News
QSPectral presentation at Machine Learning and Machine Intelligence Conference, July, 2025
Quantum Pre-Processing for Image Recognition
This study investigates a method for image classification using hybrid quantum-classical neural networks. A quantum pre-processing step is introduced to enhance the feature extraction process. A quantum convolutional algorithm is first applied to the images, and a classical neural network is employed to perform classification. Results confirm that preparing input data using quantum methods enables faster convergence and provides better accuracy in the early learning phases. However, it was also noticed that when this approach is paired with a traditional Convolutional Neural Network (CNN), while showing improvements, they aren’t as pronounced, indicating that there might be some overlap in feature extraction capabilities. The research emphasises the importance of quantum pre-processing for image classification, especially when using noisy or distorted data, marking a distinct contribution by evaluating robustness under combined noise types—a relatively unexplored aspect in prior studies on Quanvolutional networks.
QSPectral:Invited presentation at Quantum Australia Conference, Brisbane, March 2025
Quantum Pre-processing for Machine Learning Applications in Industry
Preprocessing Examples
1. Classifying data before applying other methods
2. Image recognition with noisy data
3. Protein sequencing for processing Rare Earth Elements (REE)
QSPectral presentation at Machine Learning and Machine Intelligence Conference, Osaka, Japan, 2024
Machine Learning and Quantum Intelligence for Challenging Data Scenarios
The advent of more accessible Machine Learning (ML) methods has opened new avenues for ML-driven applications in traditionally data-scarce domains. One such challenging area that we explore in this presentation is predicting the success of start-ups where data scarcity and quality pose significant obstacles. In this paper, we propose leveraging Machine Learning and specifically Quantum Machine Learning, to address the complexities of decision-making. We argue that the unique features of the dataset can be effectively processed using Quantum Machine Learning (QML). This offers advantages in creating feature spaces not achievable through classical methods, thereby compensating for the limitations of the data. Moreover, the potential acceleration of processing capability with the advent of large-scale quantum processors adds another dimension to the advantages.
QSPectral presented a talk on “Artificial Intelligence for Predicting Start-up Success” at the Australian Centre for Entrepreneurship Research Exchange (ACERE) conference in February 2023.
QSPectral presents on Quantum Machine Learning at the Australian Institute of Physics Congress
QSPectral presented on Quantum Computing for Simulating Successful Ventures: Investor Portfolio Optimisation at IFORS 2021
We propose that by simulating the startup eco-system as a Brownian motion model with data driven drift and diffusion parameters analogous to methods used in stock analysis, we can provide more rigour and insight in this challenging decision-making scenario.
Given that the model is analogous to a well understood physical system that has quantum effects, we will also show that this method is amenable to using more computationally competitive quantum algorithms that have the potential to solve larger scale and currently intractable scenarios.
QSPectral presents invited talk at Emerging Big Data Technologies Summit
Data Driven Decision Making from Remote Health Monitoring to Venture Prediction
Health systems face an increasing demand on resources, as the world’s population ages. Technology in the form of sensor-based remote monitoring systems has emerged as a viable option to mitigate this asymmetry between supply and demand enabling vulnerable individuals to live at home safely and independently. We will present novel data driven systems for sensor based human activity recognition, chronic disease monitoring as well as health risk assessment.
The advance of analytics methods provides a means to contextualise and aggregate data from disparate devices in a meaningful manner.
In this presentation, we will also discuss applications across a number of domains including the exciting area of data driven evaluation of startup
QSPectral-Uni Adelaide Paper on Networks for Military Operations at MiLCIS 2016
Achieving Policy Defined Networking for Military Operations Authors: Hung Xuan Nguyen, Michael Webb (The University of Adelaide, Australia); Sanjeev Naguleswaran (QSPectral Systems, Australia)
Abstract. In the past few years, significant progress has been made in software defined networking in a quest to increase automation, improve network agility and security, simplify network configuration and reduce resources to establish and maintain the network. There are now a vast number of studies exploring how to utilise policies to achieve these goals. Applying outcomes of these studies to military networks requires a clear understanding of military applications and available mechanisms to implement the appropriate policies in a software defined networking environment. In this paper, we identify several military networks where automatic policy defined networking is crucial. We further present a prototype policy defined networking solution that automatically translates high-level policies into device level implementations.
At Brisbane’s new “Capital” Innovation Space
myQ clinical decision support powered by QSpectral analytics
QSpectral’s MonitorMyHealth analytics is enabling myQ Decision Support in Singapore on our partner myHealthSentinel’s Telehealth platform.
http://www.prweb.com/releases/2016/07/prweb13528473.htm
Recent Comments