Search

Kidney abnormality is one of the major concerns in modern society, and it affects millions of people around the world. To diagnose different abnormalities in human kidneys, a narrow-beam x-ray imaging procedure, computed tomography, is used, which creates cross-sectional slices of the kidneys. Several deep-learning models have been successfully applied to computer tomography images for classification and segmentation purposes. However, it has been difficult for clinicians to interpret the model’s specific decisions and, thus, creating a “black box” system. Additionally, it has been difficult to integrate complex deep-learning models for internet-of-medical-things devices due to demanding training parameters and memory-resource cost. To overcome these issues, this study proposed (1) a lightweight customized convolutional neural network to detect kidney cysts, stones, and tumors and (2) understandable AI Shapely values based on the Shapley additive explanation and predictive results based on the local interpretable model-agnostic explanations to illustrate the deep-learning model. The proposed CNN model performed better than other state-of-the-art methods and obtained an accuracy of 99.52 ± 0.84% for K = 10-fold of stratified sampling. With improved results and better interpretive power, the proposed work provides clinicians with conclusive and understandable results.

Food waste is a serious problem worldwide, including in Europe. Research efforts are being carried out to reduce food waste. In this paper, we focus on using modern digital technologies (also known as Industry 4.0 technologies) to reduce waste in food supply chains. Based on interactions with a number of food companies in Europe over the last four years using Action Research, we provide new insights on the motivations and challenges for food companies when they are engaged in the use of technologies for reducing food waste in their supply chains. Motivations for firms include improved food quality of their produce, improved reliability, support in meeting legal requirements, a green image, and improved revenues from selling the food that has been saved. However, data security issues and trust issues posed challenges in using these technologies. Since this is an emerging area of research, we look at potential business models for technology companies for working with food companies in reducing food waste, identify value propositions and value capture, and look at how these investments in technologies can improve the sustainability of food businesses. We believe technology companies can leverage the opportunities, develop new business models with value propositions around the use of technologies, and support food companies via timely alerts in case of potential quality issues. Value capture occurs via the sale of hardware and subscriptions.

As global demographics change, ageing is a global phenomenon which is increasingly of interest in our modern and rapidly changing society. Thus, the application of proper prognostic indices in clinical decisions regarding mortality prediction has assumed a significant importance for personalized risk management (i.e., identifying patients who are at high or low risk of death) and to help ensure effective healthcare services to patients. Consequently, prognostic modelling expressed as all-cause mortality prediction is an important step for effective patient management. Machine learning has the potential to transform prognostic modelling. In this paper, results on the development of machine learning models for all-cause mortality prediction in a cohort of healthy older adults are reported. The models are based on features covering anthropometric variables, physical and lab examinations, questionnaires, and lifestyles, as well as wearable data collected in free-living settings, obtained for the “Healthy Ageing Initiative” study conducted on 2291 recruited participants. Several machine learning techniques including feature engineering, feature selection, data augmentation and resampling were investigated for this purpose. A detailed empirical comparison of the impact of the different techniques is presented and discussed. The achieved performances were also compared with a standard epidemiological model. This investigation showed that, for the dataset under consideration, the best results were achieved with Random UnderSampling in conjunction with Random Forest (either with or without probability calibration). However, while including probability calibration slightly reduced the average performance, it increased the model robustness, as indicated by the lower 95% confidence intervals. The analysis showed that machine learning models could provide comparable results to standard epidemiological models while being completely data-driven and disease-agnostic, thus demonstrating the opportunity for building machine learning models on health records data for research and clinical practice. However, further testing is required to significantly improve the model performance and its robustness.

Capturing hand motions for hand function evaluations is essential in the medical field. For many allied health professionals, measuring joint range of motion (ROM) is an important skill. While the universal goniometer (UG) is the most used clinical tool for measuring joint ROM, developments in current sensor technology are providing clinicians with more measurement possibilities than ever. For rehabilitation and manual dexterity evaluations, different data gloves have been developed. However, the reliability and validity of sensor technologies when used within a smart device remain somewhat unclear. This study proposes a novel electronically controlled sensor monitoring system (ECSMS) to obtain the static and dynamic parameters of various sensor technologies for both data gloves and individual sensor evaluation. Similarly, the ECSMS was designed to closely mimic a human finger joint, to have total control over the joint, and to have an exceptionally high precision. In addition, the ECSMS device can closely mimic the movements of the finger from hyperextension to a maximum ROM beyond any person’s finger joint. Due to the modular design, the ECSMS’s sensor monitoring board is independent and extensible to include various technologies for examination. Additionally, by putting these sensory devices through multiple tests, the system accurately measures the characteristics of any rotary/linear sensor in and out of a glove. Moreover, the ECSMS tracks the movement of all types of sensors with respect to the angle values of finger joints. In order to demonstrate the effectiveness of sensory devices, the ECSMS was first validated against a recognised secondary device with an accuracy and resolution of 0.1°. Once validated, the system simultaneously determines real angles alongside the hand monitoring device or sensor. Due to its unique design, the system is independent of the gloves/sensors that were tested and can be used as a gold standard to realise more medical equipment/applications in the future. Consequently, this design greatly enhances testing measures within research contact and even non-contact systems. In conclusion, the ECSMS will benefit in the design of data glove technologies in the future because it provides crucial evidence of sensor characteristics. Similarly, this design greatly enhances the stability and maintainability of sensor assessments by eliminating unwanted errors. These findings provide ample evidence for clinicians to support the use of sensory devices that can calculate joint motion in place of goniometers.