RESUMO
Objective.Automated cell nuclei segmentation is vital for the histopathological diagnosis of cancer. However, nuclei segmentation from 'hematoxylin and eosin' (HE) stained 'whole slide images' (WSIs) remains a challenge due to noise-induced intensity variations and uneven staining. The goal of this paper is to propose a novel deep learning model for accurately segmenting the nuclei in HE-stained WSIs.Approach.We introduce FEEDNet, a novel encoder-decoder network that uses LSTM units and 'feature enhancement blocks' (FE-blocks). Our proposed FE-block avoids the loss of location information incurred by pooling layers by concatenating the downsampled version of the original image to preserve pixel intensities. FEEDNet uses an LSTM unit to capture multi-channel representations compactly. Secondly, for datasets that provide class information, we train a multiclass segmentation model, which generates masks corresponding to each class at the output. Using this information, we generate more accurate binary masks than that generated by conventional binary segmentation models.Main results.We have thoroughly evaluated FEEDNet on CoNSeP, Kumar, and CPM-17 datasets. FEEDNet achieves the best value of PQ (panoptic quality) on CoNSeP and CPM-17 datasets and the second best value of PQ on the Kumar dataset. The 32-bit floating-point version of FEEDNet has a model size of 64.90 MB. With INT8 quantization, the model size reduces to only 16.51 MB, with a negligible loss in predictive performance on Kumar and CPM-17 datasets and a minor loss on the CoNSeP dataset.Significance.Our proposed idea of generalized class-aware binary segmentation is shown to be accurate on a variety of datasets. FEEDNet has a smaller model size than the previous nuclei segmentation networks, which makes it suitable for execution on memory-constrained edge devices. The state-of-the-art predictive performance of FEEDNet makes it the most preferred network. The source code can be obtained fromhttps://github.com/CandleLabAI/FEEDNet.
Assuntos
Processamento de Imagem Assistida por Computador , Redes Neurais de Computação , Núcleo Celular , Amarelo de Eosina-(YS) , Hematoxilina , Processamento de Imagem Assistida por Computador/métodosRESUMO
Background: The prevalence of protein-energy malnutrition in surgical patients is seen in 30-50% of surgical patients with gastrointestinal disease. Malnutrition and hypalbuminaemia contribute to increased morbidity and mortality. Serum albumin level is a low-cost key element in nutritional assessment. Clavien-Dindo classification of post-operative complications enables stratification of post-operative complication.Methods: We retrospectively analysed the prospectively maintained data of 136 patients who underwent major gastro intestinal procedures from October 2019 to March 2020. We assessed the pre-operative; intra operative parameters, outcome variables and the postoperative complications were graded according to Clavien-Dindo severity, length of ICU stay, length of hospitalization and mortality.Results: Out of the 136 patients (M: F=3:2), the procedures were hepatopancreaticobiliary (n=40), colorectal (n=39), small bowel (n=36) and esophago-gastric (n=19). Pre-operative serum albumin was identified to be single most statistically significant pre-operative variable predicting post-operative complication of Clavien-Dindo severity grade III and above. The ROC curve of the serum albumin level predicting the severe post-operative complication suggested optimal cut off value of 3.1 gm/dl (AUC=0.76; 5% CI=0.64-0.87; p<0.001). Linear regression analysis of serum albumin level predicting the severe post-operative complication suggested good correlation with (r2=0.133; b=0.689; p<0.001). Further analysis of serum albumin level in predicting the length of ICU and the hospital stays suggested a significant negative correlation with both of these dependent outcome variables. The serum albumin level correlated inversely with the length of ICU stay and hospitalization.Conclusions: Low pre-operative albumin (serum level <3.1 gm/dl) in patients undergoing major gastrointestinal surgery predicts severe post-operative complications, prolonged ICU and hospital stays.
RESUMO
Background: The preference for a son over female offspring continues to be a prevailing norm within the ancient Indian family. Sons are expected to work within the fields, give bigger financial gain and look after parents. Additionally, in Indian patriarchal society, sons are a unit responsible for the preservation of the family name. The preference for a son and female feticide continues to be a prevailing norm in ancient Indian families (urban and rural) leading to declining sex ratio to an alarming level. With this background the following study was conducted in Gautam Buddha Nagar with the objective of finding out gender bias in rural and urban area.Methods: A study was carried out in rural and urban field practice area of the Department of Community Medicine, district Gautam Buddha Nagar of Uttar Pradesh. The study was conducted among 270 married women of reproductive age group. The women were interviewed using a semi structure questionnaire.Results: More than half of the respondents in the rural areas (57.7%) preferred son while in the urban areas, 43% of women preferred a son. The most important reason for preferring a son in the rural area was a support to the parents. While in urban area son was preferred because of the continuation of a family line.Conclusions: Preference for son was found to be high. Further amendments in the existing laws and quality education may help in breaking gender stereotypes and minimizing gender bias.
RESUMO
Background: Visual health is both visual function and individual’s perception of vision. This study was conducted to estimate visual function and individual’s perception of vision and to understand the impact of sociodemographic factors and treatment seeking behaviour for eye morbidities among women in our society. The aim of the study was to obtain visual acuity scores and the perceived visual function score and analyse both objective and subjective visual health.Methods: A community based cross-sectional study was conducted among 182 women aged 18-59 years residing in rural practice area of School of Medical Sciences and Research, Sharda University, Greater Noida.Results: The most common problem in the study group was moderate visual impairment (23.1%), followed by severe visual impairment (8.2%) and blindness (1%). The common causes being uncorrected refractive errors followed by cataract. Visual impairment was maximum among 49-59 years (47.46%), illiterate (54.24%) and in lower socio-economic status (54.24%).Conclusions: An intersectoral approach involving the PRI, ASHA/ANM at all levels of the health system may be appropriate. Strategies involving the communities, schools, colleges, workplaces and shifting to preventive and curative approach instead of only curative approach may help to create more awareness about common avoidable causes of blindness namely refractive errors and cataract. Comprehensive eye care services from early screening to treatment need to be made accessible, affordable and available at all levels of health system irrespective of gender, literacy, age and socioeconomic status to improve visual health and thus quality of life.
RESUMO
The case of a 9-year-old boy with hemangioendothelioma of the liver and spleen who presented with consumptive coagulopathy one month after sustaining a blunt trauma to his abdomen is reported. A contrast enhanced computed tomography scan of the abdomen showed a ruptured spleen with multiple lesions in the liver that were enhancing with contrast. On exploration, the child was found to have splenic rupture with multiple vascular lesions of the liver. A splenectomy with liver biopsy was done. The histopathologic examination found that both the liver and spleen had a similar tumor morphology characteristic of an epithelioid and spindle cell hemangioendothelioma. The child ultimately died of relentless consumptive coagulopathy.