Intelligent Robotics for Elderly Assistance in Hong Kong

This study aims to provide an intelligent robotic solution for assisting older adults' mobility and manipulability. Using a novel UC3 development approach, the investigators will systematically study the requirements of elderly users to formulate technical specifications for designing personalized wearable robotic systems. Specifically, we will develop: (1) a novel UC3 development approach to identify the requirements and constraints for the elderly's mobility and manipulability based on gerontechnological and physiological studies; (2) a personalized sensor-driven wearable robot for elderly assistance; (3) an integrated interactive testing system to support the development of intelligent robot-enabled elderly assistance, and to significantly improve performance and safety; and (4) an experimental environment to test, evaluate and demonstrate the developed systems, emphasizing significantly improvement of the mobility and manipulability of older adults.

In order to improve human mobility and manipulability, a wearable robotic device is required to have certain rigidity to support the human body or sustain an external load. Inspired by the flexibility of multiple muscle bundles in biological muscle tissues, the investigators propose to use multiple servomotors to form distributed driving joints of the wearable robot of a hybrid structure and to regulate the driving force and speed of joints by coordinating the output of each motor in real-time. The UC3 approach starts with a psycho-social study and leads to the identification of the critical needs of older adults described by kinesiology-based design parameters. This will provide the guidelines for developing intelligent elderly assistive robots by integrating novel hybrid soft/rigid soft wearable robots, an intelligent sensing system and control methods. The robots will then be tested and evaluated in a user-centric environment to enable the next phase of improvement. Our systems will be developed based on requirements at three levels: (1) the physiology level, involving individual active elements; (2) the function level of action; and (3) the behaviour level of performing tasks.

In order to inform the design of our wearable robotic device, it is essential to understand the needs and compensation requirements of the elderly users in their daily life at physiology, function, and behaviour levels. 60 elderly subjects will be recruited for examination. The user needs assessment will result in a list of categories of "movement, tasks, and behaviour" with the scenarios of "mobility problems in the home and the community" that older people think important and necessary to complete independently daily life. Based on the psycho-social studies and physical and physiological tests done in laboratory testing and evaluation platform, the investigators will conduct needs and compensation assessments quantitatively to formulate technical requirements for system design and development. Furthermore, as the measured data continue to accumulate, the investigators will develop machine learning methods to efficiently estimate reliable compensation schemes based on a small set of measured data on a new user. This will greatly accelerate the efficiency of compensation studies when the number of users increases.

To evaluate our wearable devices for improvement, the investigators will quantitatively assess the enhancement of the mobility and manipulability of the targeted elderly using a pre-and post-implementation experiment. A total of 180 older people from the out-patient clinics and/or elderly community service units will be recruited for user studies in three phases, with 60 people in each phase. Subjects to be recruited in Phase 2 and Phase 3 need to meet the following inclusion and exclusion criteria: (1) aged 65 or older; (2) SARC-F score (measurement of sarcopenia) equal to or over 4; (3) SPPB score less than 11; (4) moderate mobility impairment, i.e., ADL between 6 and 12; (5) capable of clear communication; and (6) no cognitive impairment. The participants will use the robotics system and integrated environment to perform daily self-care to uncover problems in functionality, usability, and acceptance of the robotic device prototype. The system, connected to mobile apps and the Internet, will serve as a platform to measure and record data on how the participants engage in pre-specified activities. The experiment will last six months, and the outcome will be assessed pre-and post-implementation for comparison and validation. The major testing criteria include the design specifications for functionalities and reliability, user experiences, system security and privacy. Participants' perceived ability and confidence to perform daily self-care activities, quality of life, and acceptance of the system will also be measured. The first stage of evaluation will focus on the functionality and reliability of the individual system, measuring the gait stability using walker tipping index, fatigability by handle reaction vector (HRV), and centre of gravity. The second stage will focus on evaluating the system performance and usability of the IAE.

The participation of the older adults recruited from the out-patient clinics and/or elderly community service units in the tests, surveys, and/or interviews will not affect the quality of care they are receiving. No drug usage and medical treatment will be involved in the study. Intervention, surveys, and/or interviews do not impose any physical or medical risk to participants. The only concern is that participants may feel a little tired after the tests. Participants can voluntarily withdraw from the study at any time, without giving any reasons, their medical care or original rights will not be affected.

The investigators will conduct the following analysis to suit the project needs: user analysis, physiological analysis, and robotic system development. Concerning user analysis, chi-square or independent t-tests will be used to examine the differences in the baseline characteristics between older adults with different levels of mobility and manipulability difficulties. Further, the investigators will conduct regression analysis to compare the difference in outcomes between older adults with different levels of mobility and manipulability difficulties. Recruitment rate, drop-out rate and missing data will also be examined and reported. Concerning physiological analysis, the investigators will conduct muscle anatomical analysis, motion analysis, gait analysis, foot pressure analysis, and electromyography analysis. Concerning robotic system development, the investigators will conduct analysis related to sensors development and performance.

The principal investigator will be responsible for keeping of the personal data during and after the study. The data containing personal identifiers will be kept 5 years after the publication of the first paper arising from the research project and the anonymized data will be kept 10 years after the publication of the first paper arising from the project. All information obtained will be used for research purposes only. All the data will be stored in an encrypted workstation and on password-protected online cloud storage.