Search

Simple Summary: Breast cancer is one of the leading causes of cancer-related death in women. The early detection of breast cancer with screening mammograms plays a pivotal role in reducing mortality rates. Although population-based double-reading screening mammograms have reduced mortality by over 31% in women with breast cancer in Europe, continuing this program is difficult due to the shortage of radiologists. Artificial intelligence (AI) is an emerging technology that has provided promising results in medical imaging for disease detection. This study investigates the performance of AI models on an Australian mammographic database, demonstrating how transfer learning from a USA mammographic database to an Australian one, contrast enhancement on mammographic images and the quality of training data according to radiologists' concordance can improve breast cancer diagnosis. Our proposed methodology offers a more efficacious approach for AI to contribute to radiologists' decision making when interpreting mammography images. This paper investigates the adaptability of four state-of-the-art artificial intelligence (AI) models to the Australian mammographic context through transfer learning, explores the impact of image enhancement on model performance and analyses the relationship between AI outputs and histopathological features for clinical relevance and accuracy assessment. A total of 1712 screening mammograms (n = 856 cancer cases and n = 856 matched normal cases) were used in this study. The 856 cases with cancer lesions were annotated by two expert radiologists and the level of concordance between their annotations was used to establish two sets: a 'high-concordances subset' with 99% agreement of cancer location and an 'entire dataset' with all cases included. The area under the receiver operating characteristic curve (AUC) was used to evaluate the performance of Globally aware Multiple Instance Classifier (GMIC), Global-Local Activation Maps (GLAM), I&H and End2End AI models, both in the pretrained and transfer learning modes, with and without applying the Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm. The four AI models with and without transfer learning in the high-concordance subset outperformed those in the entire dataset. Applying the CLAHE algorithm to mammograms improved the performance of the AI models. In the high-concordance subset with the transfer learning and CLAHE algorithm applied, the AUC of the GMIC model was highest (0.912), followed by the GLAM model (0.909), I&H (0.893) and End2End (0.875). There were significant differences (p < 0.05) in the performances of the four AI models between the high-concordance subset and the entire dataset. The AI models demonstrated significant differences in malignancy probability concerning different tumour size categories in mammograms. The performance of AI models was affected by several factors such as concordance classification, image enhancement and transfer learning. Mammograms with a strong concordance with radiologists' annotations, applying image enhancement and transfer learning could enhance the accuracy of AI models. [ABSTRACT FROM AUTHOR]

Recent failures of bridges worldwide highlight the need for robust structural health surveillance systems to provide alerts for early damage and prevent severe loss. Structural health monitoring (SHM) continues to evolve and gain wide adoption among asset managers to mitigate catastrophic failure risks of structural systems. With the prolific integration of SHM with artificial intelligence and internet of things systems becoming prevalent, several data-driven approaches have gained popularity in SHM. However, it is challenging to obtain ground-truth-labelled damage samples for supervised machine learning methods. A class of shallow machine learning methods known as unsupervised methods (e.g. the one-class support vector machine (OCSVM)), which only uses unlabelled data from a healthy state, has been shown to achieve promising results. However, shallow machine learning methods require laborious expert-crafted feature inputs for optimal performance. Inspired by the automatic feature-extraction ability of deep neural networks, unsupervised methods (deep learning models trained without supervisory ground truth labels) such as autoencoder (AE)-based methods were explored to detect structural damage. Results obtained on both real-world data from Sydney Harbour Bridge, one of Australia's iconic structures, and data collected from laboratory specimens demonstrate the effectiveness of the AE-based method over conventional shallow methods such as the OCSVM. [ABSTRACT FROM AUTHOR]

The launch of OpenAI's ChatGPT model in late 2022, as most Australian universities wound down for summer holidays, elicited varied responses from higher education practitioners, policy makers and commentators that ranged from heightened concern and proscriptive impulses through to cautious excitement about the potentially disruptive, deceptive impact of university student use of AI chatbots (Skeat and Ziebell, 2023). Generative AI has both transformative and disruptive implications for conventional university assessment practices. Simultaneously, we observed a tension between university teaching and learning imperatives of digital literacy, academic integrity, student employability, and data security and privacy. Large Language Models (LLMs) run on deep learning programming, trained to process data in a way modelled on human brain cognition, to generate human-like responses to natural language prompts. Generative AI can answer and compose questions, write narratives, summarise documents, and construct essays, reports, reviews etc, and perform reflective writing capabilities (Li et al., 2023). Importantly, generative AI performs these tasks with substantially different degrees of accuracy, biases, and relevance potentially with each prompt. These dynamic and iterative learning abilities have significantly, sometimes imperceptibly, compromised the integrity and reliability of conventional university assessment types. Moreover, generative AI is improving incrementally, increasingly integrated into everyday software, platforms and apps (Liu and Bridgeman, June 2023). Nor is it only traditional written assessments that are at risk of disruption and invalidation. AI image generators like OpenAI's DALL-E can produce high-quality, realistic and fantastical artworks. ChatGPT-like AI models are designed for conversational and dialogic user experiences, programmed on natural and intuitive patterns of language use. Even without any targeted training in ethical, effective and critical 'prompt engineering' (cf. Liu, 2023) students can output passable assessment content. As well as concerns around digital literacy, academic integrity and meaningful learning, prompting, performed rudimentarily at least, blurs the lines between a student's original thinking (and integration of sources) and machine-generated output. The foundational challenge being in determining whether a student's submission is a result of their applied understanding or the AI's algorithmic capabilities. Yet, this GenAI interactional, iterative user experience can also be harnessed by educators to design, facilitate and assess socially constructivist, authentic, analytical, and innovative approaches to student learning (Liu and Bridgeman, June 2023). We report on a research project that implemented an iterative, nested, and collaborative assessment redesign (Lodge et al. 2023) as an alternative to a 2000-word Final Research Report due in the semester's penultimate week. For the redesign, we partially broke the one submission down into three, smaller critical reflections due across a semester. For the first, students used ChatGPT before, and then after, learning a prompt engineering approach (cf. Liu, 2023). Secondly, students reflected on their engagement with generative AI as a collaborator in comparison to their collaboration with peers on a task. The final critical reflection required students to anticipate how generative AI might impact their professional practices drawing on the subject's key topics. With ethics approval granted, our research findings are drawn from the roughly 10% of all students (n = 83) that chose the redesigned option. We analyse their three submissions in terms of existing themes in the literature (cf. Skeat and Ziebell, 2023) around academic integrity, digital literacy, institutional messaging and student belonging, and generative AI as 'study buddy' (Skeat and Ziebell, 2023). [ABSTRACT FROM AUTHOR]

Predicting electricity demand data is considered an essential task in decisions taking, and establishing new infrastructure in the power generation network. To deliver a high-quality electricity demand prediction, this paper proposes a hybrid combination technique, based on a deep learning model of Convolutional Neural Networks and Echo State Networks, named as CESN. Daily electricity demand data from four sites (Roderick, Rocklea, Hemmant and Carpendale), located in Southeast Queensland, Australia, have been used to develop the proposed hybrid prediction model. The study also analyzes five other machine learning-based models (support vector regression, multilayer perceptron, extreme gradient boosting, deep neural network, and Light Gradient Boosting) to compare and evaluate the outcomes of the proposed deep learning approach. The results obtained in the experimental study showed that the proposed hybrid deep learning model is able to obtain the highest performance compared to other existing models developed for daily electricity demand data forecasting. Based on the statistical approaches utilized in this study, the proposed hybrid approach presents the highest prediction accuracy among the compared models. The obtained results showed that the proposed hybrid deep learning algorithm is an excellent and accurate electricity demand forecasting method, which outperformed the state of the art algorithms that are currently used in this problem. [Display omitted] • A Hybrid deep learning model is proposed for daily electricity demand prediction. • Convolutional Neural Networks and Echo State Networks are integrated into a single forecasting algorithm. • The proposed deep learning-based approach is excellent in electricity demand prediction. • The proposed algorithm has practical use in inter-daily electricity demand prediction. [ABSTRACT FROM AUTHOR]

The energy generated by a solar photovoltaic (PV) system depends on uncontrollable factors, including weather conditions and solar irradiation, which leads to uncertainty in the power output. Forecast PV power generation is vital to improve grid stability and balance the energy supply and demand. This study aims to predict hourly day-ahead PV power generation by applying Temporal Fusion Transformer (TFT), a new attention-based architecture that incorporates an interpretable explanation of temporal dynamics and high-performance forecasting over multiple horizons. The proposed forecasting model has been trained and tested using data from six different facilities located in Germany and Australia. The results have been compared with other algorithms like Auto Regressive Integrated Moving Average (ARIMA), Long Short-Term Memory (LSTM), Multi-Layer Perceptron (MLP), and Extreme Gradient Boosting (XGBoost), using statistical error indicators. The use of TFT has been shown to be more accurate than the rest of the algorithms to forecast PV generation in the aforementioned facilities. [ABSTRACT FROM AUTHOR]