Re-evaluation of a reported increased micronucleus frequency in lymphocytes of workers occupationally exposed to formaldehyde

A replicate evaluation of increased micronucleus (MN) frequencies in peripheral lymphocytes of workers occupationally exposed to formaldehyde (FA) was undertaken to verify the observed effect and to determine scoring variability. May–Grünwald–Giemsa-stained slides were obtained from a previously performed cytokinesis-block micronucleus test (CBMNT) with 56 workers in anatomy and pathology laboratories and 85 controls. The first evaluation by one scorer (scorer 1) had led to a highly significant difference between workers and controls (3.96 vs 0.81 MN per 1000 cells). The slides were coded before re-evaluation and the code was broken after the complete re-evaluation of the study. A total of 1000 binucleated cells (BNC) were analysed per subject and the frequency of MN (in ‰) was determined. Slides were distributed equally and randomly between two scorers, so that the scorers had no knowledge of the exposure status. Scorer 2 (32 exposed, 36 controls) measured increased MN frequencies in exposed workers (9.88 vs 6.81). Statistical analysis with the two-sample Wilcoxon test indicated that this difference was not significant (p = 0.17). Scorer 3 (20 exposed, 46 controls) obtained a similar result, but slightly higher values for the comparison of exposed and controls (19.0 vs 12.89; p = 0.089). Combining the results of the two scorers (13.38 vs 10.22), a significant difference between exposed and controls (p = 0.028) was obtained when the stratified Wilcoxon test with the scorers as strata was applied. Interestingly, the re-evaluation of the slides led to clearly higher MN frequencies for exposed and controls compared with the first evaluation. Bland–Altman plots indicated that the agreement between the measurements of the different scorers was very poor, as shown by mean differences of 5.9 between scorer 1 and scorer 2 and 13.0 between scorer 1 and scorer 3. Calculation of the intra-class correlation coefficient (ICC) revealed that all scorer comparisons in this study were far from acceptable for the reliability of this assay. Possible implications for the use of the CBMNT in human biomonitoring studies are discussed.

Tipificação de fundações de torres eólicas em parques industriais, para diversos tipos de solos

Dissertação para obtenção do grau de Mestre em Engenharia Civil na Área de Especialização de Edificações

Paired indices for clustering evaluation correction for agreement by chance

In the present paper we focus on the performance of clustering algorithms using indices of paired agreement to measure the accordance between clusters and an a priori known structure. We specifically propose a method to correct all indices considered for agreement by chance - the adjusted indices are meant to provide a realistic measure of clustering performance. The proposed method enables the correction of virtually any index - overcoming previous limitations known in the literature - and provides very precise results. We use simulated datasets under diverse scenarios and discuss the pertinence of our proposal which is particularly relevant when poorly separated clusters are considered. Finally we compare the performance of EM and KMeans algorithms, within each of the simulated scenarios and generally conclude that EM generally yields best results.

Comparing clustering solutions: the use of adjusted paired indices

In the present paper we compare clustering solutions using indices of paired agreement. We propose a new method - IADJUST - to correct indices of paired agreement, excluding agreement by chance. This new method overcomes previous limitations known in the literature as it permits the correction of any index. We illustrate its use in external clustering validation, to measure the accordance between clusters and an a priori known structure. The adjusted indices are intended to provide a realistic measure of clustering performance that excludes agreement by chance with ground truth. We use simulated data sets, under a range of scenarios - considering diverse numbers of clusters, clusters overlaps and balances - to discuss the pertinence and the precision of our proposal. Precision is established based on comparisons with the analytical approach for correction specific indices that can be corrected in this way are used for this purpose. The pertinence of the proposed correction is discussed when making a detailed comparison between the performance of two classical clustering approaches, namely Expectation-Maximization (EM) and K-Means (KM) algorithms. Eight indices of paired agreement are studied and new corrected indices are obtained.