Auckland Bioengineering Institute

Project

Motor impairment is a common symptom after stroke and recovery of motor function is important for regaining the patient’s independence in activities of daily living. Being able to predict motion recovery and outcomes soon after stroke could support clinicians, patients and families to set proper goals for rehabilitation and appropriate plans of time and resource allocation. In this project, multimodal medical imaging data such as MRI and/or CT will be used to predict the recovery of upper limb function after stroke. The deep learning algorithms will be developed to predict the post stroke recovery based on the quantitative imaging biomarkers derived from the multimodal medical images.

Desired skills: Ideal candidates will have an interest in image processing, and basic programming skills in Matlab, C++, or Python. Experience in machine/deep learning and/or medical image analysis will be beneficial.

Project

This project aims to develop an advanced deep learning-based approach for accurately segmenting brain lesions from MRI/CT scans. By leveraging state-of-the-art techniques such as convolutional neural networks (CNNs) or attention-based models, we aim to improve the accuracy and efficiency of brain lesion segmentation compared to traditional methods. The project will involve dataset acquisition, preprocessing, model architecture design, training, evaluation, and clinical integration. The successful implementation of this project will have a significant impact on healthcare, enabling early detection, precise diagnosis, and improved treatment planning for neurological conditions.

Desired skills: Ideal candidates will have an interest in image processing, and basic programming skills in Matlab, C++, or Python. Experience in machine/deep learning and/or medical image analysis will be beneficial.

Project

This project aims to investigate brain connectivity patterns using functional Magnetic Resonance Imaging (fMRI) and Diffusion Tensor Imaging (DTI). By analyzing the functional and structural connections within the human brain, we aim to gain a deeper understanding of the underlying neural networks and their relationship to cognitive processes and neurological disorders. Through the integration of advanced data analysis techniques, including graph theory and machine learning, we seek to unravel the complexities of brain connectivity and contribute to advancements in neuroscience research and clinical applications.

Desired skills: Ideal candidates will have an interest in image processing, and basic programming skills in Matlab, C++, or Python. Experience in machine/deep learning and/or medical image analysis will be beneficial.

Project

This project aims to develop a deep learning-based approach for MRI harmonization, addressing the challenges associated with variability across different MRI scanners and acquisition protocols. By leveraging advanced deep learning techniques, such as generative adversarial networks (GANs) or domain adaptation methods, we aim to standardize and harmonize MRI images, enabling improved comparability and robustness in clinical and research settings. The project will involve data collection, model development, training, and evaluation using diverse multi-site MRI datasets, with the ultimate goal of facilitating reliable and consistent interpretation of MRI scans and enhancing the accuracy and generalizability of neuroimaging studies.

Desired skills: Ideal candidates will have an interest in image processing, and basic programming skills in Matlab, C++, or Python. Experience in machine/deep learning and/or medical image analysis will be beneficial.

Project

Post-stroke recovery is a complex and heterogeneous process influenced by various factors, including the location and extent of brain damage. Accurately predicting recovery outcomes plays a crucial role in informing rehabilitation strategies and optimizing patient care. This project aims to create a comprehensive brain atlas by integrating multimodal neuroimaging data and clinical information, with the objective of improving the prediction of post-stroke recovery. By leveraging advanced machine learning techniques, we will extract meaningful features from neuroimaging data, such as structural, functional, and connectivity measures, and combine them with clinical variables. The developed brain atlas will serve as a valuable tool for clinicians.

Desired skills: Ideal candidates will have an interest in image processing, and basic programming skills in Matlab, C++, or Python. Experience in machine/deep learning and/or medical image analysis will be beneficial.

The required skills for the project include proficiency in electromagnetic theory, expertise in using the COMSOL simulation platform for finite element analysis, and knowledge of biomedical engineering and neuroscience principles related to neurostimulation techniques. Strong problem-solving abilities, data analysis skills, and effective communication are also essential for optimizing the design of the multi-target temporal interference magnetic stimulation helmet with array coils.

Project

This project aims to develop a multi-target temporal interference magnetic stimulation helmet with array coils, utilizing the simulation platform COMSOL. Through the electromagnetic field module of COMSOL, a comprehensive finite element analysis framework, the project seeks to optimize the helmet design by evaluating its mechanical integrity, electromagnetic characteristics, acoustic effects, and heat transfer properties. By iteratively refining the design based on simulation results and experimental validation, the goal is to create a safe and effective helmet for precise and targeted magnetic stimulation, with potential applications in neuroscience research and therapeutic interventions.

The required skills for the project include proficiency in electromagnetic theory, expertise in using the COMSOL simulation platform for finite element analysis, and knowledge of biomedical engineering and neuroscience principles related to neurostimulation techniques. Strong problem-solving abilities, data analysis skills, and effective communication are also essential for optimizing the design of the multi-target temporal interference magnetic stimulation helmet with array coils.

Project

We are offering an exciting internship opportunity for a duration of four- five months (480 hours), focusing on game development. The project entails creating a game similar to Whac-A-Mole (see link: https://en.wikipedia.org/wiki/Whac-A-Mole), targeted young children for vision screening. The game will feature a background with pulsating dots and lovable rabbit/sable characters for children entertainment.

Required Skills:
Familiar in C++ or C#, HTML and JavaScript programming languages.
Experience with Unity or Unreal Engine game development frameworks.

Project

Head-mounted miniaturised microscopes (miniscopes) enable brain activity to be recorded in freely moving rodents. By using miniscopes, we can directly decipher how brain cell activity underlies behaviour as it happens in the awake behaving animal. This project aims to use deep learning techniques to decode brain activity patterns while studying the underlying neuronal connectivity using an inverse modelling approach. We are examining the cellular mechanisms that underlie behavioural deficits in mouse models of Autism Spectrum Disorders (ASD) and Alzheimer’s disease (AD). We utilise miniscopes to examine cellular activity in the hippocampus during spatial memory tasks. The hippocampus contains place cells which display location-specific firing to encode spatial environments and are important for spatial memory. With the developed methodologies of this project, we will be able to decode mouse position and velocity. We predict that deficits in decoding will correlate with spatial memory impairments.

Requirements: We are looking for highly motivated students with a physics, engineering, or mathematics background, prior programming experience (ideally Python) and a strong interest in microscopy, image analysis, deep learning and neuroscience.

Skills taught:
• Modelling techniques for neurodynamics;
• Integration of custom written software with open-source software.

Project

This project aims to develop a software platform to simulate brain activity for training, educational and research purposes. In terms of research, we are examining the cellular mechanisms that underlie behavioural deficits in Alzheimer’s disease (AD). We will use photometry datasets describing brain activity in multiple parts of the brain to calibrate our neurodynamic models in AD and healthy scenarios. Then, we will develop an interactive interface to manipulate and influence the neurodynamic simulation to understand the effect of different interventions.

Requirements: We are looking for highly motivated students with a physics, engineering, or mathematics background, prior programming experience (ideally Python) and a strong interest in photometry, image analysis, computational modelling and neuroscience.

Skills taught:
• Modelling techniques for neurodynamics;
• Integration of custom-written software with open-source software;
• Image processing techniques.

Project

The dairy sector currently contributes over 10 billion New Zealand dollars. NZ operates mainly on pasture-based farming, with large herds and large-scale processing facilities. Improving operational efficiency and animal welfare requires automated, non-invasive techniques that can remotely observe cows in the farm.

Raising global temperatures increase the risk of heat stress in cows; which has significant negative economic and welfare impacts. This project will focus on implementing machine learning methods to detect heat stress on portable processors (Jetson nano), coupling them with machine vision cameras and testing them in the field.

Suited for candidates with interest in image processing, machine learning and animal behaviour.

Project

The uterus is a smooth muscle organ which undergoes contractions of various magnitudes and frequencies at different points in time, depending on the stage of the menstrual cycle or during pregnancy. Conditions which can arise when contractions are poorly initiated or uncoordinated include endometriosis, extreme cramps, and preterm labour. Determining how contraction occurs and how it changes can help to understand what is occurring in these situations

For this summer project, you will be involved with developing and improving new experimental approaches for making measurements from uterus muscle. You will be working on constructing and improving design of ex vivo tissue measurement systems. This will involve working within our bioengineering workshop, as well as exposure to experimental design and participation in tissue experiments which will aim to measure one or more of electrophysiology, force, or calcium in living tissue.

Project

Cardiovascular disease (CVD) affects millions worldwide and is the leading cause of mortality. Devices that can efficiently and non-invasively provide early and clinically useful diagnostic information may improve patient quality of life and help reduce CVD morbidity. Our group has created a camera-based imaging system that estimates the carotid artery and jugular venous pressure (JVP) waveform by measuring the deformation of the skin on the neck due to the vessel pulsations.

This summer studentship project will focus on two key developments for the current system. Firstly, developing and testing the use of near-infrared (NIR) imaging to see deeper into the neck. This will enhance the detection of blood vessels and potentially allow vessels to be distinguished based on the wavelength-dependent absorption of oxygenated and deoxygenated blood. Secondly, evaluate the use of this imaging technique with the high-speed video recording capabilities of recent model mobile phones. Aspects of mobile phone imaging and methods to extract locations of maximal pulsation will be explored. Overall these advancements will enable the translation of this research toward a home-based healthcare tool.

Project

The endometrium is a highly dynamic tissue that lines the uterus and sheds each month as part of the menstrual cycle. The outer layer of the endometrium is an epithelial layer consisting of polarized epithelial cells, similar to other organs. Cell polarity (asymmetry in structure and function) is critical for healthy epithelial tissue function, and a loss of epithelial cell polarity is known to play a significant role in tumour progression.

Unlike other epithelial layers, the endometrial epithelium exhibits significant changes in epithelial cell behaviour over the menstrual cycle. One such change is a temporary loss of polarity during the ‘window of implantation’. This loss of polarity is necessary to facilitate implantation of an embryo, however, changes to polarity could also contribute to the progression of uterine diseases, such as endometriosis.

In this project the student will explore potential mathematical models for cell polarity at both subcellular and cellular scales. The goal of the project will be to incorporate a mechanistic model(s) of cell polarity into a multicellular model and simulate an endometrial epithelial layer with dynamic cell polarity.

Required background: mathematics, engineering, or physics.
Desired skills: numerical solutions of differential equations (e.g. ENGSCI 111/211, MATHS 260 or 270), programming

Project

Disorders of gut-brain interaction (DGBI), such as irritable bowel syndrome, are a rising concern, yet their underlying mechanisms remain poorly defined. At least 30% of the population has a DGBI, many of whom have no detectable disease but suffer from abnormal digestive processes.

This summer research project would contribute to a nationwide effort to understand the link between gastrointestinal symptoms, metabolism, physiology, and the microbiome to predict food-health relationships. Thanks to our esteemed national collaborators, we have access to a comprehensive data set consisting of food and symptom diaries, microbiome, and metabolomics information from healthy and DGBI patients. Preliminary analyses of these data have revealed distinct variations at the microbiome and metabolic levels. We now aim to use computational techniques to explore the possible functional implications of these differences.

The ideal candidate will be interested in progressing modern healthcare through computational modelling (experience not necessary). You will join a multidisciplinary team of bioengineers, clinicians, and food scientists, so an ability to work both independently and as part of a wider team is beneficial.

If you want to modernise the clinical approach toward a modern health mystery, please contact Jarrah Dowrick or Tim Angeli-Gordon to learn more.

Project

Heart muscle tissue can be dissected and studied in an experimental device to provide a better understanding of the underlying physiology of the heart. However, typical measurement devices allow only one piece of muscle to be studied at a time. We have developed a new instrument that allows the study of multiple muscles isolated from the same heart. Each muscle is placed in a separate well and ‘printed’ into place using light-curable hydrogel. Each muscle can independently be stretched and mechanically or chemically manipulated for experimental purposes.

We’d like to study muscles over multiple days but first need to understand whether the hydrogel changes significantly during this period. In this project, you will work with a team of bioengineers and physiologists to (i) characterise the change of gel stiffness under such conditions, and (ii) optimise a protocol for hydrogel preparation that minimises the deformation of the hydrogel, or maintains its properties.

This project suits a candidate with an interest in instrumentation, hydrogel bio-fabrication and muscle experimentation.

Project

Heart muscle tissue can be dissected and studied in an experimental device to provide a better understanding of the underlying physiology of the heart. However, typical measurement devices allow only one piece of muscle to be studied at a time. We have developed a new instrument that allows the study of up to 8 muscles isolated from the same heart, potentially for several days. Each muscle can independently be stretched and mechanically or chemically manipulated for experimental purposes.

In this project, you will design and construct a new flow-through fluid system for our new instrument. The system will provide saline fluid to each muscle in each well, thereby providing nutrients and oxygen to maintain cardiac muscle performance for multi-day experiments.

This project will be suitable for students with an interest in bioinstrumentation design, electronics and embedded control, and biomedical experimentation.

Project

This project will involve pre-clinical experimental research to assess changes in the structural properties of Type 2 diabetic hearts. The successful candidate will utilize high-resolution confocal laser scanning imaging to characterise cellular structures. Specifically, the structures of mitochondria (using Tom20 dye), cell membrane (using Wheat germ agglutinin), and cell contractile proteins (using phalloidin). To prepare muscles for imaging, chemical fixing, chemical staining and antibody labelling will be conducted.

The results will be used to interpret existing functional characteristics, hence providing a bigger picture of mechanisms underlying the energy utilisation by cardiac contractile proteins and the energy supply pathway in the diabetic heart. This will be important for diagnosis and treatment of heart disease.

Project

Gastric motility is coordinated by bioelectric activities and dysrhythmic activity has been linked with motility disorders. Magnetogastrography (MGG) is the non-invasive measurement of the biomagnetic fields generated by the gastric bioelectrical activities and can be used to develop biomarkers to diagnose gastric motility disorders. However, characterisation of bioelectrical activity using MGG is challenging because source models are not well developed and the impact of anatomical variation is not well understood.

We have recently developed a method to spatially co-register subject-specific anatomical models during simultaneous bioelectrical mapping and MGG measurement, which helps us to better understand the impact of anatomical details in the recorded biomagnetic data.

This project aims to develop techniques to characterise the gastric bioelectrical activities from biomagnetic recordings, where anatomical information will be used to constrain the solution space in the source localisation process.

Specific Aims:
- Explore existing source localisation techniques in the cardiac and brain fields.
- Define source (dipole) models for the existing stomach geometries and perform MGG simulations.
- Develop anatomically constrained source localisation techniques for the MGG data.
- Perform data visualisation using ParaView.

Preferred Skills:
- No background knowledge of physiology or anatomy is required, but ideal candidates should have interest in electrophysiology and mathematical modelling.
- Working knowledge of MATLAB is preferred.

Project

The mixing and grinding of food particles inside the stomach is carried out by muscular wave-like contractions. Magnetic resonance imaging (MRI) has been used to image these muscle wave-like contractions to develop dynamic 3D models of gastric motility. These models have subsequently been used to develop subject specific Computational Fluid Dynamics (CFD) models of gastric motility. But these CFD models currently lack pyloric sphincter activity and thus do not model any emptying. Co-ordinated pylorus function with contractions of the stomach is crucial for proper gastric emptying. Incorporation of gastric emptying into these subject specific CFD models of gastric motility is essential in an in-sillico model of food breakdown within the stomach.

This project aims to develop an automated framework to infer pyloric sphincter contractions/activity based on existing MRI recordings of gastric motility in the antrum.

Specific Aims:
- Analyse and segment (ML) existing MRI data for pylorus motions.
- Use segmented pyloric motion to derive pylorus contraction timings.
- Develop techniques to model/simulate 3D deformations of the pyloric sphincter coupled with existing gastric wall contractions.
- Perform CFD simulations and develop visualisation techniques.

Preferred Skills:
- No background knowledge in physiology or anatomy is required.
- Ideal candidates should have interest in image processing and segmentation.
- No previous experience will be required but familiarity with MATLAB is preferable.

Project

Breast cancer affects 1 in 9 women in Aotearoa New Zealand. A key challenge for clinicians is determining where tumours displace as the breast deforms during changes in patient positioning during treatment procedures such as surgery or radiotherapy.
If the patient’s skin can be tracked, biomechanical models that are being developed by the Auckland Bioengineering Institute can be used to recover information about the tumour location by tracking the deformation process. Cutting edge machine learning motion tracking methodologies, like “Track everything everywhere all at once” (as shown in the figure) are capable of achieving high performance on traditional tasks such as tracking images of buildings, cars, giraffes, and crowds. However, no data sets exist which allow the evaluation of these methods on human skin.

This project aims to collect a novel, high resolution, dataset of skin motion measurements and use it to compare the performance of these cutting edge methods against traditional methods of feature tracking.

Students will join a multi-disciplinary team of bioengineers, clinicians, computer vision and instrumentation specialists.

This project will suit students with an interest in computer vision, machine learning and computer programming.

Project

The Auckland Bioengineering Institute is offering funded summer research scholarships for high-achieving and ambitious Māori university students studying science, engineering, maths, medicine, population health, or similar, to conduct biomedical engineering research.

Students will work with established research groups at the Auckland Bioengineering Institute, doing cutting-edge biomedical engineering research in the fields of medical device design and development, experimental physiology, computational modelling, software / algorithm development, clinical translation of new medical technologies, or similar.

Students will be integrated into a cohort of Māori students as well as other students at ABI, and will be well-supported in ABI’s whānau-oriented environment.

Project

The Auckland Bioengineering Institute is offering funded summer research scholarships for high-achieving and ambitious Pacific university students studying science, engineering, maths, medicine, population health, or similar, to conduct biomedical engineering research.

Students will work with established research groups at the Auckland Bioengineering Institute, doing cutting-edge biomedical engineering research in the fields of medical device design and development, experimental physiology, computational modelling, software / algorithm development, clinical translation of new medical technologies, or similar.

Students will be integrated into a cohort of Pacific students as well as other students at ABI, and will be well-supported in ABI’s whānau-oriented environment.

Projects

Transforming into a powerful contracting organ during delivery, dramatic changes of the human uterus from a quiet organ are caused by mechanisms that remain unclear. Much like electrical signals in the heart, membrane voltage waves traveling over the uterine tissue trigger steep rises in intracellular calcium, which binds with and activates the cell’s contractile machinery. However, the way contractions are coordinated over the whole uterine tissue remains something of a mystery. Studying the maternal uterus as it changes in preparation for delivery is clinically complicated. To help understand these mechanisms so we can detect problems with pregnancy well before delivery, we use computational and mathematical modeling techniques

Project

This project will use medical images and video data to construct a computational model of a generic infant’s musculoskeletal system. This model will be used as part of our larger project assessing the health of infant’s neuromotor development from videos of their early movement patterns, from which risk of developing cerebral palsy and related conditions can be predicted.

The selected summer scholar will work alongside graduate students and research fellows in the Department of Exercise Sciences and the Auckland Bioengineering Institute.

Requirement

Students applying for this project should have a demonstrated interest in at least one of the following subjects: biomechanics, biomedical engineering, exercise science, or neuroscience.

Apply for this project in the Faculty of Science application form.

Project

This project will use medical images and video data to construct a computational model of a generic infant’s musculoskeletal system. This model will be used as part of our larger project assessing the health of infant’s neuromotor development from videos of their early movement patterns, from which risk of developing cerebral palsy and related conditions can be predicted.

The selected summer scholar will work alongside graduate students and research fellows in the Department of Exercise Sciences and the Auckland Bioengineering Institute.

Requirement

Students applying for this project should have a demonstrated interest in at least one of the following subjects: biomechanics, biomedical engineering, exercise science, or neuroscience.

Apply for this project in the Faculty of Science application form.