Project No.: R1011-25
Project Title: Development of Decoupled Radiant Cooling Technique for Advancing Building Energy Efficiency under Hot and Humid Climates
Project Coordinator: Prof Gongsheng HUANG (CityU)
Abstract
Traditional air-conditioning systems are energy inefficient, using blown air to cool an entire indoor space and consuming up to 60% of the total electricity used by buildings in Hong Kong. In contrast, radiant cooling uses cooling panels filled with cold water to absorb heat away from occupants directly through radiant heat exchange, potentially reducing cooling energy use by 20%–40%. However, radiant cooling suffers heavily from limited cooling capacity as the temperature of cooling panels must be higher than the dew point to prevent condensation. Current achievable cooling capacity is less than 60 W/m2 in practice, much lower than 100 W/m2, a typical cooling demand in hot and humid climates. To achieve a cooling capacity of 100 W/m2, we have developed a decoupled radiant cooling (DRC) technique. This technique solves the condensation issue by using an air layer sealed with a high-infrared-transparency membrane in the cooling panel to separate the radiant cooling surface from the air-contract surface. Such a DRC panel can achieve a cooling capacity of 110 W/m2 without condensation at a radiant cooling temperature of 5◦C. However, there are two main challenges to the commercialization of DRC panels: a lack of suitable membrane materials; and a lack of knowledge on effective deployment of DRC systems. To address these two challenges, we will carry out the following research: (1) to develop membrane materials with the desired infrared transmittance and durability, and a method for manufacturing these materials at scale; (2) to develop models for the proposed DRC panels, evaluate their cooling performance, and provide optimization strategies for the deployment of DRC panels; (3) to characterize the thermal environment and develop intelligent control strategies for DRC panels to enhance energy efficiency in operation; and (4) to develop a DRC panel design toolkit to facilitate the optimal design and deployment of DRC panels and demonstrate the energy savings and carbon emission reduction of DRC systems and compare them with conventional air-conditioning systems. The main deliverables of this project – materials, models, methods, strategies, and the design toolkit – will be valuable resources for other researchers and other professionals in the building management and heating, ventilation, and air-conditioning (HVAC) industries. In the long term, installed DRC panels will enhance the energy efficiency of buildings in hot and humid climates, lowering costs for building managers and owners; reduce carbon emissions, contributing to Hong Kong's Climate Action Plan 2050 and the UN’s SDGs, including Industry, Innovation, and Infrastructure; Sustainable Cities and Communities; Responsible Consumption and Production; and improve the health and well-being of building users and occupants.
Project No.: R1018-25
Project Title: CuCloud: AI-Enabled Green Ecosystem
Project Coordinator: Prof WU Dapeng Oliver (CityU)
Abstract
CuCloud is an innovative Artificial Intelligence (AI) platform designed to make AI services more flexible, environmentally friendly, and accessible to society. Today, many AI systems require frequent hardware replacements to accommodate new functions, resulting in high costs, electronic waste, and increased carbon emissions. CuCloud addresses this problem by allowing computing hardware to be purchased once and then reconfigured repeatedly to support different AI services through software, rather than requiring device replacements.
Utilizing an open and green architecture, CuCloud can automatically create and deploy AI services tailored to user needs. For example, the same hardware can be used for respiration monitoring today and heart or fall monitoring tomorrow. The system relies on cloud intelligence and edge devices, enabling services to be updated, improved, and personalized over time without disrupting daily use.
A major application of CuCloud is in healthcare, particularly in elderly care. The platform can support continuous, contactless monitoring of vital signs, including breathing, heart activity, and falls, and automatically send alerts to caregivers when abnormalities are detected. This helps reduce the burden on families, caregivers, and healthcare systems while improving safety and quality of life for older adults.
Beyond healthcare, CuCloud can be applied to various areas, including telemedicine, smart city safety, disaster alerts, and education. By reducing electronic waste, saving energy, and enabling smarter public services, CuCloud aims to contribute to a more sustainable, resilient, and people-centered digital society.
Project No.: R1038-25
Project Title: The UN-endorsed Global Estuaries Monitoring (GEM) Programme: Occurrence and environmental risks of contaminants of emerging concern in the world's estuaries
Project Coordinator: Prof Kenneth Mei Yee LEUNG (CityU)
Abstract
Marine pollution is a major threat to biodiversity and sustainability of ecosystem services. Rapid urbanization, industrialization, intensive agriculture, and population growth in coastal regions have intensified human activities, increasing pollution in estuaries worldwide. However, the global status of contaminants of emerging concern (CECs) in estuaries remains largely unknown. Existing studies on CEC monitoring in estuaries vary in methods and target chemicals, and mostly focus on coastal areas in developed countries, leaving many underdeveloped regions unmonitored and hampering fair global comparisons.
To address these gaps, the Global Estuaries Monitoring (GEM) Programme, led by the State Key Laboratory of Marine Environmental Health at City University of Hong Kong, aims to establish a global monitoring network for quantifying CECs in urban estuaries using standardized methods. Hence, GEM supports the UN Sustainable Development Goals (SDG 14.1 to reduce marine pollution, and SDG 17 to enhance the global partnership). As one of the 63 Action Programs of the UN Decade of Ocean Science for Sustainable Development, GEM’s initial phase (2021-2025) successfully developed a robust standardized method to measure 65 pharmaceuticals in water samples from over 190 estuaries, engaging 120 scientists from 50 countries to assess contamination status.
This proposed research embodies the second phase of GEM, which will establish a specialized Ecotoxicity Testing Laboratory for generating ecotoxicity data and deriving predicted no-effect concentrations (PNECs) of priority CECs to assess their environmental risks in global estuaries. This laboratory will also develop novel bioassays using marine species from previously untested taxonomic groups, and will provide rapid ecotoxicity tests and deliver essential data to the government for timely risk assessment and management. The project will also deploy novel passive samplers (i.e., Artificial Mussels and Smart Sponges), invented by the team to monitor a broader spectrum of CECs including per- and polyfluoroalkyl substances, halogenated flame retardants, liquid crystal monomers, and radionuclides. Furthermore, standardized methods will be developed to monitor environmental microbiomes in estuarine sediments, focusing on antibiotic resistance genes, pathogens, and biodegradation enzymes, alongside CEC monitoring.
These innovative tools, especially the novel passive samplers will be promoted for global utilization, supporting their commercialization and Hong Kong’s role as the International Innovation and Technology Hub. Based on monitoring outcomes, we will co-create strategies with global partners and stakeholders to combat estuarine pollution, aiming to make estuaries cleaner and safer for communities and ecosystems. This research, therefore, positions Hong Kong as a global leader in marine environmental surveillance and policy influence.
Project No.: R1040-25
Project Title: Integrative Study on Needleless (Wire) Electrospinning for Massive Production
Project Coordinator: Prof Hanxiong LI (CityU)
Abstract
Nanofiber materials bring novel application in medicine, energy, advanced textiles and environmental science due to their exceptional physical properties. Currently, advanced needleless electrospinning technology faces difficulties in achieving both high output and high performance. This is because the spinning process is affected by the complex coupling of multiple physical fields, leading the industry to rely on repeated experimentation for process optimization for a long time.
This project aims to investigate key scientific issues and explore solutions in needleless electrospinning through interdisciplinary collaboration, integrating physical and chemical materials, process modeling and digital twins, and artificial intelligence (AI) driven optimization.
Research approach: 1) First, based on the demand for high-throughput spinning, study the optimized formulation of spinning solutions based on AI technology; 2) At the same time, realize the dynamic monitoring of the spinning process via Internet of Things (IoT) sensors network; 3) Construct digital twin models of key processes and optimize the models with AI technology. Relying on this interdisciplinary intelligent integration framework, we will explore the process optimization methods for high-performance spinning, and apply them to actual production to achieve high-yield and high-performance spinning.
Project No.: R4022-25
Project Title: Development of an In-situ Sensing Platform for Deep-sea Exploration
Project Coordinator: Prof Wei REN (CUHK)
Abstract
The exploration of the deep sea, Earth’s final frontier, is crucial for enhancing our understanding of early life origins, sustainable resource development, and ecosystem surveillance. Analyzing dissolved gases such as carbon dioxide, methane, hydrogen sulfide and their isotopes offers insights into hidden chemical and biological processes driving deep-sea phenomena. However, traditional analytical methods, which involve collecting discrete seawater samples for subsequent laboratory analysis, hinder the capability to capture intricate spatial and temporal variations in deep-sea environments. This research will bridge scientific and technical gaps in deep-sea exploration by merging cutting-edge laser spectroscopy with manned submersibles to create a novel in situ sensing platform. A key innovation is a fiber-enhanced photothermal gas sensor that measures laser-absorption-induced variations in gas refractive index. The miniaturized gas sensors, designed with hollow-core fibers and fiber-tip cavities, enable high-sensitivity detection with tiny sample volumes (microliters or nanoliters). Performance tests in laboratory and simulated deep-sea conditions will assess sensing parameters and confirm operational reliability and pressure resistance. In the planned sea trial at the Haima cold seep field (around 1,400 m deep) in the South China Sea, we aim to conduct long-term stationary monitoring of dissolved CO2, CH4, H2S and their isotopes. Mounted on a manned submersible, our sensing platform will trace gas sources and migration patterns, with further deployment planned in hadal trenches (over 5,800 m). This platform will enable the first assessment of the spatial distributions of deep-sea dissolved gases, advancing our understanding of biogeochemical cycles, chemosynthetic ecosystems, and global climate dynamics.
Project No.: R4039-25
Project Title: Consolidating Clinical Translation of the First Locally Developed Advanced Therapy Product (ATP) – Engineered Osteochondral Tissue (eOCT) for Treating Patients with Traumatic Cartilage Injuries
Project Coordinator: Prof Barbara CHAN (CUHK)
Abstract
Cartilage injuries severely impair joint function and quality of life, while current surgical treatments are either ineffective or cause significant side effects. Advanced Therapy Products hold promises but existing options like autologous chondrocyte implantation and mesenchymal stem cell injections fail to consistently regenerate high-quality hyaline cartilage. Building on over a decade of research, supported by multiple funded projects and patents, the research team has developed an engineered osteochondral tissue (eOCT), a single implantable graft that mimics native joint structure, using patients’ own bone marrow stem cells. Preclinical studies demonstrated superior outcomes compared to current therapies. With previous RIF support, the research team bridged translational gaps and proceeded with a first-in-human (FIH) safety trial by partnering with a Good Manufacturing Practice (GMP) facility in Singapore, completing validations, manufacturing clinical batches and successfully implanting eOCT in four patients. Preliminary results show good safety and promising functional improvements. The current project aims to transfer the validated manufacturing process back to Hong Kong and conduct an exploratory efficacy trial on eOCT, paving the way to subsequent clinical trials and registration of the first locally invented, developed and manufactured ATP for cartilage repair.
Project No.: R5001-25
Project Title: Spatial and High-throughput Immunopeptidomics Enabled by Integrated Microfluidics and Proteogenomics (MAP)
Project Coordinator: Prof Qian ZHAO (PolyU)
Abstract
Cancer immunotherapy, teaching the body’s immune system to fight tumors, is one of the most promising frontiers in modern medicine. Although cancer cells are often masters of disguise, they inadvertently reveal identity by displaying tiny molecular "flags" on their surface, known scientifically as immunopeptides or neoantigens. These flags are the keys to the next generation of cancer treatments: if we can identify exactly which flags a tumor is flying, we can design personalized vaccines or cell therapies that train the immune system to attack only the cancer.
The problem we face today is that our current tools for finding these flags are blunt and inefficient. They require large amounts of patient tissue, often more than is safe to remove surgically, and they cannot tell us where in the tumor these flags are located. This leaves us with an incomplete map of the battlefield.
Our Solution is the MAP Platform, which is a "lab-on-a-chip". By using advanced mass spectrometry combined with microfluidics, we can automate the search for these immune flags. This technology is a game-changer for two reasons: 1) Sensitivity: We can detect these flags using a tissue sample a few hundred times smaller than previously required. 2) High Throughput: We can process multiple samples simultaneously, drastically speeding up the discovery process.
We will apply this technology in two critical ways. First, we will create "spatial maps" of tumors, allowing us to see exactly where immune cells interact with cancer cells. Second, we will focus on Glioblastoma, an aggressive and deadly brain cancer. By analyzing hundreds of samples, we aim to find new targets for vaccines that could extend or save lives.
Project No.: R5048-25
Project Title: Towards Life-Cycle Intelligent Predictive Maintenance for Railway Tracks: Advancing with Deep Learning Algorithms and Digital Twin Technology
Project Coordinator: Prof You DONG (PolyU)
Abstract
Railways — including high-speed trains, normal trains, and subways — are essential for daily travel and economic activity. Large systems such as China’s vast rail network and Hong Kong’s busy MTR carry millions of passengers every day. However, maintaining railway tracks is extremely expensive because the metal rails slowly crack and wear out over time, especially as trains become heavier and faster. Current inspection methods often miss early damage or cannot accurately predict when a rail will fail, which leads either to unexpected problems or unnecessary maintenance.
This project aims to make railway maintenance smarter and more efficient. First, this project will study how rails actually get damaged by testing different types of cracking and wear in controlled experiments. Next, they will develop the intelligent monitoring system that combines sensors, cameras, robotic, and artificial intelligence to detect both surface and hidden defects. They will also create computer models that can predict how rails will deteriorate under real operating conditions so maintenance can be planned at the right time.
Finally, the project will build a “digital twin”, a real-time virtual copy of the railway that continuously uses live data and predictions to warn operators early and guide maintenance decisions. Overall, the goal is to improve safety, reduce service disruptions, extend the life of tracks, and lower maintenance costs for modern rail system.
Project No.: R6005-25
Project Title: ZKDI: A Zero-Knowledge Proof-based Data Intelligence System for Financial and Web3 Data Analytics
Project Coordinator: Prof Shuai WANG (HKUST)
Abstract
With the rapid expansion of Web3 and FinTech, there is an urgent need for advanced data analytics that can balance data utility with stringent privacy requirements. This project develops ZKDI, a pioneering Zero-Knowledge Proof (ZKP)-based Data Intelligence system tailored for the evolving financial and Web3 landscapes. Traditional analytics often necessitate exposing sensitive data, posing significant security and compliance risks, especially in complex cross-entity and cross-border collaborations.
To address these challenges, the project focuses on three core pillars: (1) designing ZKP-friendly data intelligence algorithms for secure analytics; (2) optimizing high-performance proof systems for large-scale data; and (3) implementing real-world applications including DeFi protocol auditing, financial risk management, and secure cross-border data sharing. By leveraging innovative ZKP technologies, ZKDI enables financial institutions and regulatory bodies to perform sophisticated data analytics and manage institutional assets without revealing underlying raw data. The successful deployment of ZKDI will significantly enhance the security, transparency, and regulatory compliance of Hong Kong’s financial technology ecosystem, solidifying its position as a leading global hub for both Web3 and FinTech innovation.
Project No.: R7023-25
Project Title: Gallium Nitride Chip Technology for Vertical Power Delivery to AI Processors in Data Centres
Project Coordinator: Prof Yuhao ZHANG (HKU)
Abstract
Data centers that support artificial intelligence (AI) and cloud computing consume an increasing share of global electricity. A significant amount of this energy is wasted as heat during the final stage of delivering power to AI processors, limiting performance and increasing cost and environmental impact.
This project aims to develop a new generation of power semiconductor chips based on gallium nitride (GaN), a material that can convert electricity faster and more efficiently than conventional silicon. A novel “multi-channel” GaN architecture will be created to deliver higher current with lower power loss, while integrating power switches and control circuits on a single chip, referred to as “DrGaN”.
The chips will be tested in practical data-center power supplies. If successful, the technology could significantly reduce energy losses and enable faster, greener, and more cost-effective AI computing.
Project No.: R7025-25
Project Title: Single-base m6A epitranscriptomic profiling: A new frontier in cancer biomarker discovery and precision medicine
Project Coordinator: Prof Chun Ming WONG (HKU)
Abstract
Cancer is a leading cause of death worldwide. While genetic mutations in cancer are well studied, recent research shows that N6-methyladenosine (m6A) modifications on RNA, also play an important role in cancer development. This project will use a new method, GLORI-Seq, to map m6A changes in RNA from lung, breast, and liver cancers with single-base resolution and quantification. We will investigate how these m6A changes relate to disease characteristics and patient outcomes. Using gene-editing tools, we will study how specific m6A marks affect cancer-related genes. Finally, we will develop a sensitive test to detect abnormal m6A marks in blood samples, potentially enabling new, non-invasive methods for early cancer detection and monitoring. By combining advanced m6A mapping, functional studies, and biomarker development, this work aims to create a detailed map of m6A changes in cancers, identify new cancer subtypes, and develop improved diagnostic tools to enhance cancer treatment and patient care.
Project No.: R7057-25
Project Title: Saving Hong Kong oysters from environmental disaster to sustain the livelihoods of growers
Project Coordinator: Prof VENGATESEN, Thiyagarajan (HKU)
Abstract
Aquaculture of the Hong Kong oyster species supports the livelihoods of coastal communities in south China and contributes over 30% of global production. However, this species, which produces tasty white meat, is struggling. Climate change — including ocean warming and acidification—and farming in saltier waters, are causing mass mortality events. Even the surviving oysters often fail to develop their signature plump texture and flavor in time for the winter festival harvest season. This has hurt farmers' incomes and reduced the species' share of both national and global production. This project aims to fix the problem in two main steps: Identify and breed tougher oysters that can withstand winter mortality and future climate stresses. This will be done using a genomic selection pipeline that includes a genetic "fingerprint" test (SNP chip) and machine learning tools. Move the bred oysters to saltier ponds and feed them a special, nutrient-rich algal diet to plump them up, enhancing their sweet, umami taste and quality. Within a few years, the project aims to provide farmers with the technology to cultivate resilient, high-quality oysters. This will allow farming to expand into new areas, sustain incomes, supply better meat for premium oyster sauce, and help the industry adapt to a changing climate.
Project No.: R8003-25
Project Title: Revolutionizing Adapted Physical Activity for Individuals with Physical Disabilities: Examining the Health Impact of and Promoting Sitting Light Volleyball in the Greater Bay Area
Project Coordinator: Dr Ka Man LEUNG (EdUHK)
Abstract
People with physical disabilities (PWPD) are more susceptible to physical inactivity, leaving them more vulnerable to lower health and fitness levels than their physically able-bodied peers. Importantly, PWPD numbers in Hong Kong, China and Chinese Mainland have tripled over the last 20 years. In 2018, the project coordinator co-developed a new adapted physical activity, Sitting Light Volleyball (SLVB), which combines Paralympic Sitting Volleyball and light volleyball with PWPDs. Our team also examined the effectiveness of an SLVB intervention on health outcomes among PWPDs in Hong Kong. The SLVB group demonstrated significant improvements in cardiovascular endurance, body composition, and physical activity enjoyment compared to the control group. To enhance PWPD’s health in Hong Kong, China and Chinese Mainland, the present study investigates the effectiveness of an SLVB intervention on physical and psychological health outcomes in about 220 adults with physical disabilities in Hong Kong using a mixed-methods design. We will then further promote SLVB in Hong Kong, China and Chinese Mainland by organising (i) train-the-trainer workshops, (ii) SLVB classes and (iii) SLVB competitions.