The Smart Healthcare Centre is a multi-functional and digitalised platform with an aim of helping elder to have a smarter living area with preventive healthcare by leveraging emerging technologies, for example, wristband, bed sensor, physical training system, healthcare app and devices, interactive games and etc. Wristband of elder equips with indoor location tracking function which is able to detect elder’s duration of stay in different rooms (such as staying too long in washroom and alert the front desk), The wristband is also a smart key for their personal locker. Moreover, bed sensor is an non-contact equipment for measuring elderly’s heart and respiratory rate for preventive healthcare. At the same time, it is able to detect presence of elderly on bed. Caregivers can use mobile, and view those collected data to further enhance patient care and family support effectively.
eForm is an electronic form system which is easy to create forms by end users and easy to fill in various platforms. It provides web based form editor, with templates for faster form creation. The form can be used in iOS, Android, Windows, Mac, Webchat and browsers. There is criteria customisation in eForm for the form validating criteria, shorten turnaround time between fillers and reviewers, minimise human errors, secure environment (can be dedicated server and even intranet) and even know some form are randomly or unintentionally filled (such as unreasonable 1 second for 1 question).
YouGet is a platform for the community to help each other. With the idea of sharing economy, YouGet utilizes various skills and free time of different individuals in the society, enabling them to contribute back into their neighborhood. Each user on YouGet can be a service receiver, as well as a service provider at the same time. Our electronic system will swiftly match a service provider with a service receiver by effective classification of service time, service location and providers’ skillsets, resulting in a convenient, systematic and interactive user experience. We also provide data analytics such as monthly trend of services requested and distribution of participants by trial centres for better management strategies.
ASTRI’s Smart Wealth Management Platform is an online wealth management service platform that provides automated and algorithm-based portfolio management advice without the use of human financial planners, which can get client’s risk profile and risk tolerance level based on KYC (know your customer) analysis, can get optimised asset allocation and portfolio weights based on application of modern portfolio theory, stochastic simulated data analysis, machine learning, etc., and can provide excellent user experience based on IT technology, data visualisation techniques, and gamification techniques developed from finance and banking industry. The front page of the platform is illustrated in Figure 1 as follows.
The front Page of SWMP
The following figure 2 is the overall system architecture of the “Smart Wealth Management Platform” system, which is composed with three levels sub-platforms with various functionalities. From Bottom to up they are:
Cloud Platform (Bottom Level)
Input: Financial market data and trading data with multiple time frames (minutely, daily) from three exchanges (HKEX, SSE, SZSE).
Methodology: Complex Event Process (CEP) and Distributed Stream Computing engine.
Output: Financial market data with multiple time frames and asset-specific events.
Optimisation and Market Data Platform (Middle Level)
Input: Financial market data with multiple time frames.
Methodology: Fundamental analysis, Technical analysis, Probabilistic and Statistical analysis, Optimisation algorithm.
Output: Structured Market Information Database and Optimisation engine.
Quant Platform (Top Level)
Input: Structured Market Information Database and Optimisation engine.
Methodology: Modern finance theory, Financial engineering models, Behavioral finance, and AI (Machine learning) and Big data analytics.
Output: Module of financial big data analytics, Module of quantitative trading strategy, Module of portfolio optimisation, and Module of KYC.
In light of catastrophic traffic accidents that has shocked Hong Kong in the past year, this technology provides the means for public transportation to improve the safety. By integrated analysis of the vehicle’s “black-box” data, as well as the video recording of the surrounding scenarios and circumstances, it helps the transportation operator identify high-risk drivers for continuing professional development. The technology is also meant to help an organization with a fleet of vehicles and a team of drivers to manage.
Integrated Data and Video Analytics
On this platform, both data and video analytics are used for driver evaluation. Based on machine learning, bad driving habits are learned from historic operation data of the vehicles and labelled, and used as reference models to classify newly collected data of current drivers. At the same time, the surrounding video captured can provide confirmations for anomalies analyzed or reveal additional potentially risky events.
ASTRI’s work in smart mobility is about applying artificial intelligence, big-data analytics, video analytics and graph theory to solve traffic problems. They can be applied to multiple application domains, such as robotics and transportation.
As the autonomous vehicle industry racing from zero to warp speed, it calls for technology innovations in every aspect of this field. For automated guided vehicle (AGV), one type of autonomous vehicles, its global market alone is projected to be US$2.68B by 2022.
Automated Guided Vehicles
An intelligent platform for scheduling and routing for AGV is developed to address the technical challenges in this area. In an environment of hundreds or even thousands of vehicles, the technical challenges include looking for the most efficient routes for all vehicles and the optimising the assignment of vehicles to perform their tasks. Issues of scalability, avoiding traffic jams, detecting obstacles ahead etc. must be addressed. In this project, deep learning is incorporated to improve on dynamic route optimization, large-scale vehicle planning and task scheduling. ASTRI’s initial application will be to AGVs for distribution centers.
The ASTRI’s smart water IoT system, an ICT award winner in 2017, allows operators to monitor water system in real time. The system features a smart valve with built-in sensors to transmit critical data (e.g. pressure, flow rate) that helps to ensure efficient leakage management. Using both wired and wireless detectors, the system facilities data analysis, and allows timely access to water quality information. The water leakage alert systems, along with the leakage detection hardware, use Complex Event Processing to improve service providers’ operational efficiency. By installing a mobile app, users can check and settle payment accordingly.
ASTRI’S Hierarchical Distributed Ledger Technology Platform (HDLT) features a chain router for cross-chain communication. The solution addresses multiple blockchain challenges such as hierarchical consensus algorithm and speeds up processing of private transactions in permissioned based blockchains. A feature diagram is shown as followed:
ASTRI’s Distributed Ledger Technology (DLT) technologies have been proven in many industrial cases including mortgage alert, KYC etc. Please also refer to the White Papers from the Hong Kong Monetary Authority.
1. HKMA – Whitepaper on Distributed Ledger Technology 1.0
Natural language processing is an important direction in the field of computer science and artificial intelligence. ASTRI has established the core competence and technologies on NLP and applied these technologies on various applications.
Technology Highlights
I. Artificial Intelligence Chatbot for Cantonese/English Mixed Languages
There are many chatbots in the market already. The major differences in ASTRI Chatbot are:-
ASTRI chatbots can handle a mix of Cantonese and English voice input. The unique cultural practice to speak in Cantonese embedded with some English words is also prevalent in the Big Bay Area. ASTRI chatbots can also recognise speech with mixed Cantonese and English or a slight accent.
Chinese have words that share the same sound, and words that have similar forms. This could cause a lot of confusion in the context. ASTRI’s natural language processor can correct these common mistakes based on the context of the sentences.
Usually chatbots obtain the answers for questions based solely on keywords. ASTRI chatbots extract the meaning from the full sentences in addition to the keywords discovered. By doing so, sentences with negation or ambiguity can be managed.
ASTRI chatbots can learn new phrases and new answers through a learning process.
Chatbots have wide applications, ranging from intelligent home, smart city to AI telephone customer service. It can also conduct categorization on the questions raised in emails. ASTRI is working with financial, insurance, banking, and retail industry in the development of chatbots.
Chatbot for general banking inquiry
II. Multi-lingual address recognition for smart phone
ASTRI team has successfully developed an address recognition system that worked with and implemented the regular expressions for English, French, German, Italian, Spanish and Portuguese languages in short messages on mobile phones. By using this technology, the mobile phones are enabled to recognise addresses inside the short messages received. The accuracy and recall rate are all above 85%, verified on hundreds of thousands annotated sentences in each language. Based on this, user can easily tab the address and search on maps.
Specialises in behavior analysis and modeling, applies for smart healthcare, elderly, document management, intelligent data management system, smart eForm platform etc.
Intelligent Companion for Elderly (ICE), a new therapeutic robot provides intelligent user interaction to elderly through various sensors and actuators, which links with caregiver’s App and web portal. Together with the Intelligent Data Management System [IDMS], it simplifies the development of systems involving iOS/Android/Web App, cloud/server software, data visualisation and machine learning modules. ICS, powered by “Elderly companion engine”, is a scalable and extensible platform to support different services for elderly care.
Elderly companion engine
Smart Healthcare System also includes health condition tracking device for elderly home and healthcare wristbands. The team installs smart gateways in the elderly home or centre with NFC and Bluetooth wristbands, it provides the elderly location tracking function and shown on the smart location map. Besides, bed sensors and healthcare apps are also used to detect elderlies’ vital signs. Elderly’s presence, heart and respiratory rate can be detected by non-contact measurement. Thus, accidents can be detected, and alerts would be sent to caregivers.
Healthcare wristbands are used to detect fall and make further analysis on fall incidents. Elderly wears the wristbands which detects the fall. It spots out elderly with high fall risk in order to minimise the risk of serious and fatal accidents.
Healthcare Centre
A Community Support Platform is an intelligent platform named “YouGet”. It makes use of share economy idea which allows the release of individual resources in community to help elderly for their daily needs such as cleaning and accompanying elderly for clinic. With the trusted members of a local community or organisation, the support from each other can be closely linked with minimise travelling or overhead.
In addition, operational effort to manage those community support program can be minimised with the aid of technology. Meanwhile, information collected throughout the process can help social service providers better understand how and what services are needed in the community. Thus, they can plan a better development strategy to enhance their works.
