AI outperformed every dermatologist in dermoscopic melanoma diagnosis, using an optimized deep-CNN architecture with custom mini-batch logic and loss function

Melanoma, one of the most dangerous types of skin cancer, results in a very high mortality rate. Early detection and resection are two key points for a successful cure. Recent researches have used artificial intelligence to classify melanoma and nevus and to compare the assessment of these algorithms to that of dermatologists. However, training neural networks on an imbalanced dataset leads to imbalanced performance, the specificity is very high but the sensitivity is very low. This study proposes a method for improving melanoma prediction on an imbalanced dataset by reconstructed appropriate CNN architecture and optimized algorithms. The contributions involve three key features as custom loss function, custom mini-batch logic, and reformed fully connected layers. In the experiment, the training dataset is kept up to date including 17,302 images of melanoma and nevus which is the largest dataset by far. The model performance is compared to that of 157 dermatologists from 12 university hospitals in Germany based on the same dataset. The experimental results prove that our proposed approach outperforms all 157 dermatologists and achieves higher performance than the state-of-the-art approach with area under the curve of 94.4%, sensitivity of 85.0%, and specificity of 95.0%. Moreover, using the best threshold shows the most balanced measure compare to other researches, and is promisingly application to medical diagnosis, with sensitivity of 90.0% and specificity of 93.8%. To foster further research and allow for replicability, we made the source code and data splits of all our experiments publicly available.

From the Balanced Measure of Psychological Needs (BMPN) to Employee Engagement: Indicators that Matter

The purpose of this study was to investigate if confidence, interest, authenticity, and loneliness as independent variables, could help predict employee engagement, the dependent variable. In this setting, the independent variables were indicators of the Balanced Measure of Psychological Needs (BMPN) and the dependent variable was obtained using the Utrecht Work Engagement Scale (UWES-9). After surveying 151 participants in the United States, 17 responses were removed from the final dataset during data and assumptions validation. A multiple regression model was created using the remaining 134 valid cases. Our findings confirmed the existence of a statistically significant relationship between confidence, interest, and authenticity in predicting employee engagement. Only, we could not establish a statistically significant relationship between loneliness and engagement, in contrast to some prior research studies. These findings have significant implications for practitioners and researchers as documented in this article. For example, the findings can be useful for employees in determining their future career path, as they need to first look at what interests them. Indeed, interest was identified as the greatest determinant of engagement as compared to the other three predictors. These findings also suggest that managers can keep their employees engaged by assigning them functions or tasks that are aligned with their interests.

Chasing criminal wealth: broken expectations for the criminalization of illicit enrichment in Lithuania

Purpose This study aims at elaborating whether the criminalization of illicit enrichment has the potential to be an efficient and well-balanced measure against profiting from serious crime. Design/methodology/approach This study offers a normative analysis of the concept of the criminalization of illicit enrichment, particularly from the perspective of the presumption of innocence. This paper supplements theoretical considerations using Lithuania, where illicit enrichment was criminalized a decade ago, as a case study. It analyses data of all 28 known criminal cases on illicit enrichment that resulted in judgements in Lithuania in 2015–2019. Findings The author concludes that on neither a fundamental nor practical level can the assumption that the criminalization of illicit enrichment efficiently carries out the task for which it was created be supported. Moreover, efforts to implement this legal strategy might unbalance the system of illicit asset recovery measures and obstruct the implementation of other promising legal instruments. Originality/value As very few states with developed democracy and strong rule of law traditions have implemented criminalization of illicit enrichment, there are very few if any empirical data or analysis of practices on this issue that could contribute to the theoretical discussion. This paper aims at contributing to fulfillment of this gap by presenting relevant data and insights from the perspective of Lithuanian criminal justice system.

Agility: an essential element of leadership for an evolving educational landscape

Defined as the ability to think and move quickly and easily, the importance of agility as an essential element in the move forward for leaders of schools and systems postpandemic, as a result of the impact of COVID-19 on children, is examined. The smartness of a leader’s continuous interactions with the multi-faceted features of their environment, the very nature of the ever-evolving educational landscape of today, is of tremendous value for the leadership of tomorrow. Through the prioritization of strategic objectives in balanced measure, connectivity through relationships and partnership building, proactivity for effective change management, ingenuity in the optimization of resources over time, and the cultivation of systemness throughout the organization—as aspects of agility—educational leaders have the bona fide chance of a lifetime to transform school systems in the pursuit of achievement, equity, and well-being for the benefit of all students, staff, and school communities. Additional considerations, including barriers to agility, are also addressed as are recommendations for leaders of schools and systems as they navigate the shifts in organizational terrain caused by the disruption.

The Structure of Need Fulfillment

Self-Determination Theory predicts that fulfillment of the three psychological needs for autonomy, competence, and relatedness predicts well-being. Fulfillment of these needs has long been considered a uni-dimensional construct consisting of need satisfaction and (reverse coded) need dissatisfaction. Recent research suggests that satisfaction and dissatisfaction should be separated. We tested whether need satisfaction and dissatisfaction can be distinguished psychometrically and whether they have unique effects in predicting well-being. We used data from a daily-diary study of 135 participants over the course of 42 days. A six factor solution (with one satisfaction and one dissatisfaction factor per need) for the Balanced Measure of Psychological Needs scale (BMPN) fitted the data best at both the between-person and the within-person level of analysis. We concluded that (a) the BMPN can be used to reliably assess satisfaction and dissatisfaction of the three needs specified by Self-Determination Theory; (b) need satisfaction and dissatisfaction can and should be separated psychometrically; (c) these findings hold at both the between-person and the within-person level of analysis; (d) all three needs predict well-being at the within-person level, but only competence and relatedness predict well-being at the between-between-person level; and (e) need satisfaction and dissatisfaction predict unique variance in well-being.

Virtual Socialization among College Students: An Analysis through Self-Esteem, Psychological Needs and Narcissism

The objective of the present study is to determine whether there is a relationship between self-esteem, psychological needs and narcissism among college students who are frequent and infrequent virtual socializers. A non-probability purposive sampling technique was employed to select 200 college students (100 boys and 100 girls), aged between 18-25 years. An Online Social Networking Scale, Rosenberg’s Self-Esteem scale (Rosenberg, 1965), Balanced Measure of Psychological Needs Scale (Sheldon, 2012) and Narcissism Personality Inventory (Raskin and Terry, 1988) were administered to measure the frequency of virtual socialization, level of self-esteem, psychological needs and narcissism respectively. Significant differences were found between frequent and infrequent virtual socializers with respect to the dimensions of self-deprecation, authority, self-sufficiency and exhibitionism (p<0.05). Significant gender differences were also found with respect to the exhibitionism and vanity dimensions of narcissism (p<0.05). The dimensions of self-esteem, psychological needs and narcissism were significantly correlated in both the groups (p<0.05). In a technologically advanced time such as the present era, it is imperative to be aware of the influence of virtual socialization on behavioral traits of narcissism and self-esteem in college students and the effect on the creation of their concepts of themselves and how they protect their self-esteem and issues of identity.