Machine learning (ML) models are becoming integral in healthcare technologies, presenting a critical need for formal assurance to validate their safety, fairness, robustness, and trustworthiness. These models are inherently prone to errors, potentially posing serious risks to patient health and could even cause irreparable harm. Traditional software assurance techniques rely on fixed code and do not directly apply to ML models since these algorithms are adaptable and learn from curated datasets through a training process. However, adapting established principles, such as boundary testing using synthetic test data can effectively bridge this gap. To this end, we present a novel technique called Mix-Up Boundary Analysis (MUBA) that facilitates evaluating image classifiers in terms of prediction fairness. We evaluated MUBA for two important medical imaging tasks -- brain tumour classification and breast cancer classification -- and achieved promising results. This research aims to showcase the importance of adapting traditional assurance principles for assessing ML models to enhance the safety and reliability of healthcare technologies. To facilitate future research, we plan to publicly release our code for MUBA.
Coronary angiography analysis is a common clinical task performed by cardiologists to diagnose coronary artery disease (CAD) through an assessment of atherosclerotic plaque's accumulation. This study introduces an end-to-end machine learning solution developed as part of our solution for the MICCAI 2023 Automatic Region-based Coronary Artery Disease diagnostics using x-ray angiography imagEs (ARCADE) challenge, which aims to benchmark solutions for multivessel coronary artery segmentation and potential stenotic lesion localisation from X-ray coronary angiograms. We adopted a robust baseline model training strategy to progressively improve performance, comprising five successive stages of binary class pretraining, multivessel segmentation, fine-tuning using class frequency weighted dataloaders, fine-tuning using F1-based curriculum learning strategy (F1-CLS), and finally multi-target angiogram view classifier-based collective adaptation. Unlike many other medical imaging procedures, this task exhibits a notable degree of interobserver variability. %, making it particularly amenable to automated analysis. Our ensemble model combines the outputs from six baseline models using the weighted ensembling approach, which our analysis shows is found to double the predictive accuracy of the proposed solution. The final prediction was further refined, targeting the correction of misclassified blobs. Our solution achieved a mean F1 score of $37.69\%$ for coronary artery segmentation, and $39.41\%$ for stenosis localisation, positioning our team in the 5th position on both leaderboards. This work demonstrates the potential of automated tools to aid CAD diagnosis, guide interventions, and improve the accuracy of stent injections in clinical settings.
Over the past few years, surgical data science has attracted substantial interest from the machine learning (ML) community. Various studies have demonstrated the efficacy of emerging ML techniques in analysing surgical data, particularly recordings of procedures, for digitizing clinical and non-clinical functions like preoperative planning, context-aware decision-making, and operating skill assessment. However, this field is still in its infancy and lacks representative, well-annotated datasets for training robust models in intermediate ML tasks. Also, existing datasets suffer from inaccurate labels, hindering the development of reliable models. In this paper, we propose a systematic methodology for developing robust models for surgical tool detection using noisy data. Our methodology introduces two key innovations: (1) an intelligent active learning strategy for minimal dataset identification and label correction by human experts; and (2) an assembling strategy for a student-teacher model-based self-training framework to achieve the robust classification of 14 surgical tools in a semi-supervised fashion. Furthermore, we employ weighted data loaders to handle difficult class labels and address class imbalance issues. The proposed methodology achieves an average F1-score of 85.88\% for the ensemble model-based self-training with class weights, and 80.88\% without class weights for noisy labels. Also, our proposed method significantly outperforms existing approaches, which effectively demonstrates its effectiveness.
Recent advancements in technology, particularly in machine learning (ML), deep learning (DL), and the metaverse, offer great potential for revolutionizing surgical science. The combination of artificial intelligence and extended reality (AI-XR) technologies has the potential to create a surgical metaverse, a virtual environment where surgeries can be planned and performed. This paper aims to provide insight into the various potential applications of an AI-XR surgical metaverse and the challenges that must be addressed to bring its full potential to fruition. It is important for the community to focus on these challenges to fully realize the potential of the AI-XR surgical metaverses. Furthermore, to emphasize the need for secure and robust AI-XR surgical metaverses and to demonstrate the real-world implications of security threats to the AI-XR surgical metaverses, we present a case study in which the ``an immersive surgical attack'' on incision point localization is performed in the context of preoperative planning in a surgical metaverse.
The recent progress in unmanned aerial vehicles (UAV) technology has significantly advanced UAV-based applications for military, civil, and commercial domains. Nevertheless, the challenges of establishing high-speed communication links, flexible control strategies, and developing efficient collaborative decision-making algorithms for a swarm of UAVs limit their autonomy, robustness, and reliability. Thus, a growing focus has been witnessed on collaborative communication to allow a swarm of UAVs to coordinate and communicate autonomously for the cooperative completion of tasks in a short time with improved efficiency and reliability. This work presents a comprehensive review of collaborative communication in a multi-UAV system. We thoroughly discuss the characteristics of intelligent UAVs and their communication and control requirements for autonomous collaboration and coordination. Moreover, we review various UAV collaboration tasks, summarize the applications of UAV swarm networks for dense urban environments and present the use case scenarios to highlight the current developments of UAV-based applications in various domains. Finally, we identify several exciting future research direction that needs attention for advancing the research in collaborative UAVs.
Radars are widely used to obtain echo information for effective prediction, such as precipitation nowcasting. In this paper, recent relevant scientific investigation and practical efforts using Deep Learning (DL) models for weather radar data analysis and pattern recognition have been reviewed; particularly, in the fields of beam blockage correction, radar echo extrapolation, and precipitation nowcast. Compared to traditional approaches, present DL methods depict better performance and convenience but suffer from stability and generalization. In addition to recent achievements, the latest advancements and existing challenges are also presented and discussed in this paper, trying to lead to reasonable potentials and trends in this highly-concerned field.
As a result of increasing population and globalization, the demand for energy has greatly risen. Therefore, accurate energy consumption forecasting has become an essential prerequisite for government planning, reducing power wastage and stable operation of the energy management system. In this work we present a comparative analysis of major machine learning models for time series forecasting of household energy consumption. Specifically, we use Weka, a data mining tool to first apply models on hourly and daily household energy consumption datasets available from Kaggle data science community. The models applied are: Multilayer Perceptron, K Nearest Neighbor regression, Support Vector Regression, Linear Regression, and Gaussian Processes. Secondly, we also implemented time series forecasting models, ARIMA and VAR, in python to forecast household energy consumption of selected South Korean households with and without weather data. Our results show that the best methods for the forecasting of energy consumption prediction are Support Vector Regression followed by Multilayer Perceptron and Gaussian Process Regression.
Under unexpected conditions or scenarios, autonomous vehicles (AV) are more likely to follow abnormal unplanned actions, due to the limited set of rules or amount of experience they possess at that time. Enabling AV to measure the degree at which their movements are novel in real-time may help to decrease any possible negative consequences. We propose a method based on the Local Outlier Factor (LOF) algorithm to quantify this novelty measure. We extracted features from the inertial measurement unit (IMU) sensor's readings, which captures the vehicle's motion. We followed a novelty detection approach in which the model is fitted only using the normal data. Using datasets obtained from real-world vehicle missions, we demonstrate that the suggested metric can quantify to some extent the degree of novelty. Finally, a performance evaluation of the model confirms that our novelty metric can be practical.
Breast cancer is one of the most common cause of deaths among women. Mammography is a widely used imaging modality that can be used for cancer detection in its early stages. Deep learning is widely used for the detection of cancerous masses in the images obtained via mammography. The need to improve accuracy remains constant due to the sensitive nature of the datasets so we introduce segmentation and wavelet transform to enhance the important features in the image scans. Our proposed system aids the radiologist in the screening phase of cancer detection by using a combination of segmentation and wavelet transforms as pre-processing augmentation that leads to transfer learning in neural networks. The proposed system with these pre-processing techniques significantly increases the accuracy of detection on Mini-MIAS.
Federated learning (FL) is becoming a popular paradigm for collaborative learning over distributed, private datasets owned by non-trusting entities. FL has seen successful deployment in production environments, and it has been adopted in services such as virtual keyboards, auto-completion, item recommendation, and several IoT applications. However, FL comes with the challenge of performing training over largely heterogeneous datasets, devices, and networks that are out of the control of the centralized FL server. Motivated by this inherent setting, we make a first step towards characterizing the impact of device and behavioral heterogeneity on the trained model. We conduct an extensive empirical study spanning close to 1.5K unique configurations on five popular FL benchmarks. Our analysis shows that these sources of heterogeneity have a major impact on both model performance and fairness, thus sheds light on the importance of considering heterogeneity in FL system design.