-
Research Highlight
Mobilizing Artificial Intelligence for Maternal Healthcare in Bangladesh
This UNESCAP-APRU-Goolge.org funded Project Supports the Bangladeshi Government to Address Challenges in Pregnancy Monitoring & Health Care Technologies using Artificial Intelligence (AI). The United Nations ESCAP-APRU-Google.org ‘AI for Social Good’ project team includes Dr. Olivia Jensen, Mr. Nathaniel Tan from the National University of Singapore’s Institute for Public Understanding of Risks (NUS IPUR), and Dr. Cornelius Kalenzi from KAIST KPC4IR as lead researchers to support Bangladesh government in strengthening the maternal care health workforce and achieving SDG 3, which will transform methods of treatment as well as the relationships among health professionals and patients. Dr. Jensen, Mr. Tan, and Dr. Cornelius Kalenzi will address challenges in perceptions and reception of incorporating AI into continuous pregnancy monitoring systems, as well as related tech challenges. They will investigate technological issues in Bangladesh’s health care sector and how these impact AI-based data analysis and decision-making processes. Project outcomes will support the Bangladeshi Government in developing policies to promote and enable AI policy frameworks and AI capability-building for social good....read more
-
Research Highlight
Opto-vTrap, an optogenetic trap for reversible inhibition of vesicular release, synaptic transmission, and behavior
Spatiotemporal control of brain activity by optogenetics has emerged as an essential tool to study brain function. To silence brain activity, optogenetic probes, such as halorhodopsin and archaerhodopsin, inhibit transmitter release indirectly by hyperpolarizing membrane potentials. However, these probes cause an undesirable ionic imbalance and rebound spikes. Moreover, they are not applicable to use in non-excitable glial cells. In this study we engineered Opto-vTrap, a light-inducible and reversible inhibition system to temporarily trap the transmitter-containing vesicles from exocytotic release. Light activation of Opto-vTrap caused full vesicle clusterization and complete inhibition of exocytosis within 1min, which recovered within 30 min of light off. We found a significant reduction in synaptic and gliotransmission upon activation of Opto-vTrap in acute brain slices. Opto-vTrap significantly inhibited hippocampus-dependent memory retrieval with full recovery within an hour. We propose Opto-vTrap as a next-generation optogenetic silencer to control brain activity and behavior with minimal confounding effects....read more
-
Research Highlight
Development and Verification of Terahertz Ultra-wideband Beamforming Solution for Next-Generation Mobile Communication
We develop and validate new THz broadband beamforming systems for 6G mobile communications. The controllable beamforming system realizes the characteristics of ultra-wideband and low loss at the same time....read more
-
Research Highlight Top Story
Computational design of a neutralizing antibody against all SARS-CoV-2 variants
Prof. Oh’s group has developed a neutralizing antibody against SARS-CoV-2 by computational method. The antibody binds to all SARS-CoV-2 variants, including Omicron, with pico- to femto-molar binding affinity....read more
-
Research Highlight
Normal tissue complication probability (NTCP) calculator development for research and clinical activities
Radiation therapy is one of the most important tools in the fight against cancer. Generally, tumor control is mainly considered for treatment planning. However, radiation-induced side effects are considered as well. The NTCP is the probability of critical normal tissue having complications as calculated from the combined effect of the radiation dose and volume. Some normal tissue complications are very important for a normal quality of life after radiation therapy. The NTCP calculator is software developed at KI IT Convergence to calculate the NTCP with various methods in a clinical format, DICOM. The Lyman-Kutcher-Burman NTCP model is implemented with a flexible parameter input interface. The NTCP calculator will be used at local university hospitals for research and clinical purposes....read more
-
Research Highlight
KAIST Racing Team finished as a semi-finalist in CES 2022 Autonomous Racing
Researchers at KI Robotics entered CES 2022 Autonomous Racing using their Indy AV-21 autonomous vehicle and competed with the world’s top researchers. Our team developed a cutting-edge autonomous racing software stack from scratch and successfully finished the race as a semi-finalist, only to lose to the top European teams....read more
-
Research Highlight
New chiral ceramic nanostructures: Broad chiroptical activity from ultraviolet (UV) to short-wave infrared (SWIR)
Prof. Yeom’s group has developed new chiral nanostructures with broad chiroptical activity from the UV to the SWIR region to expand the choice of material platforms. ...read more
-
Research Highlight
Fibril-to-Fiber Structure: Hierarchically Fabricated Amyloid Fiber via Evaporation-induced Self-assembly
Prof. Yoon’s group has suggested a facile way to produce protein-based amyloid aggregates, allowing long-range orientational and positional order. Using evaporation-induced self-assembly, a rationalized hierarchical structure will open new opportunities for developing novel functional biomaterials....read more
-
Research Highlight
Tomographic measurement of dielectric tensors at the optical frequency
Prof. YongKeun Park’s group has developed a label-free tomographic method for measuring and reconstructing 3D dielectric tensors, which have not been directly measured thus far. ...read more
-
Research Highlight
Effective functionalization of porous polymer fillers to enhance the CO2/N2 separation performance of mixed-matrix membranes
Prof. Bae’s group has developed a series of porous organic polymers possessing various functional groups, which were incorporated into an in-house polyimide to fabricate mixed-matrix membranes for CO2/N2 separation. Their systematic study revealed that both the structural properties and chemical functionality should be optimized together to obtained an ideal filler material for CO2-selective mixed-matrix membranes. The optimized membrane showed ultrahigh CO2 permeability of 2988 Barrers (158% higher than that of a pure polymer membrane) with a decent enhancement of the CO2/N2 selectivity....read more
291 Daehak-ro Yuseong-gu Daejeon, 34141, Republic of Korea
Partnered with KAIST Breakthroughs and KAIST Compass