Projects

Capstone thesis project description

University of Sydney

1.Digital Heart


Title: Development of AI-based Digital Heart

 

This project will develop an AI based digital heart to identify the strengths and weaknesses of the current technology which will be served as a foundation for improvement and development of future AI-enabled ECG in deep-learning convolutional neural networks. The most desirable outcome is a fully automated AI-enabled ECG system which mimics human-like interpretation of ECGs to assist medical practitioners in making greater diagnosis and clinical decisions.

Here you need to develop an APP for Android & IOS system, the basic function is that you can use the APP to connect with the wearable ECG patch and perform some simple data analysis, etc. priority is given to whom familiar with Android studio or Xcode programming.

Similar to the first project, now the platform is multi-platform crossed, specially  co-development for user interface,priority is given to those who have project development experience

Develop AI based computer software which can automatically diagnose heart diseases. The java or C# programming is necessary, priority is given to whom familiar with .net programming

 

The 3D heart geometry including the four cardiac valves and the main vessels need to be built. Finally, virtual heart model coping with all the main features of the cardiovascular function needs to be built with GPU.priority is given to those who have project development experience (source from 1.GPU accelerated digital twins of the human heart open new routes for cardiovascular research | Scientific Reports 2.https://academic.oup.com/europace/article/25/2/469/6825231)

 

Supervisor:  Zihuai Lin and Branka Vucetic.  Email: branka.vucetic@sydney.edu.au; zihuai.lin@sydney.edu.au

 

Requirements: up to 6 students are required for this project. The students participating in this project should have good knowledge on smart phone APP development. Programming skills are essential.


Reference

1. L. Meng, K. Ge, Y. Song, D. Yang, and Z. Lin, Long-term wearable electrocardiogram signal monitoring and analysis based on convolutional neural network, the IEEE Transactions on Instrumentation & Measurement, Vol. 70, April 2021, DOI:10.1109/TIM.2021.3072144

2. Z. Chen, Z. Lin, P. Wang, and M. Ding, Negative-ResNet: noisy ambulatory electrocardiogram signal classification scheme, Neural Computing and Applications, Vol. 33, Issue 14, July 2021, pp. 8857-8869

3. P, Wang, Z. Lin, Z. Chen, X. Yan, M. Ding, A Wearable ECG Monitor for Deep Learning-Based Real-Time Cardiovascular Disease Detection, https://arxiv.org/abs/2201.10083

4. X. Yan, Z. Lin, P. Wang, Wireless Electrocardiograph Monitoring Based on Wavelet Convolutional Neural Network, Proceedings of the IEEE WCNC 2020.

2.Digital Twin


Title: Development of digital twin generative AI

 

This project will invent, train, and deploy AI models that produce accurate, comprehensive forecasts of a patient’s future clinical outcomes, which we call their digital twin, predicting disease progression

 

Develop AI-based computer software that can automatically predict disease progression scores. Java or C# programming skills are necessary; priority is given to those familiar with server programming.

 

Similar to the first project, now the platform is multi-platform crossed, specially  co-development for user interface.

 

Developing a novel way to machine-learning generative AI of the digital twin, priority is given to those familiar with NLP or LLM programming experience.

 

Supervisor:  Zihuai Lin and Branka Vucetic.  Email: zihuai.lin@sydney.edu.au; branka.vucetic@sydney.edu.au

 

Requirements: up to 4 students are required for this project. The students participating in this project should have good knowledge on smart phone APP development. Programming skills are essential.

3.Natural Language Processing


Title: Development of clinical trial recruitment system based on large language model (LLM) generative AI

 

This project will invent, train, and deploy AI models that produce accurate, comprehensive matching feedback of a patient’s clinical trial outcomes—which we call Patient recruitment system for global medicine companies

 

Develop AI based computer software which can automatically output clinical trial matching score. The java or .C# programming is necessary, priority is given to whom familiar with server programming.

 

Similar to the first project, now the platform is multi-platform crossed, specially  co-development for user interface.

 

Developing and deploying LLM generative AI to clinical trial recruitment system, priority is given to whom familiar with NLP or LLM programming experience.

 

Supervisor:  Zihuai Lin and Branka Vucetic.  Email: zihuai.lin@sydney.edu.au; branka.vucetic@sydney.edu.au  

 

Requirements: up to 4 students are required for this project. The students participating in this project should have good knowledge on smart phone APP development. Programming skills are essential.

4.AI - Glass

 

Title: Development of monocle with a suite of AI capabilities

This project will consist of hardware development, firmware development and app development, ideally, the monocle is capable of using AR technology and MicroPython on monocle to connect the real world, at very low cost instead of costly Apple vision pro.

 

Develop server system which can automatically provide service to hardware. The python programming is necessary, priority is given to whom familiar with server programming.

 

Choosing project components at reasonable low costs, drawing PCB schematics and making PCB fabrication, priority is given to whom familiar with hardware developing experience.

 

Developing Bluetooth connection with hardware , priority is given to whom familiar with firmware coding experience.

 

Supervisor:  Zihuai Lin and Branka Vucetic.  Email: zihuai.lin@sydney.edu.au; branka.vucetic@sydney.edu.au  

 

Requirements: up to 4 students are required for this project. The students participating in this project should have good knowledge on smart phone APP development. Programming skills are essential.

5.Healthcare IoT --- Pressure Injury & NLP

 

Title 1: Internet of Things (IoT) in Smart Hospitals for Pressure Injury Monitoring

This project will develop novel solutions for IoT-based pressure injury monitoring to enable smart hospitals, smart healthcare and provide in-patients with good treatment experience. IoT is a key enabler for our future smart hospital and healthcare to effectively overcome the major problems, such as nursing staff shortage, impractical physical care environments and difficulties in identifying patients’ needs, etc.  In order to enable a fast uptake of the IoT pressure injury monitoring systems, key issues, such as data collection and sensing, pressure injury prediction modelling,  and hospital validation, should be addressed. These problems are the major technological obstacles which are preventing industrial partners from further expanding their businesses in the hospital IoT area.

1.    A hospital-specific pressure injury IoT platform: We will develop a hospital IoT with a large-scale hybrid cloud system (combining Edge and Cloud computing) to meet the QoS requirements of the provided services and addressing the unexpected transmission latency caused by Internet connections.

2.      Smart mat: We will develop a smart mat for pressure injury monitoring. The proposed smart mat consisting of a pressure sensor array and a moisture sensor array will be tested and validated in the Westmead Hospital. The collected data will be transmitted to the cloud via WiFi or other wireless networks for data processing and analysis. Our developed deep learning algorithms for smart mats will be used to obtain the weight and moisture distribution of the inpatient’s body as well as the respiratory rate.

3.       Pressure injury prediction: Based on the data collected in steps 1 - 2 and the PI prediction model obtained in Task 3, we will predict pressure injury  development in a real scenario. The results from different sensing techniques will be compared with those obtained using Waterlow scale. The comparison will provide insight to further improve the proposed algorithms and designs. 

4.      Hospital validation: Our developed pressure injury IoT system will be validated at Westmead Hospital.

Supervisor: Zihuai Lin and Branka Vucetic.  Email: zihuai.lin@sydney.edu.au; branka.vucetic@sydney.edu.au

 

Requirements: up to 2 students are required for this project. The students participating in this project should have good knowledge on wireless communications, RF engineering, wireless local area networks and wireless sensor networks. Matlab C++ and Python programming skills are essential.

 

Title 2: Hospital-Acquired Pressure Injury Clinical Large Language Model for Downstream Clinical Information Extraction

 

The project focuses on developing specialised Large Language Models (LLMs) for efficiently extracting Hospital-Acquired Pressure Injuries (HAPI) information from clinical notes. The students are required to address the challenge of scarce labelled data and privacy concerns in healthcare by creating a synthetic dataset of clinical texts, which served as a foundation for training and evaluating four advanced NLP models: LLaMA Base, Alpaca, MedAlpaca, and Asclepius. A significant part of the thesis project lies in the comprehensive analysis and benchmarking of the models against real-world data, setting a new standard in the field of large language models in clinical settings. The students must work closely with Westmead Local Health District clinicians, ensuring that the synthetic notes accurately represent clinical scenarios and fine-tuning the project's output with practical insights.

 

Supervisor:  Zihuai Lin and Branka Vucetic.  Email: zihuai.lin@sydney.edu.au; branka.vucetic@sydney.edu.au

 

Requirements: up to 2 students are required for this project. The students participating in this project should have good knowledge of NLP and AI. Programming skills are essential.

6.WiFi Sensing and Positioning


Title 1: Development of hand-held devices based on UWB, mmWave, THz or WiFi signals to see through walls for human activity monitoring

In this project, we aim to develop a low cost X-ray vision wireless hand-held device. The developed hand-held device can be used to see through walls to track moving human bodies. The technique would be based on a concept similar to radar and sonar imaging, instead of using high power signal, this one would use low power Wi-Fi or mmWave/THz, UWB signals to track the movement of people behind walls and closed doors.  When a RF signal is transmitted towards a wall, due to the absorbing property of the walls, only a small part of the signal can be penetrated through the wall and can be reflected back when the signal reaches any objects that happen to be moving around in the other room. Based on the reflected signal, we can detect the moving objects.

Supervisor: Zihuai Lin and Branka Vucetic.  Email: zihuai.lin@sydney.edu.au; branka.vucetic@sydney.edu.au

 

Requirements: up to 2 students are required for this project. The students participating in this project should have good knowledge on wireless communications, RF engineering, wireless local area networks and wireless sensor networks. Matlab C++ and Python programming skills are essential.

 

Title 2: Precise Indoor Positioning based on moving Robots and UWB/mmWave/THz/WiFi techniques

 

Recently, indoor positioning is attracting considerable attention from both research and industry. Logistics, health-care applications, search and rescue, military services, tracking of objects and people, gaming and entertainment are a few examples of applications which can benefit from having precise localization information. However, in indoor environments, traditional services provided by e.g. GPS usually are not available, unreliable or inaccurate. For this reason, alternative solutions need to be developed. In this project, we aim to using Ultra-Wideband (UWB)/mmWave/THz/WiFi techniques and/or moving robots to realize precise indoor positioning.

 

Supervisor: Zihuai Lin and Branka Vucetic.  Email: zihuai.lin@sydney.edu.au; branka.vucetic@sydney.edu.au

 

Requirements: up to 3 students are required for this project. The students participating in this project should have good knowledge on wireless sensor networks, communication theory and signal processing. Matlab, C++, Python programming skills are essential.

Reference


Title 3: Augmented Reality (AR) mobile APP development for wayfinding

 

This project is to develop an Internet of Things network and platform architecture suitable for the Westmead Precinct consisting of location sensors, people counting sensors and an interactive way-finding app e.g. an augmented reality (AR) mobile app. The design of the proposed platform and data processing architecture will aim to future-proof IoT network capabilities to allow more connected devices including environmental sensors to be incorporated such as temperature, humidity, and air quality. The project is based on a multi-sensor data fusion artificial intelligence algorithm, which can geolocate compatible smartphones inside buildings. This algorithm achieves higher accuracy by leveraging pre-existing information of the environment (Bluetooth, WiFi…) combined with sensors that allow inferring the movement of the user (compass, gyroscope, accelerometer, barometer…). The information can be fed into the indoor wayfinding and navigation solution to guide hospital visitors to always find the most suitable route to their destination based on their stated preference.

 

Supervisors:  Zihuai Lin and Branka Vucetic.  Email: zihuai.lin@sydney.edu.au; branka.vucetic@sydney.edu.au

 

Requirements: Up to 2 students are required for this project. The students participating in this project should have good knowledge of smartphone APP development, wireless communications, and indoor positioning. Programming skills are essential.

7.RADAR Sensing

 

Title: Development of hand-held devices based on UWB, mmWave, THz or WiFi signals to see-through walls for human activity monitoring

In this project, we aim to develop a low-cost X-ray vision wireless hand-held device. The developed hand-held device can be used to see through walls to track moving human bodies. The technique would be based on a concept similar to radar and sonar imaging, instead of using high power signal, this one would use low power Wi-Fi or mmWave/THz, UWB signals to track the movement of people behind walls and closed doors. When an RF signal is transmitted towards a wall, due to the absorbing property of the walls, only a small part of the signal can penetrate through the wall and can be reflected back when the signal reaches any objects that happen to be moving around in the other room. Based on the reflected signal, we can detect the moving objects.

Supervisor: Zihuai Lin. Email: zihuai.lin@sydney.edu.au

Requirements: Up to 2 students are required for this project. The students participating in this project should have good knowledge of wireless communications, RF engineering, wireless local area networks and wireless sensor networks. Matlab C++ and Python programming skills are essential.

Reference

1.                 X. Wang and Z. Lin, “Nonrandom microwave ghost imaging,” IEEE Transactions on Geoscience and Remote Sensing, vol. 56, no. 8,pp. 4747–4764, 2018.

2.                 X. Wang and Z. Lin, “Microwave Surveillance based on Ghost Imaging and Distributed Antennas” IEEE Antennas and Wireless Propagation Letters, Volume: 15, March 2016. pp. 1831 - 1834

3.                 R. Luo, Z. Zhang, X. Wang, Z. Lin, “Wi-Fi Based Device-free Microwave Ghost Imaging Indoor Surveillance System” accepted at 2018 28th International Telecommunication Networks and Applications Conference (ITNAC).

4.                 Z. Zhang, R. Luo, X. Wang, and Z. Lin, “Microwave Ghost Imaging via LTE-DL Signals, accepted by International Conference on Radar 2018 August 2018, Brisbane

8.Communication Systems

 

Title: Reflective Intelligence System (RIS) for 6G THz Cellular Networks

 

With explosive growth of wireless networks and the Internet, network resource utilization becomes one of the critical design issues. RIS has emerged as a promising tool for the design of next generation communication networks. There is a need to develop systematic design of RIS THz array techniques for real wireless energy efficient networks. In this project, we plan to systematically design RIS channel estimation, beamforming mechanisms for cellular networks and related signal processing algorithms, as well as demonstrate the feasibility and benefits of integrating the developed RIS schemes into practical systems. The research outcomes are likely to result in significant improvements in network performance

 

Supervisor: Zihuai Lin and Branka Vucetic.  Email: zihuai.lin@sydney.edu.au; branka.vucetic@sydney.edu.au

 

Requirements: up to 3 students are required for this project. The students participating in this project should have good knowledge on wireless cellular networks, communication theory and signal processing. Matlab,  C++ and Python programming skills are essential. The students with average marks above 75 are preferred.

 

Reference

      2. Y. Hu, P. Wang, Z. Lin, M. Ding, "Performance Analysis of Reconfigurable Intelligent Surface Assisted Wireless System with Low-Density Parity-Check Code," accepted  by IEEE Communications Letters, June, 2021.