A5042053820- AI in cancer detection
- Radiomics and Machine Learning in Medical Imaging
- Hearing Loss and Rehabilitation
- Medical Image Segmentation Techniques
- Image Retrieval and Classification Techniques
- Hearing, Cochlea, Tinnitus, Genetics
- Image Processing Techniques and Applications
- Neural dynamics and brain function
- Image and Signal Denoising Methods
- Advanced Vision and Imaging
- MRI in cancer diagnosis
- Advanced Neural Network Applications
- EEG and Brain-Computer Interfaces
- Imbalanced Data Classification Techniques
- Phonocardiography and Auscultation Techniques
- Neuroscience and Neural Engineering
- Cell Image Analysis Techniques
- Artificial Intelligence in Healthcare and Education
- Neural Networks and Applications
- Machine Learning and Data Classification
- Blind Source Separation Techniques
- Advanced Data Compression Techniques
- Noise Effects and Management
- Advanced Image and Video Retrieval Techniques
- Anomaly Detection Techniques and Applications
Queensland University of Technology
2015-2025
University of the Sunshine Coast
2023-2025
George Washington University
2023
George Washington University Hospital
2023
Sofie Biosciences (United States)
2022
Oxford Brookes University
2022
The University of Queensland
2011-2021
University of Nottingham
2021
Sandia National Laboratories
2021
The University of Adelaide
2014-2017
Diabetes mellitus is a chronic disease and major public health challenge worldwide. According to the International Federation, there are currently 246 million diabetic people worldwide, this number expected rise 380 by 2025. Furthermore, 3.8 deaths attributable diabetes complications each year. It has been shown that 80% of type 2 can be prevented or delayed early identification at risk. In context, several data mining machine learning methods have used for diagnosis, prognosis, management...
International challenges have become the standard for validation of biomedical image analysis methods. Given their scientific impact, it is surprising that a critical common practices related to organization has not yet been performed. In this paper, we present comprehensive conducted up now. We demonstrate importance and show lack quality control consequences. First, reproducibility interpretation results often hampered as only fraction relevant information typically provided. Second, rank...
In this paper, we present an improved algorithm for the segmentation of cytoplasm and nuclei from clumps overlapping cervical cells. This problem is notoriously difficult because degree overlap among cells, poor contrast cell presence mucus, blood, inflammatory Our methodology addresses these issues by utilizing a joint optimization multiple level set functions, where each function represents within clump, that have both unary (intracell) pairwise (intercell) constraints. The constraints are...
Mass detection from mammograms plays a crucial role as pre- processing stage for mass segmentation and classification. The of masses is considered to be challenging problem due their large variation in shape, size, boundary texture also because low signal noise ratio compared the surrounding breast tissue. In this paper, we present novel approach detecting using cascade deep learning random forest classifiers. first classifier consists multi-scale belief network that selects suspicious...
We describe an automated methodology for the analysis of unregistered cranio-caudal (CC) and medio-lateral oblique (MLO) mammography views in order to estimate patient's risk developing breast cancer. The main innovation behind this lies use deep learning models problem jointly classifying mammogram respective segmentation maps lesions (i.e., masses micro-calcifications). This is a holistic that can classify whole mammographic exam, containing CC MLO maps, as opposed classification...
In this paper, we introduce and evaluate the systems submitted to first Overlapping Cervical Cytology Image Segmentation Challenge, held in conjunction with IEEE International Symposium on Biomedical Imaging 2014. This challenge was organized encourage development benchmarking of techniques capable segmenting individual cells from overlapping cellular clumps cervical cytology images, which is a prerequisite for next generation computer-aided diagnosis cancer. particular, these automated must...
Precision medicine approaches rely on obtaining precise knowledge of the true state health an individual patient, which results from a combination their genetic risks and environmental exposures. This approach is currently limited by lack effective efficient non-invasive medical tests to define full range phenotypic variation associated with health. Such critical for improved early intervention, better treatment decisions, ameliorating steadily worsening epidemic chronic disease. We present...
As autonomous vehicles and racing rise in popularity, so does the need for faster more accurate detectors. While our naked eyes are able to extract contextual information almost instantly, even from far away, image resolution computational resources limitations make detecting smaller objects (that is, that occupy a small pixel area input image) genuinely challenging task machines wide-open research field. This study explores how popular YOLOv5 object detector can be modified improve its...
Proximal femoral fractures are an important clinical and public health issue associated with substantial morbidity early mortality. Artificial intelligence might offer improved diagnostic accuracy for these fractures, but typical approaches to testing of artificial models can underestimate the risks intelligence-based systems.We present a preclinical evaluation deep learning model intended detect proximal in frontal x-ray films emergency department patients, trained on from Royal Adelaide...
Cells within the tumour microenvironment (TME) can impact development and influence treatment response. Computational approaches have been developed to deconvolve TME from bulk RNA-seq. Using scRNA-seq profiling breast tumours we simulate thousands of mixtures, representing purities cell lineages, compare performance nine deconvolution methods (BayesPrism, Scaden, CIBERSORTx, MuSiC, DWLS, hspe, CPM, Bisque, EPIC). Some are more robust in deconvolving mixtures with high purity levels. Most...
Muscle fatigue is often a result of unhealthy work practice. It has been known for some time that there significant change in the spectrum electromyography (EMG) muscle when it fatigued. Due to very complex nature this signal however, difficult use information reliably automate process onset determination. If such implementation were feasible, could be used as an indicator reduce chances work-place injury. This research report on effectiveness wavelet transform applied EMG means identifying...
In this paper, we describe a model of the human visual system (HVS) based on wavelet transform. This is largely previously proposed model, but has number modifications that make it more amenable to potential integration into image compression scheme. These include use separable transform instead cortex transform, application contrast sensitivity function (CSF), and simplified definition subband allows one predict noise visibility directly from coefficients. Initially, outline luminance,...
In this paper, we propose a novel algorithm for rule extraction from support vector machines (SVMs), termed SQRex-SVM. The proposed method extracts rules directly the vectors (SVs) of trained SVM using modified sequential covering algorithm. Rules are generated based on an ordered search most discriminative features, as measured by interclass separation. Rule performance is then evaluated rates true and false positives area under receiver operating characteristic (ROC) curve (AUC). Results...