Principal speaker
Associate Professor David Ackland
Other speakers
Associate Professor David John Saxby Dr Damith Senanayake Dr Claudio Pizzolato
2022 Menzies Health Institute Queensland Seminar Series
Griffith Centre of Biomedical and Rehabilitation Engineering (GCORE) Workshop
Presenters:
Associate Professor David Ackland and Dr Damith Senanayake, University of Melbourne
Associate Professor David John Saxby and Dr Claudio Pizzolato, Griffith University
Title: Artificial Intelligence and Computational Automation in Orthopaedics
Seminar Schedule -
08:30 - 09.00 Prof David Lloyd - Introduction
09:00 - 09:30 A/Prof David Ackland
09:30 - 10:00 A/Prof David John Saxby
10:00 - 10:30 Morning Tea
10:30 - 11:00 Dr Damith Senanayake
11:00 - 11:30 Dr Claudio Pizzolato
11:30 - 12:00 Q&A Forum - Moderator: Dr Azadeh Nasseri
12.00 - 1:00 Light Lunch
Presenters -
Associate Professor David Ackland -
Workshop Abstract - Real-time conversion of IMU data to joint angles using artificial neural networks
Inertial Measurement Units (IMUs) are low-cost, wireless motion measurement devices capable of data acquisition in and outside of the laboratory; however, joint angle measurement using IMUs depends on sensor alignment with body segments, or joint calibration (Lebleu et al. 2020), and accurate out-of-plane joint motion is challenging to achieve. The aim of this study was to employ a new deep neural network model to convert IMU data to joint angles in real time such that the resultant angles data were indistinguishable from those derived from a reference standard video motion capture system. This modelling approach, which produced high joint angle predictability across a range of contrasting motion speeds and planes of joint motion, has application in remote diagnostics and rehabilitation i.e., telemedicine, personal sports and exercise industries, defence, and the film and animation industry.
Biography -
A/Prof David Ackland is the Deputy Director of the ARC Training Centre for Medical Implant Technologies, and an ARC Future Fellow. He graduated with a Bachelor of Science (Neuroscience) and Bachelor of Engineering (Mechanical), and went on to complete a PhD and postdoctoral studies in musculoskeletal biomechanics at the University of Melbourne. A/Prof Ackland undertakes research focusing on computational modelling and simulation of human movement biomechanics and data analytics, particularly in orthopaedics applications. He employs medical imaging, human motion experiments, musculoskeletal modelling, and in vitro joint-biomechanics experiments as his primary research techniques.
Associate Professor David John Saxby -
Workshop Abstract - Applications for neural networks and automated surgery planning to improve clinical efficiency and guide surgical installation.
(Topic 1) Currently, when presenting to an orthopaedic surgeon for consultation regarding anterior cruciate ligament rupture, ~25% of consultation time is spent on manual measurements performed by the surgeon using the patient's medical imaging. By reducing this time burden, clinical throughput can be increased which for private surgeons would result in improved revenue and for public surgeons would reduce waitlists. We present a neural network solution to this problem, where the clinical measurements are performed rapidly, automatically, and accurately.
(Topic 2) Rupture of the wrist's scapholunate interosseous ligament (SLIL) is common in athletes, laborers, and the elderly. Current commercially available orthopaedic solutions for SLIL rupture are inadequate, due to poor clinical outcomes, unreliable performance, and surgical complexity. Over the past 5 years, we have developed a novel synthetic scaffold system which includes design, testing, and fabrication of a SLIL reconstruction implant and associated installation tools. This product, termed bioSLIL, fixes the dissociated scaphoid and lunate, can withstand the mechanical demands of everyday use, and degrades as biological ingrowth occurs. We present a computational method to precision-install this implant to achieve the lowest risk of failure during rehabilitation.
Biography -
A/Prof David John Saxby, DECRA Fellow of the Australian Research Council, is a research lead within the Griffith Centre for Biomedical and Rehabilitation Engineering (GCORE), Griffith University. Dr Saxby graduated from the University of Ottawa in 2008 and 2010 with Bachelors and Masters degrees, respectively, and from Griffith University in 2016 with a Doctorate. His research is focused on the development and application of digitally integrated technologies for orthopaedics, musculoskeletal health, defense, and ergonomics.
Dr Damith Senanayake -
Workshop Abstract - Predicting fall and fracture risk using DEXA scans and patient clinical examination data: A deep-learning study
Bone Mineral Density scans (e.g., DEXA) are often used as a tool for evaluating bone health in individuals susceptible to falls and fractures, especially in older Australians affected by bone pathology. Analysis of DEXA scans is typically carried out by a trained expert, and prognosis based on image features such as pixel intensity in specific regions of interest (ROI); however, such an approach provides only a subjective measure which may not always correlate with fracture risk. This study seeks to employ artificial neural networks to achieve accurate fall and fracture risk predictions by combining DEXA images together with clinical examination data including mineral levels (calcium, vitamin D), behavioural factors (smoking, alcohol consumption, exercise frequency) and demographic data (sex, age, etc.). This modelling approach, which is the first of its kind to the best of our knowledge, integrates a range of data formats and account for low data abundance while providing increased accuracy and automation in fracture and fall risk assessment. The findings may ultimately be used to inform patient management and clinical decision making to mitigate falls and fracture.
Biography -
Dr Damith Senanayake is a Research Fellow in the Departments of Biomedical Engineering and Mechanical Engineering at the University of Melbourne. His expertise is in machine learning applied to high-dimensional biological data and human motion analysis. He obtained a Bachelor's degree in Computer Science and Engineering from the University of Moratuwa, Sri Lanka in 2015 and a PhD in Engineering from the University of Melbourne in 2020.
Dr Claudio Pizzolato -
Workshop Abstract - Leveraging computational models and AI to enable real-time monitoring of musculoskeletal tissue states outside the laboratory.
Mechanics is a fundamental driver of musculoskeletal adaptation, affecting material properties and morphology of tissues. Inappropriate mechanical loading is associated to maladaptation and consequent initiation and progression of numerous musculoskeletal conditions, such as tendinopathy and osteoarthritis. Conversely, appropriate loading leads to improved tissue mechanical properties, and it is a desirable target to guide efficacious and personalized rehabilitation interventions. Recent experimental and validation studies from our group demonstrated the ability of our computational models to predict physiologically plausible mechanics of selected musculoskeletal tissues. The current challenge is enabling this same level of precision outside the research laboratory.
We present our approach combining garment embedded wearable sensors, computational modelling of the neuromusculoskeletal system, and artificial intelligence to enable monitoring of tissue states (i.e., stress and strain) during activities of daily living and sport.
Biography -
Dr Claudio Pizzolato is a research lead within the Griffith Centre of Biomedical and Rehabilitation Engineering (GCORE), focusing on real-time computational modelling of neural and musculoskeletal systems, and their application to rehabilitation. He received a Master of Mechatronic Engineering from the University of Padova (Italy) in 2011 and a PhD in computational biomechanics from Griffith University (Australia) in 2016.
RSVP by Monday 28 March 2022 -
https://forms.office.com/r/rfJg3bSbgJ
Seminar Flyer -
Download the flyer for this seminar here
Microsoft Teams meeting link -
or
https://bit.ly/GCORE-ARCCMITWorkshop
Event categories
RSVP
RSVP on or before Monday 28 March 2022 13.52 pm, by email mhiq@griffith.edu.au , or by phone 07 5678 0907 , or via https://forms.office.com/r/rfJg3bSbgJ