In this paper, we describe our method for the ISIC 2019 Skin Lesion Classification Challenge. The challenge comes with two tasks. For task 1, skin lesions have to be classified based on dermoscopic images. For task 2, dermoscopic images and additional patient meta data have to be used. A diverse dataset of 25 000 images was provided for training, containing images from eight classes. The final test set contains an additional, unknown class. We address this challenging problem with a simple, data driven approach by including external data with skin lesions types that are not present in the training set. Furthermore, multi-class skin lesion classification comes with the problem of severe class imbalance. We try to overcome this problem by using loss balancing. Also, the dataset contains images with very different resolutions. We take care of this property by considering different model input resolutions and different cropping strategies. To incorporate meta data such as age, anatomical site, and sex, we use an additional dense neural network and fuse its features with the CNN. We aggregate all our models with an ensembling strategy where we search for the optimal subset of models. Our best ensemble achieves a balanced accuracy of 74.2 % using five-fold crossvalidation. On the official test set our method is ranked first for both tasks with a balanced accuracy of 63.6 % for task 1 and 63.4 % for task 2.
Introduction: Visceral leishmaniasis (VL) is one of the fatal parasitic diseases, currently threatening 500,000 new cases worldwide each year. The major front line drugs available for treating VL are toxic and develop resistance due to their long treatment regimen. With this assumption, present work has been designed to formulate miltefosine (MF) as microparticulate drug delivery system to target macrophages where leishmania parasite resides. Methodology: MF is formulated using albumin as carrier and spray drying as method for the preparation of microparticles of desired size (£6 mm) for intravenous administration. Microspheres were characterized morphologically and evaluated for particle size, product yield, and for drug polymer interaction. The ability of macrophages to phagocytosis these microspheres were investigated in RAW 264.7 cell line. In vitro hemolysis study was carried out to determine hemotoxicity of prepared microsphere formulation. Results: The average size of the prepared microspheres were found to be nearly 3 mm with most of particles in the range of 2-5 mm. Differential scanning calorimetry (DSC) and Fourier transform infrared spectroscopy (FTIR) confirmed no significant chemical interaction between the albumin and MF. Uptake study in macrophage cell line confirmed the targetability of prepared microspheres to macrophages. In vitro hemolysis study indicated reduced hemotoxicity and suitability of prepared microspheres for parenteral administration. Conclusions: MF loaded microspheres of desired size were prepared successfully using spray drying method with thermal stabilization. In vitro hemolysis and uptake studies confirmed the suitability of prepared microspheres for targeting macrophages with reduced hemotoxicity.
Software Development is a complex and multidimensional task. Often software development faces serious problems of meeting key constraints of cost and time. Big projects which are well planned and analyzed, can end up in a disaster because of mismanagement in cost estimation and time allocation. Program slicing has unique importance in addressing the issues of cost and time. It is broadly applicable static program analysis technique which provides mechanism to analyze and understand the program behavior for further restructuring and refinement. In this paper, authors investigate the relationship between program slicing and software development phases on the basis of empirical studies conducted in the past and also establish the fact that how program slicing can be helpful in making software system cost and time effective.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.