagement in behavioral interventions recommended by such apps (known as behavioral compliance or intervention adherence). The distinctive stages of user can be divided into adoption, abandonment, and attrition. >
< Figure 6. Trends of changes in frequency of app usage and adherence to behavioral intervention over 8 weeks, ● SC: Self-Control Ability (High-SC: user group with high self-control, Low-SC: user group with low self-control) ● BD: Boredom-Proneness (High-BD: user group with high boredom-proneness, Low-BD: user group with low boredom-proneness). The app usage frequencies were declined over time, but the adherence rates of those participants with High-SC and Low-BD were significantly higher than other groups. >
2024.10.25 View 5055 -
KAIST begins full-scale cooperation with Taiwan’s Formosa Group
< (From left) Senior Vice President for Planning and Budget Kyung-Soo Kim, and Professor Minee Choi of the Department of Brain and Cognitive Sciences of KAIST along with Chairman of Formosa Group Sandy Wang and KAIST President Kwang-Hyung Lee, and Dean Daesoo Kim of KAIST College of Life Science and Bioengineering >
KAIST is pursuing cooperation in the fields of advanced biotechnology and eco-friendly energy with Formosa Plastics Group, one of Taiwan's three largest companies.
To this end, Chairman Sandy Wang, a member of Formosa Group's standing committee and leader of the group's bio and eco-friendly energy sector, will visit KAIST on the 13th of this month. This is the first time that the owner of Formosa Group has made an official visit to KAIST.
Cooperation between the two institutions began last March when our university signed a memorandum of understanding on comprehensive exchange and cooperation with Ming Chi University of Science and Technology (明志科技大學), Chang Gung University(長庚大學), and Chang Gung Memorial Hospital(長庚記念醫院), three of many institutions established and supported by Formosa Group.
Based on this, Chairman Sandy Wang, who visits our university to promote more exchanges and cooperation, talked about ‘the education of children and corporate social return and practice of his father, Chairman Yung-Ching Wang,’ through a special lecture for the school leadership as a part of the Monthly Lecture on KAIST’s Leadership Innovation Day.
She then visited KAIST's research and engineering facilities related to Taiwan's future industries, such as advanced biotechnology and eco-friendly energy, and discussed global industry-academic cooperation plans. In the future, the two organizations plan to appoint adjunct professors and promote practical global cooperation, including joint student guidance and research cooperation. We plan to pursue effective mid- to long-term cooperation, such as conducting battery application research with the KAIST Next-Generation ESS Research Center and opening a graduate program specialized in stem cell and gene editing technology in connection with Chang Gung University and Chang Gung Memorial Hospital. The newly established cooperative relationship will also promote Formosa Group's investment and cooperation with KAIST's outstanding venture companies related to bio and eco-friendly energy to lay the foundation for innovative industrial cooperation between Taiwan and Korea.
President Kwang-Hyung Lee said, “The Formosa Group has a global network, so we regard it to be a key partner that will position KAIST’s bio and engineering technology in the global stages.” He also said, “With Chairman Sandy Wang’s visit, Taiwan is emerging as a global economic powerhouse,” and added, “We expect to continue our close cooperative relationship with the company.”
Formosa Group is a company founded by the late Chairman Yung-Ching Wang, the father of Chairman Sandy Wang. As the world's No. 1 plastic PVC producer, it is leading the core industries of Taiwan's economy, including semiconductors, steel, heavy industry, bio, and batteries. Chairman Yung-Ching Wang was respected by the Taiwanese people by setting an example of returning his wealth to society under the belief that the companies and assets he built ‘belonged to the people.’ Chang Gung University, Chang Gung Memorial Hospital, and Ming Chi University of Technology, which are pursuing cooperation with our university, were also established as part of the social contribution promoted by Chairman Yung-Ching Wang and are receiving financial support from Formosa Group.
2024.05.09 View 4515 -
KAIST Research Team Develops Sweat-Resistant Wearable Robot Sensor
New electromyography (EMG) sensor technology that allows the long-term stable control of wearable robots and is not affected by the wearer’s sweat and dead skin has gained attention recently. Wearable robots are devices used across a variety of rehabilitation treatments for the elderly and patients recovering from stroke or trauma.
A joint research team led by Professor Jae-Woong Jung from the KAIST School of Electrical Engineering (EE) and Professor Jung Kim from the KAIST Department of Mechanical Engineering (ME) announced on January 23rd that they have successfully developed a stretchable and adhesive microneedle sensor that can electrically sense physiological signals at a high level without being affected by the state of the user’s skin.
For wearable robots to recognize the intentions behind human movement for their use in rehabilitation treatment, they require a wearable electrophysiological sensor that gives precise EMG measurements. However, existing sensors often show deteriorating signal quality over time and are greatly affected by the user’s skin conditions. Furthermore, the sensor’s higher mechanical hardness causes noise since the contact surface is unable to keep up with the deformation of the skin. These shortcomings limit the reliable, long-term control of wearable robots.
< Figure 1. Design and working concept of the Stretchable microNeedle Adhesive Patch (SNAP). (A) Schematic illustration showing the overall system configuration and application of SNAP. (B) Exploded view schematic diagram of a SNAP, consisting of stretchable serpentine interconnects, Au-coated Si microneedle, and ECA made of Ag flakes–silicone composite. (C) Optical images showing high mechanical compliance of SNAP. >
However, the recently developed technology is expected to allow long-term and high-quality EMG measurements as it uses a stretchable and adhesive conducting substrate integrated with microneedle arrays that can easily penetrate the stratum corneum without causing discomfort. Through its excellent performance, the sensor is anticipated to be able to stably control wearable robots over a long period of time regardless of the wearer’s changing skin conditions and without the need for a preparation step that removes sweat and dead cells from the surface of their skin.
The research team created a stretchable and adhesive microneedle sensor by integrating microneedles into a soft silicon polymer substrate. The hard microneedles penetrate through the stratum corneum, which has high electrical resistance. As a result, the sensor can effectively lower contact resistance with the skin and obtain high-quality electrophysiological signals regardless of contamination. At the same time, the soft and adhesive conducting substrate can adapt to the skin’s surface that stretches with the wearer’s movement, providing a comfortable fit and minimizing noise caused by movement.
< Figure 2. Demonstration of the wireless Stretchable microNeedle Adhesive Patch (SNAP) system as an Human-machine interfaces (HMI) for closed-loop control of an exoskeleton robot. (A) Illustration depicting the system architecture and control strategy of an exoskeleton robot. (B) The hardware configuration of the pneumatic back support exoskeleton system. (C) Comparison of root mean square (RMS) of electromyography (EMG) with and without robotic assistance of pretreated skin and non-pretreated skin. >
To verify the usability of the new patch, the research team conducted a motion assistance experiment using a wearable robot. They attached the microneedle patch on a user’s leg, where it could sense the electrical signals generated by the muscle. The sensor then sent the detected intention to a wearable robot, allowing the robot to help the wearer lift a heavy object more easily.
Professor Jae-Woong Jung, who led the research, said, “The developed stretchable and adhesive microneedle sensor can stability detect EMG signals without being affected by the state of a user’s skin. Through this, we will be able to control wearable robots with higher precision and stability, which will help the rehabilitation of patients who use robots.”
The results of this research, written by co-first authors Heesoo Kim and Juhyun Lee, who are both Ph.D. candidates in the KAIST School of EE, were published in Science Advances on January 17th under the title “Skin-preparation-free, stretchable microneedle adhesive patches for reliable electrophysiological sensing and exoskeleton robot control”.
This research was supported by the Bio-signal Sensor Integrated Technology Development Project by the National Research Foundation of Korea, the Electronic Medicinal Technology Development Project, and the Step 4 BK21 Project.
2024.01.30 View 6818 -
KAIST Professor Jiyun Lee becomes the first Korean to receive the Thurlow Award from the American Institute of Navigation
< Distinguished Professor Jiyun Lee from the KAIST Department of Aerospace Engineering >
KAIST (President Kwang-Hyung Lee) announced on January 27th that Distinguished Professor Jiyun Lee from the KAIST Department of Aerospace Engineering had won the Colonel Thomas L. Thurlow Award from the American Institute of Navigation (ION) for her achievements in the field of satellite navigation.
The American Institute of Navigation (ION) announced Distinguished Professor Lee as the winner of the Thurlow Award at its annual awards ceremony held in conjunction with its international conference in Long Beach, California on January 25th. This is the first time a person of Korean descent has received the award.
The Thurlow Award was established in 1945 to honor Colonel Thomas L. Thurlow, who made significant contributions to the development of navigation equipment and the training of navigators. This award aims to recognize an individual who has made an outstanding contribution to the development of navigation and it is awarded to one person each year. Past recipients include MIT professor Charles Stark Draper, who is well-known as the father of inertial navigation and who developed the guidance computer for the Apollo moon landing project.
Distinguished Professor Jiyun Lee was recognized for her significant contributions to technological advancements that ensure the safety of satellite-based navigation systems for aviation. In particular, she was recognized as a world authority in the field of navigation integrity architecture design, which is essential for ensuring the stability of intelligent transportation systems and autonomous unmanned systems. Distinguished Professor Lee made a groundbreaking contribution to help ensure the safety of satellite-based navigation systems from ionospheric disturbances, including those affected by sudden changes in external factors such as the solar and space environment.
She has achieved numerous scientific discoveries in the field of ionospheric research, while developing new ionospheric threat modeling methods, ionospheric anomaly monitoring and mitigation techniques, and integrity and availability assessment techniques for next-generation augmented navigation systems. She also contributed to the international standardization of technology through the International Civil Aviation Organization (ICAO).
Distinguished Professor Lee and her research group have pioneered innovative navigation technologies for the safe and autonomous operation of unmanned aerial vehicles (UAVs) and urban air mobility (UAM). She was the first to propose and develop a low-cost navigation satellite system (GNSS) augmented architecture for UAVs with a near-field network operation concept that ensures high integrity, and a networked ground station-based augmented navigation system for UAM. She also contributed to integrity design techniques, including failure monitoring and integrity risk assessment for multi-sensor integrated navigation systems.
< Professor Jiyoon Lee upon receiving the Thurlow Award >
Bradford Parkinson, professor emeritus at Stanford University and winner of the 1986 Thurlow Award, who is known as the father of GPS, congratulated Distinguished Professor Lee upon hearing that she was receiving the Thurlow Award and commented that her innovative research has addressed many important topics in the field of navigation and her solutions are highly innovative and highly regarded.
Distinguished Professor Lee said, “I am very honored and delighted to receive this award with its deep history and tradition in the field of navigation.” She added, “I will strive to help develop the future mobility industry by securing safe and sustainable navigation technology.”
2024.01.26 View 5908 -
North Korea and Beyond: AI-Powered Satellite Analysis Reveals the Unseen Economic Landscape of Underdeveloped Nations
- A joint research team in computer science, economics, and geography has developed an artificial intelligence (AI) technology to measure grid-level economic development within six-square-kilometer regions.
- This AI technology is applicable in regions with limited statistical data (e.g., North Korea), supporting international efforts to propose policies for economic growth and poverty reduction in underdeveloped countries.
- The research team plans to make this technology freely available for use to contribute to the United Nations' Sustainable Development Goals (SDGs).
The United Nations reports that more than 700 million people are in extreme poverty, earning less than two dollars a day. However, an accurate assessment of poverty remains a global challenge. For example, 53 countries have not conducted agricultural surveys in the past 15 years, and 17 countries have not published a population census. To fill this data gap, new technologies are being explored to estimate poverty using alternative sources such as street views, aerial photos, and satellite images.
The paper published in Nature Communications demonstrates how artificial intelligence (AI) can help analyze economic conditions from daytime satellite imagery. This new technology can even apply to the least developed countries - such as North Korea - that do not have reliable statistical data for typical machine learning training.
The researchers used Sentinel-2 satellite images from the European Space Agency (ESA) that are publicly available. They split these images into small six-square-kilometer grids. At this zoom level, visual information such as buildings, roads, and greenery can be used to quantify economic indicators. As a result, the team obtained the first ever fine-grained economic map of regions like North Korea. The same algorithm was applied to other underdeveloped countries in Asia: North Korea, Nepal, Laos, Myanmar, Bangladesh, and Cambodia (see Image 1).
The key feature of their research model is the "human-machine collaborative approach," which lets researchers combine human input with AI predictions for areas with scarce data. In this research, ten human experts compared satellite images and judged the economic conditions in the area, with the AI learning from this human data and giving economic scores to each image. The results showed that the Human-AI collaborative approach outperformed machine-only learning algorithms.
< Image 1. Nightlight satellite images of North Korea (Top-left: Background photo provided by NASA's Earth Observatory). South Korea appears brightly lit compared to North Korea, which is mostly dark except for Pyongyang. In contrast, the model developed by the research team uses daytime satellite imagery to predict more detailed economic predictions for North Korea (top-right) and five Asian countries (Bottom: Background photo from Google Earth). >
The research was led by an interdisciplinary team of computer scientists, economists, and a geographer from KAIST & IBS (Donghyun Ahn, Meeyoung Cha, Jihee Kim), Sogang University (Hyunjoo Yang), HKUST (Sangyoon Park), and NUS (Jeasurk Yang). Dr Charles Axelsson, Associate Editor at Nature Communications, handled this paper during the peer review process at the journal.
The research team found that the scores showed a strong correlation with traditional socio-economic metrics such as population density, employment, and number of businesses. This demonstrates the wide applicability and scalability of the approach, particularly in data-scarce countries. Furthermore, the model's strength lies in its ability to detect annual changes in economic conditions at a more detailed geospatial level without using any survey data (see Image 2).
< Image 2. Differences in satellite imagery and economic scores in North Korea between 2016 and 2019. Significant development was found in the Wonsan Kalma area (top), one of the tourist development zones, but no changes were observed in the Wiwon Industrial Development Zone (bottom). (Background photo: Sentinel-2 satellite imagery provided by the European Space Agency (ESA)). >
This model would be especially valuable for rapidly monitoring the progress of Sustainable Development Goals such as reducing poverty and promoting more equitable and sustainable growth on an international scale. The model can also be adapted to measure various social and environmental indicators. For example, it can be trained to identify regions with high vulnerability to climate change and disasters to provide timely guidance on disaster relief efforts.
As an example, the researchers explored how North Korea changed before and after the United Nations sanctions against the country. By applying the model to satellite images of North Korea both in 2016 and in 2019, the researchers discovered three key trends in the country's economic development between 2016 and 2019. First, economic growth in North Korea became more concentrated in Pyongyang and major cities, exacerbating the urban-rural divide. Second, satellite imagery revealed significant changes in areas designated for tourism and economic development, such as new building construction and other meaningful alterations. Third, traditional industrial and export development zones showed relatively minor changes.
Meeyoung Cha, a data scientist in the team explained, "This is an important interdisciplinary effort to address global challenges like poverty. We plan to apply our AI algorithm to other international issues, such as monitoring carbon emissions, disaster damage detection, and the impact of climate change."
An economist on the research team, Jihee Kim, commented that this approach would enable detailed examinations of economic conditions in the developing world at a low cost, reducing data disparities between developed and developing nations. She further emphasized that this is most essential because many public policies require economic measurements to achieve their goals, whether they are for growth, equality, or sustainability.
The research team has made the source code publicly available via GitHub and plans to continue improving the technology, applying it to new satellite images updated annually. The results of this study, with Ph.D. candidate Donghyun Ahn at KAIST and Ph.D. candidate Jeasurk Yang at NUS as joint first authors, were published in Nature Communications under the title "A human-machine collaborative approach measures economic development using satellite imagery."
< Photos of the main authors. 1. Donghyun Ahn, PhD candidate at KAIST School of Computing 2. Jeasurk Yang, PhD candidate at the Department of Geography of National University of Singapore 3. Meeyoung Cha, Professor of KAIST School of Computing and CI at IBS 4. Jihee Kim, Professor of KAIST School of Business and Technology Management 5. Sangyoon Park, Professor of the Division of Social Science at Hong Kong University of Science and Technology 6. Hyunjoo Yang, Professor of the Department of Economics at Sogang University >
2023.12.07 View 7250 -
A KAIST research team identifies a cause of mental diseases induced by childhood abuse
Childhood neglect and/or abuse can induce extreme stress that significantly changes neural networks and functions during growth. This can lead to mental illnesses, including depression and schizophrenia, but the exact mechanism and means to control it were yet to be discovered.
On August 1, a KAIST research team led by Professor Won-Suk Chung from the Department of Biological Sciences announced the identification of excessive synapse removal mediated by astrocytes as the cause of mental diseases induced by childhood abuse trauma. Their research was published in Immunity, a top international journal in the field of immunology.
The research team discovered that the excessive astrocyte-mediated removal of excitatory synapses in the brain in response to stress hormones is a cause of mental diseases induced by childhood neglect and abuse. Clinical data have previously shown that high levels of stress can lead to various mental diseases, but the exact mechanism has been unknown. The results of this research therefore are expected to be widely applied to the prevention and treatment of such diseases.
The research team clinically screened an FDA-approved drug to uncover the mechanism that regulates the phagocytotic role of astrocytes, in which they capture external substances and eliminate them. As a result, the team found that synthetic glucocorticoids, namely stress hormones, enhanced astrocyte-mediated phagocytosis to an abnormal level. Glucocorticoids play essential roles in processes that maintain life, such as carbohydrate metabolism and anti-inflammation, but are also secreted in response to external stimuli such as stress, allowing the body to respond appropriately. However, excessive and long-term exposure to glucocorticoids caused by chronic stress can lead to various mental diseases including depression, cognitive disorders, and anxiety.
< Figure 1. Results of screening for compounds that increase astrocyte phagocytosis
(A) Discovered that synthetic glucocorticoid (stress hormone) increases the phagocytosis of astrocytes through screening of FDA-approved clinical compounds. (B-C) When treated with stress hormones, the phagocytosis of astrocytes is greatly increased, but this phenomenon is strongly suppressed by the GR antagonist (Mifepristone). CORT: corticosterone (stress hormone), Eplerenone: mineralocorticoid receptor (MR) antagonist, Mifepristone: glucocorticoid receptor (GR) antagonist >
To understand the changes in astrocyte functions caused by childhood stress, the research team used mice models with early social deprivation, and discovered that stress hormones bind to the glucocorticoid receptors (GRs) of astrocytes. This significantly increased the expression of Mer tyrosine kinase (MERK), which plays an essential role in astrocyte phagocytosis. Surprisingly, out of the various neurons in the cerebral cortex, astrocytes would eliminate only the excitatory synapses of specific neurons. The team found that this builds abnormal neural networks, which can lead to complex behavioral abnormalities such as social deficiencies and depression in adulthood.
The team also observed that microglia, which also play an important role in cerebral immunity, did not contribute to synapse removal in the mice models with early social deprivation. This confirms that the response to stress hormones during childhood is specifically astrocyte-mediated.
To find out whether these results are also applicable in humans, the research team used a brain organoid grown from human-induced pluripotent stem cells to observe human responses to stress hormones. The team observed that the stress hormones induced astrocyte GRs and phagocyte activation in the human brain organoid as well, and confirmed that the astrocytes subsequently eliminated excessive amounts of excitatory synapses. By showing that mice and humans both showed the same synapse control mechanism in response to stress, the team suggested that this discovery is applicable to mental disorders in humans.
< Figure 2. A schematic diagram of the study published in Immunity. Excessive stress hormone secretion in childhood increases the expression of the MERTK phagocytic receptor through the glucocorticoid receptor (GR) of astrocytes, resulting in excessive elimination of excitatory synapses. Excessive synaptic elimination by astrocytes during brain development causes permanent damage to brain circuits, resulting in abnormal neural activity in the adult brain and psychiatric behaviors such as depression and anti-social tendencies. >
Prof. Won-Suk Chung said, “Until now, we did not know the exact mechanism for how childhood stress caused brain diseases. This research was the first to show that the excessive phagocytosis of astrocytes could be an important cause of such diseases.” He added, “In the future, controlling the immune response of astrocytes will be used as a fundamental target for understanding and treating brain diseases.”
This research, written by co-first authors Youkyeong Byun (Ph.D. candidate) and Nam-Shik Kim (post-doctoral associate) from the KAIST Department of Biological Sciences, was published in the internationally renowned journal Immunity, a sister magazine of Cell and one of the best journal in the field of immunology, on July 31 under the title "Stress induces behavioral abnormalities by increasing expression of phagocytic receptor MERTK in astrocytes to promote synapse phagocytosis."
This work was supported by a National Research Foundation of Korea grant, the Korea Health Industry Development Institute (KHIDI), and the Korea Dementia Research Center (KDRC).
2023.08.04 View 6699 -
A KAIST research team develops a high-performance modular SSD system semiconductor
In recent years, there has been a rise in demand for large amounts of data to train AI models and, thus, data size has become increasingly important over time. Accordingly, solid state drives (SSDs, storage devices that use a semiconductor memory unit), which are core storage devices for data centers and cloud services, have also seen an increase in demand. However, the internal components of higher performing SSDs have become more tightly coupled, and this tightly-coupled structure limits SSD from maximized performance.
On June 15, a KAIST research team led by Professor Dongjun Kim (John Kim) from the School of Electrical Engineering (EE) announced the development of the first SSD system semiconductor structure that can increase the reading/writing performance of next generation SSDs and extend their lifespan through high-performance modular SSD systems.
Professor Kim’s team identified the limitations of the tightly-coupled structures in existing SSD designs and proposed a de-coupled structure that can maximize SSD performance by configuring an internal on-chip network specialized for flash memory. This technique utilizes on-chip network technology, which can freely send packet-based data within the chip and is often used to design non-memory system semiconductors like CPUs and GPUs. Through this, the team developed a ‘modular SSD’, which shows reduced interdependence between front-end and back-end designs, and allows their independent design and assembly.
*on-chip network: a packet-based connection structure for the internal components of system semiconductors like CPUs/GPUs. On-chip networks are one of the most critical design components for high-performing system semiconductors, and their importance grows with the size of the semiconductor chip.
Professor Kim’s team refers to the components nearer to the CPU as the front-end and the parts closer to the flash memory as back-end. They newly constructed an on-chip network specific to flash memory in order to allow data transmission between the back-end’s flash controller, proposing a de-coupled structure that can minimize performance drop.
The SSD can accelerate some functions of the flash translation layer, a critical element to drive the SSD, in order to allow flash memory to actively overcome its limitations. Another advantage of the de-coupled, modular structure is that the flash translation layer is not limited to the characteristics of specific flash memories. Instead, their front-end and back-end designs can be carried out independently. Through this, the team could produce 21-times faster response times compared to existing systems and extend SSD lifespan by 23% by also applying the DDS defect detection technique.
< Figure 1. Schematic diagram of the structure of a high-performance modular SSD system developed by Professor Dong-Jun Kim's team >
This research, conducted by first author and Ph.D. candidate Jiho Kim from the KAIST School of EE and co-author Professor Myoungsoo Jung, was presented on the 19th of June at the 50th IEEE/ACM International Symposium on Computer Architecture, the most prestigious academic conference in the field of computer architecture, held in Orlando, Florida. (Paper Title: Decoupled SSD: Rethinking SSD Architecture through Network-based Flash Controllers)
< Figure 2. Conceptual diagram of hardware acceleration through high-performance modular SSD system >
Professor Dongjun Kim, who led the research, said, “This research is significant in that it identified the structural limitations of existing SSDs, and showed that on-chip network technology based on system memory semiconductors like CPUs can drive the hardware to actively carry out the necessary actions. We expect this to contribute greatly to the next-generation high-performance SSD market.” He added, “The de-coupled architecture is a structure that can actively operate to extend devices’ lifespan. In other words, its significance is not limited to the level of performance and can, therefore, be used for various applications.”
KAIST commented that this research is also meaningful in that the results were reaped through a collaborative study between two world-renowned researchers: Professor Myeongsoo Jung, recognized in the field of computer system storage devices, and Professor Dongjun Kim, a leading researcher in computer architecture and interconnection networks.
This research was funded by the National Research Foundation of Korea, Samsung Electronics, the IC Design Education Center, and Next Generation Semiconductor Technology and Development granted by the Institute of Information & Communications Technology, Planning & Evaluation.
2023.06.23 View 6677 -
KAIST research team develops a forgery prevention technique using salmon DNA
The authenticity scandal that plagued the artwork “Beautiful Woman” by Kyung-ja Chun for 30 years shows how concerns about replicas can become a burden to artists, as most of them are not experts in the field of anti-counterfeiting. To solve this problem, artist-friendly physical unclonable functions (PUFs) based on optical techniques instead of electronic ones, which can be applied immediately onto artwork through brushstrokes are needed.
On May 23, a KAIST research team led by Professor Dong Ki Yoon in the Department of Chemistry revealed the development of a proprietary technology for security and certification using random patterns that occur during the self-assembly of soft materials.
With the development of the Internet of Things in recent years, various electronic devices and services can now be connected to the internet and carry out new innovative functions. However, counterfeiting technologies that infringe on individuals’ privacy have also entered the marketplace.
The technique developed by the research team involves random and spontaneous patterns that naturally occur during the self-assembly of two different types of soft materials, which can be used in the same way as human fingerprints for non-replicable security. This is very significant in that even non-experts in the field of security can construct anti-counterfeiting systems through simple actions like drawing a picture.
The team developed two unique methods. The first method uses liquid crystals. When liquid crystals become trapped in patterned substrates, they induce the symmetrical destruction of the structure and create a maze-like topology (Figure 1). The research team defined the pathways open to the right as 0 (blue), and those open to the left as 1 (red), and confirmed that the structure could be converted into a digital code composed of 0’s and 1’s that can serve as a type of fingerprint through object recognition using machine learning. This groundbreaking technique can be utilized by non-experts, as it does not require complex semiconductor patterns that are required by existing technology, and can be observed through the level of resolution of a smartphone camera. In particular, this technique can reconstruct information more easily than conventional methods that use semiconductor chips.
< Figure 1. Security technology using the maze made up of magnetically-assembled structures formed on a substrate patterned with liquid crystal materials. >
The second method uses DNA extracted from salmon. The DNA can be dissolved in water and applied with a brush to induce bulking instability, which forms random patterns similar to a zebra’s stripes. Here, the patterns create ridge endings and bifurcation, which are characteristics in fingerprints, and these can also be digitalized into 0’s and 1’s through machine learning. The research team applied conventional fingerprint recognition technology to this patterning technique and demonstrated its use as an artificial fingerprint. This method can be easily carried out using a brush, and the solution can be mixed into various colors and used as a new security ink.
< Figure 2. Technology to produce security ink using DNA polymers extracted from salmon >
This new security technology developed by the research team uses only simple organic materials and requires basic manufacturing processes, making it possible to enhance security at a low cost. In addition, users can produce patterns in the shapes and sizes they want, and even if the patterns are made in the same way, their randomness makes each individual pattern different. This provides high levels of security and gives the technique enhanced marketability.
Professor Dong Ki Yoon said, “These studies have taken the randomness that naturally occurs during self-assembly to create non-replicable patterns that can act like human fingerprints.” He added, “These ideas will be the cornerstone of technology that applies the many randomities that exist in nature to security systems.”
The two studies were published in the journal Advanced Materials under the titles “1Planar Spin Glass with Topologically-Protected Mazes in the Liquid Crystal Targeting for Reconfigurable Micro Security Media” and “2Paintable Physical Unclonable Function Using DNA” on May 6 and 5, respectively.
Author Information: 1Geonhyeong Park, Yun-Seok Choi, S. Joon Kwon*, and Dong Ki Yoon*/ 2Soon Mo Park†, Geonhyeong Park†, Dong Ki Yoon*: †co-first authors, *corresponding author
This research was funded by the Center for Multiscale Chiral Architectures and supported by the Ministry of Science and ICT-Korea Research Foundation, BRIDGE Convergent Research and Development Program, the Running Together Project, and the Samsung Future Technology Development Program.
< Figure 1-1. A scene from the schematic animation of the process of Blues (0) and Reds (1) forming the PUF by exploring the maze. From "Planar Spin Glass with Topologically-Protected Mazes in the Liquid Crystal Targeting for Reconfigurable Micro Security Media" by Geonhyeong Park, Yun-Seok Choi, S. Joon Kwon, Dong Ki Yoon. https://doi.org/10.1002/adma.202303077 >
< Figure 2-1. A schematic diagram of the formation of digital fingerprints formed using the DNA ink. From "Paintable Physical Unclonable Function Using DNA" by Soon Mo Park, Geonhyeong Park, Dong Ki Yoon. https://doi.org/10.1002/adma.202302135 >
2023.06.08 View 6864