Sharif Digital Repository

We present a deep neural network to classify dermatological disease from patient images. Using self-supervised learning method we have utilized large amount of unlabeled data. We have pre-trained our model on 27000 dermoscopic images gathered from razi hospital, the best dermatological hospital in Iran, along with 33000 images from ISIC 2020 dataset. We have evaluated our model performance in semi-supervised and transfer learning approaches. Our experiments show that using this approach can improve model accuracy and PRC up to 20 percent on semi-supervised setting. The results also show that pretraining can improve classification PRC up to 20 percent on transfer learning task on HAM10000... 

Industrial enzymes can generally be classified into three groups: carbohydrases, proteases and lipases, which account for a major part of the global enzyme trade. Proteases, in particular, have always been extensively studied and investigated due to their widespread use in various industries, such as food, pharmaceutical, chemical, and agricultural sectors. Proteases are a group of proteolytic enzymes that break down proteins and peptides into their constituent amino acids by hydrolyzing peptide bonds. Among the family of proteases, alkaline serine proteases have special biocatalytic abilities in both aqueous and organic environments. Their suitable resistance in alkaline environments and... 

Polymer flooding is a famous enhanced oil recovery method and is used to reduce mobility ratio of displacing fluid to displaced fluid by forming a stable front in displacing process. Due to high costs, polymer flooding faces economical risks and from this point of view, performance investigation of polymer flooding in laboratory scale was concerned. In this project, performance of 7 types of polymers which have different rheological properties were investigated in fractured/non-fractured carbonated and sandstone reservoirs. These polymers consists of 4 commercial polymers, one branched polymer, one synthetic polymer and one polymer composite which have different molecular weight,... 

Cancer treatment is one of the most challenging issues of recent years. Extensive researches have been conducted in various fields of medicine, pharmacy, and medical physics in this regard. Today, considering nanotechnology developments, researchers believe that full cancer treatment is possible through the use of nano-systems designed with scientific background to overcome the defects of previous methods. In this research, the main purpose is to design and synthesize multifunctional hybrid nanostructures with favorable optical and magnetic properties providing a way to combine chemotherapy and thermal treatment. Accordingly, two different structures were designed and experimentally... 

PMMA bone cement lacks biodegradability and the ability to bond with surrounding bone tissue. Therefore, the development of a new generation of bioactive bone cements that are biodegradable and possess adequate mechanical properties as well as desirable setting time is receiving remarkable interest.In this study, we have developed novel mineral-based bioactive bone cements. Our mineral bioactive bone cements were composed of Calcium Sulfate Hemihydrate, Bioactive Glass, and Tricalcium Silicate. Firstly, a binary system composed of Calcium Sulfate Hemihydrate and Bioactive Glass was optimized based on mechanical and setting behavior. Secondly, Tricalcium Silicate was added to the powder phase... 

Membrane processes are accounted as the newest technology for gas separation and nowadays are more applicable in different petroleum, gas and petrochemical industries because of abundant advantages. One of important application of membrane gas separation technology is syngas adjustment and hydrogen enrichment. In this research, the appropriate alumina nanocomposite membranes for this purpose are prepared and their gas separation performance is investigated. Therefore the mixed matrix membranes, containing the continues polymeric phase polyethersulfone and dispersed phase alumina nanoparticles is synthesized by using methods of experimental design. As the best of our knowledge, there is no... 

In the upstream asphaltene flow assurance community, both academics and industries are actively involved to predict the asphaltene deposit profile in wellbores and pipelines. The main objective of this study is to contribute to the thermodynamic and transport modeling of asphaltene in order to predict its precipitation, aggregation and deposition. For a better knowledge of the flow properties of asphaltene containing crude oils, it is necessary to understand how asphaltene affect the rheological properties. In this work, the information on such rheological properties of oil-asphaltene slurry systems is provided. The results of rheological experiments for two Iranian crude oils show that the... 

This thesis discusses the work on the production of a rhamnolipid-type biosurfactant by Pseudomonas aeruginosa MR01, its structural characterization and environmental applications. First, results from analytical approaches for screening of biosurfactant producers suggested the oil spreading method as the fastest, simplest and most consistent analytical method. The second part of the thesis describes the properties of the most efficient biosurfactant-producing bacteria isolated from oil fields in south of Iran and according to it’s biochemical characteristics and partial sequenced 16S rRNA gene affiliated with Pseudomonas aeruginosa. Time course study indicated that the maximum biosurfactant... 

The conventional approach to investigate the reservoir performance is to build a detailed geological model, upscale it, and finally run flow simulation which is computationally very expensive. In addition, during the early stage of life of a reservoir, due to the lake of certain data, this method is usually based on analogues or rules of thumb and not detailed reservoir modelling. Therefore, there is a great incentive to produce much simpler physically-based methodologies. The main focus of this thesis is to use percolation approach to estimate the uncertainty in the reservoir properties. This method considers a hypothesis that the reservoir can be split into either permeable (i.e....