Edited by Elysee Chung.
The winners of the National Science and Technology Council's (NSTC) “2024 Future Tech Award” have been announced, with a total of 82 technologies receiving recognition this year. Among them, 13 technologies from National Cheng Kung University (NCKU) were awarded. This year's evaluation continued the two main review criteria of “scientific breakthroughs” and “industrial applications.” The awarded technologies are highly innovative and have strong potential for future commercial development, whether in terms of technological breakthroughs or subsequent business growth.
The purpose of the Future Tech Award is to highlight cutting-edge scientific research achievements and showcase the nation's technological strength. It encourages research outcomes to enter global markets and strengthens international connections. Applications were widely invited from projects funded by the NSTC, Academia Sinica, the Ministry of Education, and the Ministry of Health and Welfare, among others. The submitted technologies were categorized into seven major areas: (1) Chemical Engineering and Materials, (2) AIoT and Smart Living Applications, (3) Green Energy, Environmental Protection, and Net-Zero Technology, (4) Electronics and Optoelectronics, (5) Biotechnology and New Pharmaceuticals, (6) Medical Devices, and (7) Humanities, Sports Technology, and Technological Arts.
Among the 13 award-winning technologies from NCKU, 2 are in the field of Humanity and Technology, 4 in Bio-tech, New Drugs and Medical Devices, 3 in AIoT and Smart Applications, 2 in Evolutionary Materials and Chemical, 1 in Electronics and Optoelectronics, and 1 in Net-Zero Technology.
Humanity and Technology Field:
AI Blended Training Programs and Learning Performance Evaluation System for Badminton Skills Learning and Improvement
Project Leader: Professor Ya-Ting Carolyn Yang (楊雅婷)
This system integrates cloud computing and AI technology, using data analysis, digital assessment, and blended learning to assist badminton training. It adjusts training based on each individual’s learning progress to provide personalized training, enhancing learning outcomes and motivation.
Multi-View Multi Player Tracking Technology
Project Leader: Professor James Jenn-Jier, Lien (連震杰)
This system uses multiple cameras to capture synchronized multi-angle footage of the court. Pose detection technology is used to identify each player's position and body joints. A specially designed multi-view, multi-dimensional trajectory correlation algorithm correlates detection results from different angles and calculates 3D coordinates through triangulation for 3D tracking. This generates the players' 3D movement trajectories. The technology can be applied to match analysis, tactical assessment, and player performance evaluation, providing valuable scientific data to players and coaches.
Bio-tech, New Drugs & Medical Devices Field:
AI Multifunctional Health Management System for Chronic Kidney Disease Index
Project Leader: Professor Lung-Ming, Fu (傅龍明)
The team developed a rapid microfluidic paper-based chip detection system, which utilizes smart IoT technology to upload test results to the cloud for management and analysis. A predictive model is built for analysis, generating personalized medical reports. This provides real-time, accurate assessments and recommendations for doctors and patients, achieving the goal of home testing.
Precision Nano-vesicle Therapy: Missile-Like Extracellular Vesicles Attack on Infectious Microbes
Project Leader: Associate Professor Wei-Hsuan, Hsu (徐瑋萱)
The team built a dynamic culture platform simulating human gut microbiota and innovatively used probiotic exosomes as natural nanocarriers. These exosomes specifically target pathogenic bacteria, precisely delivering nucleic acid drugs to prevent infections. After oral administration, they can target different tissues, effectively inhibiting the proliferation and toxicity of Clostridium difficile without impacting other gut bacteria. This technology has great commercialization potential, offering a cutting-edge research platform for the medical and food industries.
Bead-based Detection of Trace Analytes on An Automated Platform
Project Leader: Professor Han-Sheng, Chuang (莊漢聲)
Utilizing the Stokes-Einstein-Debye equation, the platform quantifies rotational Brownian motion through amphiphilic particles. Since rotational diffusion is inversely proportional to particle size cubed, antibodies attached to amphiphilic particles can capture biomarkers, altering particle size and slowing diffusion. The technology can detect bacteria, protein markers, exosomes, and many more other targets, and is applicable in industries such as medical diagnostics, food safety, environmental monitoring, animal and plant health, and basic research.
Highly Realistic, Multi-functional Phantom System Simulating Thermal Ablation Surgical Training
Project Leader: Professor Yi-Chun, Du (杜翌群)
The team developed a multifunctional high-fidelity phantom system for thermal ablation surgery simulation training using three core technologies: (1) an artificial double-network material capable of mimicking protein tissue characteristics, (2) an interactive simulated human circulatory system, and (3) embedded soft sensor technology. The system underwent feasibility verification through four clinical case studies, significantly improving surgical training effectiveness for new physicians.
AIoT and Smart Applications Field:
A Pioneer Novel Weakly-supervised Multi-instance Learning Framework for Genetic Expression Recognition and Survival Prediction in Digital Pathology Images
Project Leader: Professor Jung-Hsien, Chiang (蔣榮先)
The team successfully addressed the challenges of processing massive pixel images with current AI hardware and the issue of insufficient manual annotations. This allows digital pathology, after training with deep learning models, to better assist doctors in making more precise decisions. The research also validated that specific gene expression features and patient prognosis can be directly identified and predicted from digital pathology images.
A Multispectral Light Source Based Miniaturized Tissue Oxygenation Imaging System for Telemedicine Wound Healing Phases Recognition
Project Leader: Professor Chih-Lung, Lin (林志隆)
This technology focuses on real-time monitoring and healing assessment in chronic wound care using multispectral imaging to provide blood oxygen saturation levels, which are not visible to the naked eye. The system trains an AI algorithm using clinical patient data collected at NCKU Hospital to identify wound tissues and assess healing progress. By integrating miniaturized systems and IoT technology, it allows healthcare professionals and patients to use it in various medical settings and at home, achieving continuous wound tracking and remote medical care.
Toward Reliable Satellite Image Generative Model: Ultra-Fast Hyperspectral Image Compressed Sensing and Fusion
Project Leader: Associate Professor Chih-Chung, Hsu (許志仲)
The team developed innovative techniques for hyperspectral satellite image sensing, fusion, and secure transmission. By combining the advantages of multispectral and hyperspectral imaging, the technology enhances spatial and spectral resolution. It also integrates a mechanism for detecting deepfake images to ensure security. This technology has been published in top journals like IEEE TGRS, not only improving image transmission efficiency but also ensuring the authenticity of the images.
Evolutionary Materials & Chemical Field:
Luminous Aurora Orchid Controlled High Entropy Materials in Indoor Carbon Reduction Technology
Project Leader: Professor Yen-Hsun, Su (蘇彥勳)
The luminescent orchid displays different appearances at night, enhancing the economic value of plants. The luminous cycle can sustainably repeat every night. The technology uses non-GMO, safe biotech materials and nano high-entropy materials. It can capture an amount of carbon equivalent to 43.1% of the carbon emissions per kilowatt-hour of electricity in one month, making it a practical solution for indoor carbon reduction technologies.
Ultrafast Responsive Non-volatile Flash Photomemory and Its Application on Artificial Neural Network
Project Leader: Associate Professor Jung-Yao, Chen (陳蓉瑤)
The team developed an ultra-fast flash photomemory, with a programming time of 0.7 ms, a photoresponsivity of 1.91×10^4 mA W^-1, and 128 levels of memory functionality. This photomemory, combined with a hydrogen-sensitive gasochromic film, can record hydrogen leakage cycles and concentrations, helping to predict leakage locations in pipelines and improving gas leakage management.
Electronics and Optoelectronics Field:
High-sensitivity, Low-cost and Reliable SERS Screening Chips for Rapid Detection for Food Safety
Project Leader: Professor Chen-Kuei, Chung (鍾震桂教授)
This technology is a rapid food safety screening SERS (Surface-Enhanced Raman Spectroscopy) chip. It has demonstrated detection capabilities for melamine at concentrations as low as 0.05 ppm, bisphenol A at 1 ppb, and methylene blue at 1x10^-10 M, all below regulatory standards. The chip can also detect preservatives, antibiotics used in aquaculture, and natural food ingredients.
Net-Zero Technology Field:
Integration of and Power-to-X Technology and Circular Economy for Net-zero Carbon Energy Transition
Project Leader: Professor Kuan-Zong, Fung (方冠榮)
This project focuses on developing low-cost solid oxide electrolysis cells (SOEC) using the concept of a materials circular economy. It addresses the growing issue of discarded lithium batteries by recycling materials and using them in key SOEC technology for future Power-to-X applications. The process not only secures critical metal resources and reduces environmental pollution but also increases the value of the products, paving the way for future net-zero carbon emissions.