An improved deep convolutional neural network architecture for chromosome abnormality detection using hybrid optimization model.

Chromosomes are thread-like structures located in the cell nucleus that contains the human body blueprint. Chromosome analysis is also known as karyotyping is the test taken to detect the abnormalities identified in the human chromosome. The two types of widely known chromosome abnormalities are structural and numerical abnormalities. Manual karyotyping is complex, time-consuming, and error-prone. To overcome these complexities, automated chromosome karyotype architecture is proposed using the deep convolutional neural network (DCNN) architecture. Training the DCNN architecture from scratch needs a huge dataset and to overcome this problem a generative adversarial networks is used to create adversarial samples that resemble the images in the actual dataset. The time-consuming hyperparameter tuning in the DCNN architecture is overcome using the hybrid moth-flame optimization integrated with the hill-climbing strategy (HMFOHC). The HMFOHC algorithm is mainly utilized in this article to minimize the huge number of parameters associated with the DCNN architecture. The efficiency of the proposed methodology is evaluated using two datasets namely the BioImLab chromosome dataset and hospital dataset. The proposed HMFOHC optimized DCNN architecture is mainly utilized for multiclass classification where it differentiates five numerical chromosome abnormalities, namely Trisomy 13, Trisomy 18, Trisomy 21, Trisomy XXY syndrome, and Monosomy X. The proposed model offers an accuracy, F1-score, and kappa coefficient value of 98.65%, 98.86%, and 0.9894, respectively. The results obtained show that the proposed model achieves higher classification accuracy when compared with the different state-of-art techniques such as deep learning, random forest, and CNN. The inference time of the proposed methodology is 12.5 s which is relatively lower than the state-of-art techniques. The proposed approach can help cytogenetics forensic experts make better decisions and save time by automating manual karyotyping. RESEARCH HIGHLIGHTS: Manual karyotyping is complex, time-consuming, and error-prone. To overcome these complexities, automated chromosome karyotype architecture is proposed using the DCNN architecture The proposed method is mainly applied for multiclass classification where the different types of numerical abnormalities are addressed such as Trisomy 13, Trisomy 18, Trisomy 21, Trisomy XXY syndrome, and Monosomy X.

Florística de trepadeiras em floresta semidecidual e em mata ciliar no vale do rio Araguari, MG / Floristic of climbing plants in semideciduous and riparian forests of the Araguari river, MG

Trepadeiras são importantes componentes de florestas tropicais e contribuem para a composição e riqueza de espécies, porém são poucas as pesquisas que enfocam esse hábito de vida. O objetivo deste trabalho foi determinar a composição florística de trepadeiras em um fragmento de floresta semidecidual e outro de mata ciliar em Uberlândia-MG e comparar essa composição com a de fragmentos de outras fisionomias semelhantes. O levantamento florístico foi realizado mensalmente durante os anos de 2006 e 2007 por meio de caminhadas aleatórias. As espécies coletadas foram categorizadas segundo o hábito trepador, modo de ascensão, síndrome de dispersão e frequência de ocorrência. Para determinar as similaridades florísticas foi utilizado o método UPGMA com o índice de Jaccard. Nos dois fragmentos foram amostradas 62 espécies, 41 gêneros e 20 famílias. Na floresta semidecidual e mata ciliar foram encontradas 39 e 33 espécies, respectivamente. As famílias mais ricas em espécies foram Malpighiaceae (9 spp.), Bignoniaceae (8 spp.) e Convolvulaceae (7 spp.) e os gêneros mais ricos foram Fridericia Mart. (5 spp.), Ipomoea L. (5 spp.) e Serjania Mill. (4 spp.). Foi registrado maior número de trepadeiras lenhosas do que não-lenhosas e houve predominância da estratégia de ascensão volúvel e da dispersão anemocórica tanto na floresta semidecidual como em mata ciliar. A similaridade florística entre os nove fragmentos analisados foi baixa, com uma composição de espécies bastante heterogênea, mesmo entre áreas mais próximas, indicando uma elevada diversidade ß de trepadeiras nas áreas estudadas.

Climbers are important components of tropical forests and contribute to their species composition and richness. However only few studies have focused on this specific growth habit. The aim of this study was to determine the floristic composition of climbers in a riparian and a semideciduous forests in Uberlândia-MG, and compare these floras with other forest areas. The floristic survey was conducted monthly during 2006 and 2007 using random walks. The species collected were categorized according to the climbing habit, climbing strategy, dispersal syndrome and frequency of occurrence. Jaccard's similarity index with UPGMA was used to determine the floristic similarities. In both sites the sampling totaled 62 species, in 41 genera and 20 families. In semideciduous and riparian forests we found 39 and 33 species, respectively. The richest families in species were Malpighiaceae (9 spp.), Bignoniaceae (8 spp.) and Convolvulaceae (7 spp.) and the richest genera were Fridericia Mart. (5 spp.), Ipomoea L. (5 spp.) and Serjania Mill. (4 spp.). It was recorded a higher number of woody climbing than vines. The predominant strategies were twining-climbing mechanism and wind dispersal in both semideciduos and riparian forests. The floristic similarity between nine fragments analyzed was low, with species composition very heterogeneous, even among the closest areas, indicating a high ß diversity of climbers in these areas.