3 Investigating the machine learning approaches to de- termine differences between postural and visual be- haviour of healthy adults and adults with visually in- duced dizziness.
Organisation: NZ Dizziness and Balance Centre/AUT
Project start date: Semester 1 2022: 28 February – 23 June Number of students: Individual or Team
Student type: Full-time student (1 semester, 15 weeks)
Copyright By PowCoder代写 加微信 powcoder
Introduction
Following a vestibular disorder, some people experience ongoing symptoms of visually induced dizziness (Bronstein et al., 2020). This limits people’s ability to navigate through public places, restricting participation in daily activities and reducing the quality of life (Benecke et al., 2013). It is a commonly reported symptom in people with chronic dizziness and is characterised by nausea, dizziness, and imbalance in complex visual surroundings ( Chin, 2018). Lack of a clear understanding of its aetiology makes treating visually induced dizziness difficult. Patients with visually induced dizziness often feel anxious and frustrated at their lack of symptomatic improvement. A comprehensive body of work was conducted to explore visually induced dizziness in a range of visually provocative environments (Chaudhary et al., 2020). The exploratory study exposed forty adults (20 healthy and 20 adults with visually induced dizziness) to a range of visually complex backgrounds. Visual fixations, the centre of pressure, body and head kinematics were recorded via a mobile eye tracker, AMTI force plate and Qualisys 3D motion capture system, respectively. The study identified that visually- induced dizziness adults exhibit fixational instability, with increased postural and head sway as visual complexity increases. A detailed study protocol has been published (Chaudhary et al., 2020).
Description
The student will conduct an exploratory analysis of the human motion and eye-tracking data to determine differences between postural and visual behaviour of healthy adults and adults with visually induced dizziness associated with complex visual environments. This includes exploring body kinematic data, centre of pressure data and visual fixations data recorded with an eye tracker.
Objectives
Objective 1. To get acquainted with literature concerning this project.
Objective 2. To run descriptive statistics to understand potential associations of interest on baseline data (task 1-neutral condition) from a) adults with visually induced dizziness and b) healthy adults.
Objective 3. To determine unique patterns/relationships which are associated with altered
postural (force plate data) and visual behaviour (eye-tracker data) in adults with visually induced dizziness and with healthy adults when exposed to complex visual environments by utilising the unsupervised machine learning approaches across the whole trial and in all tasks. Objective 4. Investigating the temporal evolution of the unique patterns/relationship deter- mined in objective 3 l over time in each task and across six tasks between healthy adults and Adults with visually induced dizziness Using advanced modelling techniques.
Data available: Yes
Type of project data: a. Kinematics: 3D/6DOF. 6 Degree of Freedom (pitch, roll, yaw, x, y, z) data from user-defined rigid bodies. The 6DOF data gives information about the position and rotation of a moving body. QTM can save 6DOF data and send 6DOF data over TCP/UDP in real-time raw format.
b. Centre of Pressure: Raw COP displacement data (1200 Hz) can be accessed from Qualisys. c. Eye-tracking: Raw data in pixels can be extracted from SMIBE gaze eye-tracking software.
Constraints
Where is the company expecting the student(s) to work?
A combination of both (on–site + remotely – 20/80).
Is a ‘My Vaccine Pass’ required?
Is a non–disclosure agreement required?
Yes (to be signed in Week 0, between 21 – 25 February 2022).
程序代写 CS代考 加微信: powcoder QQ: 1823890830 Email: powcoder@163.com