Background and motivation: Lung computed tomography (CT) techniques are high-resolution and are well adopted in the intensive care unit (ICU) for COVID-19 disease control classification. Most artificial intelligence (AI) systems do not undergo generalization and are typically overfitted. Such trained AI systems are not practical for clinical settings and therefore do not give accurate results when executed on unseen data sets. We hypothesize that ensemble deep learning (EDL) is superior to deep transfer learning (TL) in both non-augmented and augmented frameworks. Methodology: The system consists of a cascade of quality control, ResNet–UNet-based hybrid deep learning for lung segmentation, and seven models using TL-based classification followed by five types of EDL’s. To prove our hypothesis, five different kinds of data combinations (DC) were designed using a combination of two multicenter cohorts—Croatia (80 COVID) and Italy (72 COVID and 30 controls)—leading to 12,000 CT slices. As part of generalization, the system was tested on unseen data and statistically tested for reliability/stability. Results: Using the K5 (80:20) cross-validation protocol on the balanced and augmented dataset, the five DC datasets improved TL mean accuracy by 3.32%, 6.56%, 12.96%, 47.1%, and 2.78%, respectively. The five EDL systems showed improvements in accuracy of 2.12%, 5.78%, 6.72%, 32.05%, and 2.40%, thus validating our hypothesis. All statistical tests proved positive for reliability and stability. Conclusion: EDL showed superior performance to TL systems for both (a) unbalanced and unaugmented and (b) balanced and augmented datasets for both (i) seen and (ii) unseen paradigms, validating both our hypotheses.

Khanna NN, et al. J Korean Med Sci. 2023 Nov;38:e395. https://doi.org/10.3346/jkms.2023.38.e395

The unplanned growth of urbanization in towns and cities has led to variations in Land Surface Temperature (LST) as the green lands are converted into impervious structures without land cover management. In this study, an effort is made to study the transformational change in natural land cover area over the years and its impact on LST in Ernakulum District of Kerala, India. As per the current statistics from one of the reports on “Observed Rainfall Variability and Changes Over Kerala State”, by Indian Metrological Department in January 2020 shows that there is a significant decreasing trend in rainy days in many districts of Kerala which has resulted in highest temperature records in recent years. To understand the change in temperature trends in the district, Landsat data is collected for four different years 2005, 2010, 2015 and 2020 that maps Land Use and Land Cover (LULC) pattern for vegetation cover and built-up areas. It is important to interpret LULC maps as converting the vegetative land in built-up areas leads to multiple reflecting surfaces and temperature trapping which serves as the most direct way to understand the influence of global warming. The major merit of this research is (i) to interpret the temporal changes in LULC for categorizing the images in different land covers, (ii) to analyze Urban Thermal Field Variance Index (UTFVI) to quantify the urban heat island (UHI) effect in the region, (iii) to assess change in temperature by analyzing the spatio-temporal variations in mean temperatures for different land cover classes, and (iv) to analyze the trend of LST wrt to vegetation spectral index i.e., Normalized Differential Vegetation Index (NDVI) and spectral building index i.e., Normalized Difference Building Index (NDBI) to predict the percentage change in temperature in successive years using the widely used regression model. The computational simulation findings show that the percentage area of vegetated land decreases from 47.91 % to 29.04 % and the LST change analysis observed for years from 2005 to 2020 is 2.55 °C. For built up land, the percentage area is increased from 6.50 % to 15.27 % and the change in LST observed from year 2005–2020 is 4.82 °C which is a significant rise in temperature. The overall accuracy achieved for four respective years comes out to be 93, 89, 90, and 92 %. It has been observed from the analysis, that there is a change in temperature over the period of time from 2005 to 2020 b 3.45 °C. This is due to unplanned use of land cover classes that show considerable change in LULC maps that can be seen in the simulated results for a given study area. If the rate of change in area appeared to constantly change in this manner, then there is a possibility of further increase in land surface temperature and that will contribute in increase of temperature. The rise in mean LST from year 2020–2025 would be 0.5 °C and 1.0 °C for year 2030.

In today’s era, there is a need for a system that can automate the process of treatment for the patient if medical facilities are out of reach. Smart healthcare can step in to make the patient more self-dependent. 6G with its features can be seen as the future of smart healthcare with IoT and AI. 6G-Enabled IoT and AI for Smart Healthcare: Challenges, Impact, and Analysis offers the fundamentals, history, reality, and challenges faced in the smart healthcare industry today. It discusses the concepts, tools, and techniques of smart healthcare as well as the analysis used. The book details the role that machine learning-based deep learning and 6G-enabled IoT concepts play in the automation of smart healthcare systems. The book goes on to presents applications of smart healthcare through various real-world examples and includes chapters on security and privacy in the 6G-enabled and IoT environment, as well as research on the future prospects of the smart healthcare industry. This book: Offers the fundamentals, history, reality, and the challenges faced in the smart healthcare industry Discusses the concepts, tools, and techniques of smart healthcare as well as the analysis used Details the role that machine learning-based deep learning and 6G-enabled IoT concepts play in the automation of smart healthcare systems Presents applications of smart healthcare through various real-world examples Includes topics on security and privacy in 6G-enabled IoT, as well as research and future prospectus of the smart healthcare industry Interested readers of this book will include anyone working in or involved in smart healthcare research which includes, but is not limited to healthcare specialists, computer science engineers, electronics engineers, systems engineers, and pharmaceutical practitioners.