Project No.: R4015-19
Project Title: Accelerating Precision Medicine Translation: Development of the First Evidence-based Clinical Exceptional Responder Gene-Drug Sensitivity Resource for Cancer
Project Coordinator: Prof Vivian Wai-yan LUI (CUHK)

Abstract

Precision medicine is a very effective modern cancer treatment impacting patient survival. In brief, a patient can be effectively treated if his/her tumor bears a specific gene defect (mutation, deletion or amplification) causing sensitivity to a specific cancer drug. Yet, as of today, there is only a handful of tumor genetic defects being translated into precision treatments for a few cancer types. In this study, we will investigate the genetics of clinical exceptional responders (ERs) who demonstrated remarkably good clinical responses to certain anti-cancer drugs, mostly targeted therapies. Scientific investigations of exceptional responder genetics and their drug response profiles will inform new clinical trials and new drug usages for cancer treatment. We will conduct solid gene/mutation-drug sensitivity and identify the drug response signatures of 100 exceptional responses. Our results will guide the establishment of the first evidence-based exceptional responder resource in Asia (in English & Chinese languages) to accelerate clinical trials design to achieve higher clinical trial successful rate, and guide future development of new precision medication for cancer.

*Exceptional Responder (ER) definition: As defined by the National Cancer Institute (NCI, USA), Exceptional Responders are patients who show complete response or remarkably good partial response (for more than 6 months) to drugs that are not intended for that indication/cancer type (with responses not expected to be seen in >10% of patients)

Project No.: R4035-19
Project Title: Innovative Diagnosis and Treatment for Shrimp Allergy
Project Coordinator: Prof LEUNG Ting-fan (CUHK)

Abstract

Allergy to shrimp and other types of shellfish affects 2% of global population. Shellfish allergy is the commonest cause of food-induced anaphylaxis, but no curative treatment for shrimp allergy is currently available. Shrimp allergy is conventionally diagnosed by suggestive clinical history and positive skin prick test or blood assay to commercial crude shrimp extract. Whereas clinical history is usually subjective and equivocal, positivity to the latter diagnostic tests does not necessarily imply clinical allergy as false positive and negative results are commonly found. Given the heavy health burden, there is a pressing need to innovate precise diagnostic tests for shrimp allergy. The emerging concept of component-resolved diagnosis (CRD) revolutionalised the precision medicine for food allergy diagnosis over the past decade. The relative contributions of different shrimp allergens as well as their usefulness in refining the accuracy of clinical diagnosis remain unclear. With our recent data implicating a promising potential of CRD for shrimp allergy, this research proposal will make use of machine learning approaches with cross validation procedures to reduce the large number of personal, family, clinical and laboratory parameters from 300 patients to an optimal multi-tier algorithm to innovate shrimp allergy diagnosis. Such approach may also obviate the need for risky and labour-intensive oral food challenges. Regarding the treatment of shrimp allergy, our group constructed two relevant novel hypoallergens of shrimp tropomyosin, and showed therapeutic potentials of hypoallergen DNA vaccine-based treatment in mice with significant reduction in allergic immune responses accompanied by enhanced regulatory immunity. The second part of this proposal, a Phase 1 clinical trial, aims to delineate the safety and immunological effects of our candidate hypoallergen DNA vaccine in human over a 17-month period. The availability of such pivotal data on the safety, tolerability, immunological and functional benefits will be critically important to address the translational potential and commercial impact of our DNA vaccine-based allergen-specific immunotherapy for shrimp allergy. We hope to translate our novel findings from precision diagnosis and therapeutics with shrimp allergens and hypoallergens to clinical impact on enhanced care of shrimp-allergic patients.

Project No.: R4036-19
Project Title: Immigration Detention and Vulnerable Migrants in Hong Kong: Evaluating the System, Facilitating Reform
Project Coordinator: Prof Surabhi CHOPRA (CUHK)

Abstract

This project will evaluate the immigration detention system in Hong Kong, particularly in relation to vulnerable groups of migrants. We will examine the functioning and effects of immigration detention law, policy and practice, with a focus on survivors of torture, victims of human trafficking, and low-income migrant workers. Based upon our findings, we will develop actionable, evidence-based guidelines for improving the system.

This project is not just timely but overdue. It is timely because the Hong Kong Government has recently indicated that it is considering expanding the use of immigration detention (Security Bureau, January 2019). It is overdue because the Government already exercises considerable powers of detention under the Immigration Ordinance (Cap 115). In 2016, 10,749 people were in detention for violations of immigration rules and regulations in Hong Kong, a slight decline from 10,972 in 2015 (Immigration Department, 2016). By comparison, the prison population in Hong Kong was approximately 8,500 in 2016 (Hong Kong Correctional Services, 2019). Despite the high number of immigration detainees, very little information about the system is publicly available, and analysis about it is extremely sparse.

Comparative experience from other countries indicates that many immigration detention regimes have tended to become harsh, procedurally unfair and over-used. Comparative experience also suggests that detainees who are politically or socio-economically marginalized are more likely to face violations of their rights to legal advice and due process as well as disruptions to education, work and healthcare.

Against this background, careful evaluation of immigration detention is urgently needed. We propose to:

1. Map, analyse and evaluate the functioning of the immigration detention system, and trace its effects on vulnerable detainees and their families in Hong Kong.

2. Compare immigration detention policy in Hong Kong to the system in the United Kingdom, to identify best practices and innovation, as well as pitfalls to be avoided.

3. Recommend concrete, evidence-based reforms to immigration detention in Hong Kong and develop operational guidelines to implement beneficial systemic change.

Not only will this be the first detailed study of immigration detention in Hong Kong, it will also add significantly to comparative analysis of immigration detention in East Asia more widely. We will use qualitative, socio-legal methods to conduct our inquiry, analysing official data obtained through applications under the Code on Access to Information; conducting semi-structured interviews with government personnel, civil society organisations, detainees, detainees’ families, and other individuals belonging to vulnerable migrant groups; investigating thematic and individual case studies; and visiting detention centres. Alongside this, we will conduct extensive multidisciplinary secondary research on immigration detention.

Our findings will be actively communicated to core stakeholders as the project progresses. We will conduct workshops disseminating our findings to government officials, civil society organizations in Hong Kong and other parts of Asia, individuals who belong to vulnerable migrant groups, as well as lawyers. Towards the end of the project, we will organize training for government officials based upon our reform proposals. We will also publicize our findings through research briefs, scholarly articles and the media, and integrate them into future advocacy and educational activities.

Project No.: R4037-19
Project Title: Corporate Sustainability Engagement and Corporate Performance: An Examination of Listed Companies in the Greater China Region
Project Coordinator: Prof Carlos Wing-hung LO (CUHK)

Abstract

This project builds on the team’s existing work on business sustainability indexes commenced in 2011 to engage listed companies in Hong Kong in business sustainability through the compilation of the Hong Kong Business Sustainability Index (HKBSI). Since its launch in 2015, the credibility of the HKBSI has been established through four rounds of compilation and announcements. The HKBSI will serve as a benchmark for introducing a CSR rating and certification scheme for the Greater China Region. This knowledge transfer mechanism will provide listed companies with timely and regular assessments of their ESG (environmental, social and governance) and sustainability achievements. To provide a sound empirical basis for establishing and promoting this CSR rating and certification scheme, the project will conduct an in-depth and comprehensive study of listed companies’ ESG and sustainability reports with a specific focus on the effect of CSR activities and performance on corporate performance.

The project objectives are as follows:

1. To use and refine the Value-Process-Impact (VPI) template for compiling and analyzing ESG/Sustainability data at the corporate level.

2. To examine the relevance of CSR in promoting corporate performance.

3. To compile and launch local and regional business sustainability indexes that provide benchmarks for CSR rating and certification.

4. To engage listed companies in the Greater China Region in business sustainability through the launch of local and regional CSR rating and certification schemes.

5. To extend the CSR rating and certification schemes to non-listed companies and SMEs.

This corporate sustainability project focuses on the CSR business case for corporations and society at large. The project investigates the mechanisms linking corporate social responsibility to business sustainability for the benefit of both corporations and society. The project will sample listed companies in the Stock Exchange of Hong Kong Limited, Shanghai Stock Exchange, Shenzhen Stock Exchange, and Taiwan Stock Exchange to create a large CSR database of annual ESG and sustainability reports for performing this investigation. Specifically, the project aims to answer three research questions: (1) Does companies’ CSR and sustainability engagement lead to better corporate performance in the Greater China context? (2) What CSR strategic focus leads to better corporate performance? (3) Are there viable reporting techniques that can improve stakeholder evaluations of firms’ CSR performance?

The research data collected in this project will support the introduction of Greater Bay Area and Greater China business sustainability indexes. Together with the HKBSI, these indices will provide a CSR rating and certification scheme that encompasses both local and regional levels. In the short-term, the project focus will be on establishing local-regional benchmarks, and pilot testing the CSR rating and certification scheme. In the medium-term, the project will focus on promoting corporate subscription to the CSR rating and certification scheme. In the long-term, the project focus will be on internationalizing this scheme as well as extending it to non-listed companies it to non-listed companies and SMEs which are active in the pursuit of CSR and business sustainability. In order to effectively transfer CSR research knowledge to practice, SGS HK Limited will support the marketing and provision of third party data verification for the CSR rating and certification scheme.

Project No.: R5013-19
Project Title: Data Storage and Retrieval using Peptides and Tandem Mass Spectrometry Sequencing
Project Coordinator: Dr YAO Zhongping (PolyU)

Abstract

This is the age of big data. We are generating a lot of data every day and in an increasing rate. These data are typically stored in digital form using electrical, magnetic or optical devices, which require larger physical storage spaces and cannot last for very long time. New data storage methods that are durable and of high density are highly desirable. Storing data using molecules is a solution for this.

In this project, we aim to develop data storage and retrieval using peptides and tandem mass spectrometry (MS/MS) sequencing. This is a new method that has never been reported. To store data in peptides, amino acids are assigned as digital data patterns. Raw data is first encoded as a long string of 0s and 1s, which corresponds to a sequence of amino acids, i.e., peptide, according to the assignment. To retrieve the data, the peptide is sequenced, and the obtained sequence is converted into bits of 0 and 1, which are then decoded as the raw data. The peptides can be commercially synthesized, and MS/MS is the state-of-the-art technique for peptide sequencing. Peptides must not be too long in order to ensure their effective synthesis and sequencing. Therefore, the encoded string is broken into smaller parts, and an address indicator is added into each part to ensure all the parts will be in their original order when they are read back. In this way, a raw data will be stored in a mixture of peptides, which can be separated and sequenced using liquid chromatography coupled with MS/MS (LC-MS/MS). Our preliminary results demonstrated that it was feasible to store data in peptides and retrieve the data using MS/MS sequencing.

Five objectives have been identified for this project: (1) To optimize the sequences of peptides for data storage and retrieval; (2) To develop and optimize protocols for LC-MS/MS analysis of data-bearing peptides; (3) To develop schemes with error-correction capability for encoding digital data into sequences of amino acids; (4) To develop and optimize a software for sequence assignment from MS/MS spectra of data-bearing peptides; (5) To combine the knowledge gained from objectives 1-4 for storage and retrieval of larger real-life data. We have formed a team, comprising of members with expertise in peptide synthesis and sequencing, data storage and retrieval, mass spectrometry, computing and bioinformatics, to tackle various problems for this interdisciplinary project.

The outcomes of this project can generate significant impacts. The implementation of this project will allow development of a new method for data storage, which could be much denser and more durable than the present methods. This project couples data storage with proteomics techniques for the first time and can boost the development of these two important fields.

Project No.: R5029-19
Project Title: Forward-looking intravascular photoacoustic/ultrasound imaging technology for the assessment and guidance of cardiovascular chronic total occlusion intervention
Project Coordinator: Dr LAM Kwok-ho (PolyU)

Abstract

Cardiovascular chronic total occlusion (CTO) is one of the most prevalent coronary heart diseases, which is due to a heavy buildup of cholesterol plaque inside the coronary artery, resulting in the complete blockage of the artery. CTO can cause chest pain, shortness of breath, stroke or even death. One of the low-risk viable options for treatment strategy is to open the blockages using special guide wires and catheters. Nevertheless, the procedure requires very high level of skills and knowledge.

To reduce the procedural time and further enhance the success rate of surgery, an imaging technology being able to assist surgeons to develop optimal treatment strategies before the procedure and provide the real-time therapy guidance during the procedure is highly desired. In this research, a novel imaging technology is proposed for being able to achieve both assessment and guidance of coronary heart disease intervention particularly for the CTO cases. The proposed technology combines the functional imaging offered by the photoacoustic effect and the structural imaging offered by the high-resolution ultrasound. Integrated with the forward-looking design, the technology can offer the real-time visualization of the vessel and plaque morphology as well as the information of plaque characteristics such as constituents and inflammation within the coronary artery.

Project No.: R5043-19
Project Title: Mapping and Characterization for Optimized Evaluation of Potential Landing Sites on the Moon and Mars to Support Future Missions
Project Coordinator: Dr WU Bo (PolyU)

Abstract

Space exploration expands mankind’s understanding of the Earth and the Universe, and promotes human civilization, social and technological progress. The Moon, as our closest neighbor, and Mars, as a once warm and wet planet, have been the focuses of exploration. Identifying a suitable landing site is vital for any in situ exploration missions to the Moon and Mars. Generally, landing site selection must consider both engineering constraints and scientific significance. To meet engineering constraints, topographical features and geomorphological features must be evaluated to identify favorable places for safe landing and surface operations. For scientific significance, the site must be suitable for addressing scientific questions related to geological and environmental processes, with spectroscopic analysis of rocks and minerals on planetary surfaces being the key to answering these questions. This project will develop innovative approaches for mapping and characterizing topographical, geomorphological and geological features, and analyzing these data jointly to enable optimized evaluation of potential landing sites on the Moon and Mars.

The objectives of the project are as follows:

(1) Development of innovative high-resolution, high-precision topographical mapping approaches to support detailed analysis of surface hazards for landing site evaluation on the Moon and Mars.

(2) Development of more automated, robust geomorphological mapping approaches to characterize typical geomorphological features for landing site evaluation on the Moon and Mars.

(3) Development of a multiple-source spectral database of rocks and minerals for spectral interpretation to characterize geological records and mineral compositions for landing site evaluation on the Moon and Mars.

(4) Optimization of joint topographical, geomorphological and geological analysis for systematic evaluation of potential landing sites on the Moon and Mars, to support future exploration missions.

Project No.: R5047-19
Project Title: Improving vision and quality of life in patients with glaucoma using non-invasive brain stimulation and perceptual learning: A Randomized Clinical Trial
Project Coordinator: Dr CHEONG Ming Yan Allen (PolyU)

Abstract

Glaucoma is a complex disease, which is caused by degeneration of the cells that connect the eye to brain, leading to progressive vision loss. It is a neurodegenerative disease, which affects not only the eyes, but also the brain itself, in particular the visual cortex. Despite intense research efforts to restore vision loss in glaucoma, current treatment strategies only arrest or slow vision loss. Progression of the disease is still common. There are no treatments that restore vision lost to glaucoma. Because of the progressive and debilitating loss of vision, in particular peripheral vision, patients with glaucoma experience difficulties with many important daily tasks, such as reading and walking, and they usually experience depression and lower quality of life.

Current vision rehabilitation for peripheral field loss involves compensation strategies such as the use of prisms to move images into central vision or eye movement training to optimize the use of remaining vision. None of these approaches results in improved vision. Recent scientific advances have discovered that neurons and neural circuits in our brain can be rewired and this process is called neuroplasticity. Emerging evidence has demonstrated that vision can be improved by harnessing brain plasticity using visual perceptual learning (i.e. intensive visual training by repeated presenting stimuli at one or more locations in the visual field) and/or neuro-modulation (i.e. alteration of brain activity through non-invasive brain stimulation). These approaches take advantage of residual input from the eye to the brain, which still occurs in many regions of the visual field affected by glaucoma.

“Vision restoration” is a paradigm shift which involves the recovery of lost vision. With this breakthrough paradigm, improving patients’ visual functions through modulating brain function has gained traction in clinical rehabilitation research. While experimentally and conceptually promising, large-scale, randomized controlled trials (RCT) are urgently needed to assess the effectiveness of this new intervention approach. To address this need, we will conduct a research program culminating in a large scale, pivotal RCT of brain function modulation for treating glaucoma.

Project No.: R5060-19
Project Title: Edge Learning: the Enabling Technology for Distributed Big Data Analytics in Cloud-Edge Environment
Project Coordinator: Prof GUO Song (PolyU)

Abstract

Nowadays, the number of edge devices (e.g., phones, tablets, sensors, etc.) is growing extremely fast, and these devices generate an excessive amount of data at a regular basis which constitutes big data. It is very significant to learn from such big data for intelligent applications, e.g., Smart buildings. Traditional cloud-based methods have made great achievements in learning from the big data, especially for those that allow long response delay and data aggregation from edge to cloud. To further extend the frontiers of distributed learning, in this research project, we propose Edge Learning, a paradigm complementary to the cloud-based methods for big data analytics in the cloud-edge environment. It is proposed and developed for moving the training and inference to the edge environment to serve the delay-sensitive and privacy-sensitive applications, of which the data cannot be gathered to the cloud. To tackle the challenges rooted by the inherent characteristics of the cloud-edge environment, we will make efforts on developing new techniques of Edge Learning in this project, including (1) collaborative learning architecture, (2) learning-oriented communication scheme, (3) fault tolerance and resilience strategy, (4) privacy protection and security guarantee, and (5) trained model deployment and update. A highlight of our proposed work is its strong emphasis on practical implementations. We will develop an Edge Learning platform that embeds all technologies above. Our proposed research and implemented platform will enable learning-based applications to be deployed in edge environment in an efficient and secure manner, resulting in the quick deployment of AI-enabled services in smart cities, which has tremendous technological and social significance.

Project No.: R6020-19
Project Title: Domain-Specific Knowledge Base Construction
Project Coordinator: Prof CHEN Lei (HKUST)

Abstract

In this proposal, we aim to build a complete and robust domain-specific knowledge base with the following motivations. First, many online applications (e.g. shopping websites) routinely receive a large number of queries regarding their products or services. At present, a common search technique for items on online shopping websites uses keyword-based queries that return ranked lists of candidate entities. However, keyword-based queries may overlook useful hidden information since they cannot discover potential relationships among these entities. We aim to construct domain-specific knowledge bases that can improve the query results by considering relationships among entities such as online products. Second, domain-specific knowledge bases can also help customers and users better understand the usage, attributes, and properties of entities. If all the real-world entities in a specific domain are properly organized, users can easily find the ones that best suit their needs. This concept can be applied to not only commercial products on online shopping websites, but also other professional domains such as medicines in medical science databases. Third, entity catalogues in domain-specific knowledge bases are also valuable, as they allow the data providers to organize, standardize, and publish information to users. A clear and organized entity taxonomy makes advertisement and recommendation more effective and efficient. To achieve these benefits, we propose a two-stage framework involving entity linkage and entity taxonomy construction, and to construct a domain-specific knowledge base that describes different kinds of entities in a specific domain based on their specifications as well as general knowledge. Using data provided by our partner organizations, we will apply the proposed techniques to implement such a knowledge base to aid intelligent entity search and recommendations.

Project No.: R7003-19
Project Title: Ultra-Light and Ultra-Low-Power MRI Technologies for Accessible Healthcare
Project Coordinator: Prof Ed Xuekui WU (HKU)

Abstract

Magnetic resonance imaging (MRI) has impacted modern healthcare tremendously, as most clinicians recognize it as a valuable medical device innovation. MRI is now a routine diagnostic procedure. Despite the clear advantages and clinical impact of MRI in healthcare, its accessibility is low and extremely inhomogeneous globally. Approximately 70% of the world’s population has little to no access to MRI and the benefits that it entails. We propose to develop ultra-light and ultra-low-power brain ultra-low-field (ULF) MRI technologies for ubiquitous and highly accessible healthcare applications. As a proof of concept, we will develop a set of ultra-low-field (ULF) MRI technologies for brain imaging. In this project, we aim to revolutionize MRI hardware by designing and optimizing a compact brain ULF MRI scanner for low weight, power, and costs; develop essential MRI protocols for clinical brain MRI with relative speed and comfort; and innovate novel data acquisition schemes and image reconstruction algorithms uniquely designed for ULF MRI. The development of such innovative technologies will create an entirely new category of MRI scanners that are compact, low-cost, ultra-light, ultra-low-power, and easy to operate.