Abstract
The purpose of this article is to identify the most impactful research on customer churn and to map the conceptual and intellectual structure of its field of study. Data were collected from the WoS database, comprising 338 articles published between 1995 and 2020. Several bibliometric techniques were applied, including analysis of co-words, co-citation, bibliographic coupling, and co-authorship networks. R software and the Bibliometrix/Biblioshiny package were used to perform the analyses. The results identify the most active and influential authors, articles, and journals on the topic. More specifically, through co-citations and bibliographic coupling, it was possible to map the oldest articles (retrospective analysis) and the current research front (prospective analysis). The retrospective analysis, based on co-citations, revealed that the foundations of this research field are constructs such as quality of service, satisfaction, loyalty, and changing behaviors. The prospective analysis, performed through bibliographic coupling, revealed that current research is embedded in predictive analysis, clusters, data mining, and algorithms. The results provide robust guidance for further investigation in this field.

Abstract
The grapevine variety plays an important role in wine chain production, thus identifying it is crucial for control activities. However, the specialists responsible for identifying the different varieties, mainly through visual analysis, are disappearing. In this scenario, Deep Learning (DL) classification techniques become a possible solution to handle professionals' scarcity. Nevertheless, previous experiments show that trained classification models use the background information to make decisions, which should be avoided. In this paper, we present a study allowing the assessment of removing background regions from the grapevine images in the improvement classification using DL models. The Xception model is trained with a normal dataset and its segmented version. The Local Interpretable Model-Agnostic Explanations (LIME), Grad-CAM, and Grad-CAM++ approaches are used to visualize the segmentation impact in classification decisions. F1-score of 0.92 and 0.94 were achieved, respectively, for segmented-dataset and normal-dataset trained models. Despite the model trained with the segmented-dataset to achieve a worse performance, the Explainable Artificial Intelligence (XAI) approaches showed that it looks into more reliable regions when making decisions.

Abstract
Neuroblastoma (NB) is the most common extracranial solid tumor in childhood. Genomic amplification of MYCN is associated with poor outcomes and is detected in 16% of all NB cases. CT scans and MRI are the imaging techniques recommended for diagnosis and disease staging. The assessment of imaging features such as tumor volume, shape, and local extension represent relevant prognostic information. Radiogenomics have shown powerful results in the assessment of the genotype based on imaging findings automatically extracted from medical images. In this work, random forest was used to classify the MYCN amplification using radiomic features extracted from CT slices in a population of 46 NB patients. The learning model showed an area under the curve (AUC) of 0.85 ± 0.13, suggesting that radiomic-based methodologies might be helpful in the extraction of information that is not accessible by human naked eyes but could aid the clinicians on the diagnosis and treatment plan definition. Clinical relevance - This approach represents a random forest-based model to predict the MYCN amplification in NB patients that could give a faster, earlier, and repeatable analysis of the tumor along the time.

Abstract
Purpose The aim of this paper is to evaluate the use of blockchain for identity management (IdM) in the context of the Internet of things (IoT) while focusing on privacy-preserving approaches and its applications to healthcare scenarios. Design/methodology/approach The paper describes the most relevant IdM systems focusing on privacy preserving with or without blockchain and evaluates them against ten selected features grouped into three categories: privacy, usability and IoT. Then, it is important to analyze whether blockchain should be used in all scenarios, according to the importance of each feature for different use cases. Findings Based on analysis of existing systems, Sovrin is the IdM system that covers more features and is based on blockchain. For each of the evaluated use cases, Sovrin and UniquID were the chosen systems. Research limitations/implications This paper opens new lines of research for IdM systems in IoT, including challenges related to device identity definition, privacy preserving and new security mechanisms. Originality/value This paper contributes to the ongoing research in IdM systems for IoT. The adequacy of blockchain is not only analyzed considering the technology; instead the authors analyze its application to real environments considering the required features for each use case.

Abstract

Abstract
Robust wireless communication networks are a cornerstone of the modern world, allowing data to be transferred quickly and reliably. Establishing such a network at sea, a Maritime Internet of Things (MIoT), would enhance services related to safety and security at sea, environmental protection, and research. However, given the remote and harsh nature of the sea, installing robust wireless communication networks with adequate data rates and low cost is a difficult endeavor. This paper reviews recent MIoT systems developed and deployed by researchers and engineers over the past few years. It contains an analysis of short-range and long-range over-the-air radio-frequency wireless communication protocols and the synergy between these two in the pursuit of an MIoT. The goal of this paper is to serve as a go-to guide for engineers and researchers that need to implement a wireless sensor network at sea. The selection criterion for the papers included in this review was that the implemented wireless communication networks were tested in a real-world scenario.