Classification of colorectal tissue images from high throughput tissue microarrays by ensemble deep learning methods

Tissue microarray (TMA) core images are a treasure trove for artificial intelligence applications. However, a common problem of TMAs is multiple sectioning, which can change the content of the intended tissue core and requires re-labelling. Here, we investigate different ensemble methods for colorectal tissue classification using high-throughput TMAs. Hematoxylin and Eosin (H&E) core images of 0.6 mm or 1.0 mm diameter from three international cohorts were extracted from 54 digital slides (n = 15,150 cores). After TMA core extraction and color enhancement, five different flows of independent and ensemble deep learning were applied. Training and testing data with 2144 and 13,006 cores included three classes: tumor, normal or “other” tissue. Ground-truth data were collected from 30 ngTMA slides (n = 8689 cores). A test augmentation is applied to reduce the uncertain prediction. Predictive accuracy of the best method, namely Soft Voting Ensemble of one VGG and one CapsNet models was 0.982, 0.947 and 0.939 for normal, “other” and tumor, which outperformed to independent or ensemble learning with one base-estimator. Our high-accuracy algorithm for colorectal tissue classification in high-throughput TMAs is amenable to images from different institutions, core sizes and stain intensity. It helps to reduce error in TMA core evaluations with previously given labels.

Slope stability analysis of the drainage tunnel portal in Tanju Dam, Dompu District, West Nusa Tenggara

This paper presents design results of the tunnel portal slopes at the Tanju Dam, Dompu, West Nusa Tenggara. The objective of this research was to analyse the stability of the tunnel portal slopes using circular failure chart (CFC) method, limit equilibrium method (LEM), and finite element method (FEM). Input parameters were obtained from drill core evaluations and laboratory tests. By considering the rock mass rating (RMR) values of rock masses, which are categorized as class II, at the two slopes, adjustments for the cohesion and inner friction angle values are made. The inlet slope (IL) have cohesion values of 350 kPa and 40º inner friction angle and the outlet slope (OL) have cohesion values of 400 kPa and 45º inner friction angle. The CFC method shows that the IL and OL have safety factor (FS) values of 3.5 and 3.44, respectively. The LEM shows that the IL and OL have the FS values of 3.69 and 3.65, respectively. Meanwhile, the FEM shows that the IL and OL have FS values of 4.78 and 4.79, respectively. The stability analysis results indicate that designed slopes are stable.

Candidates’ attractiveness in selection decisions: a laboratory experiment

Purpose Recruiters in today’s organizations, through social networks, have the opportunity to see a candidate’s overall figure, and from this, they gain a first impression of their personalities which in turn affects their decisions. The purpose of this paper is to investigate the biasing role of candidates’ attractiveness – in facial and bodily terms – on perceived main personality features (i.e. core evaluations (CEs)) in selection decisions. Design/methodology/approach An experimental study involving professional recruiters (n=150) was conducted. Participants were asked to rate bodily attractiveness (using the objectification construct), CEs, facial attractiveness and hiring scores of six candidates for an administrative position; then, a moderated mediation model was tested. Findings This study suggests that recruiters’ perception of candidates’ CEs mediates the relationship between objectification (i.e. body attractiveness) and the assigned hiring score, while facial attractiveness amplifies or reduces the effect of objectification on CEs. Originality/value The value added of this contribution lies in studying the biasing mechanism of candidates’ overall attractiveness (facial and bodily) and its effects on the perceived core personality features.

Core Self-Evaluations and Individual Strategies of Coping with Unemployment among Displaced Spanish Workers

Unemployment has negative but also positive effects on mental health and general well-being depending on which coping strategies the individual use. Our aim was to determine the contribution of core self-evaluations in explaining the coping strategies of job search and job devaluation, as well as to test the potential moderation effect of job search and mediation effect of job devaluation on the relationship between self core-evaluations and both positive and negative experience of unemployment. One hundred seventy-eight individuals who lost their jobs involuntarily for a longer period than one month completed a questionnaire while attending to employment office. Results show that there is a significant relation between core-self evaluations and job devaluation (.37**). Furthermore, core-self evaluations were positively related to positive experience of unemployment (r = .31; p < .01) and negatively related to negative experience of unemployment (r = .60; p < .01). Moreover, self-core evaluations predicted both coping with unemployment strategies (job devaluation; β = .26; p < .01 and job search β = .19; p < .05). However, job search did not moderate the relationship between core self-evaluations and experience of unemployment. But, individuals with a longer duration of the current period of unemployment and higher core self-evaluations had a more positive experience of unemployment, and job devaluation partially mediated this relation (SE = .002; p = .038). These results imply that programs interventions should include the improvement of core self-evaluations and the positive experience of unemployed people.