China
Africa
Explore chapters and articles related to this topic
Application of Artificial Intelligence in Thyroidology
Published in Sandeep Reddy, Artificial Intelligence, 2020
Lack of large medical image datasets makes it harder to use supervised learning. Yang et al. used GANs to address this shortcoming by using dual-path semi-supervised conditional generative adversarial networks for classification of thyroid nodules in ultrasound (Yang et al., 2019). Another application of GAN is the virtual staining of histopathology slides. Traditionally, staining of medical tissue is a cumbersome and time-consuming process. Rivenson et al. used GAN to create a label-free virtual staining method for tissues including thyroid (Rivenson et al., 2019). GANs can also be used to train label generating AI algorithms. For example, ultrasound images of thyroid nodules generated by GAN can be tagged with different features likes microcalcification, irregular border and hypoechogenicity. These tagged pictures could be used to create an AI model to generate tags for unseen thyroid ultrasound images. Figure 15.1 depicts synthetic ultrasound thyroid nodule images created by the author.
Robust Nuclei Segmentation using Statistical Level Set Method with Topology Preserving Constraint
Published in Ayman El-Baz, Jasjit S. Suri, Level Set Method in Medical Imaging Segmentation, 2019
Shaghayegh Taheri, Thomas Fevens, Tien D. Bui
In order to improve the visualization of cell and tissue sections and highlight the desired structures, biological specimens are usually stained with a certain number of dyes. The most widely used staining protocol is the combination of Hematoxylin and Eosin (H&E). Hematoxylin stains nuclei blue, whereas Eosin is employed to stain red blood cells, cytoplasm and extracellular structures magenta or red. Whole slide digital scanners currently available use the Red-Green-Blue-based (RGB-based) imaging sensors, thus a stain separation method is needed to calculate the contribution of each stain. Ruifrok and Johnston [32] proposed an unmixing algorithm for up to three stains that reveals uncorrelated information about stain concentration even when the stains have overlapping spectral absorption.
Laboratory tutorials
Published in Firdos Alam Khan, Biotechnology Fundamentals, 2018
Staining: Hematoxylin and eosin (H&E) is the most commonly used light microscopical stain in histology and histopathology. Hematoxylin stains nuclei blue; eosin stains the cytoplasm pink. Uranyl acetate and lead citrate are commonly used to impart contrast to tissue in the electron microscope. There are hundreds of various other techniques that have been used to selectively stain cells and cellular components. Other compounds used to color tissue sections include safranin, oil red o, Congo red, fast green fungal culture filtrate (FCF), silver salts, and numerous natural and artificial dyes that largely originated from the development dyes for the textile industry. Histochemistry refers to the science of using chemical reactions between laboratory chemicals and components within tissue. A commonly performed histochemical technique is the Perls’ Prussian blue reaction, used to demonstrate iron deposits in diseases like hemochromatosis.
Removal of Malachite Green using carbonized material derived from disposable facemasks: optimization of removal process through Box–Behnken design
Published in Environmental Technology, 2023
Subramanian Srinivasan, R.S. Kaarmukhilnilavan, Kumarasamy Murugesan
Textile, food, pharmaceutical, paper printing, leather and cosmetic sectors are the main sources of dyes [3–5]. The majority of dyes are discharged into the environment due to the inefficacy of dye molecules for complete attachment to materials [6,7]. It is very complicated to degrade dye by nature, and their presence in water bodies pollutes natural environment and causes health problems [8]. The presence of trace amount of hazardous dyes in water causes a significant environmental impact. For example, Malachite Green (MG) a cationic dye used in textile and aquaculture as a fungicide and Congo Red (CR), a diazo anionic dye used in textile dyeing (Figure 1a and b). MG is a toxic dye utilised not only for product coloration in the textile, paper, rubber and leather sectors [9] but also as an antiprotozoan, fungicide in fisheries. However, referral studies had revealed that MG is toxic (at 0.05 mg/L) to freshwater species in both acute and chronic exposure. Similarly, Congo red is another anionic azo dye mainly used in textile industries for staining silk and cotton cloths. It is also used as a staining agent in biological laboratories to visualize cells. Studies have proven that these dyes can produce serious carcinogenic and toxic effects in mammalian cells and organs such as kidney, lung, spleen and skin [10,11]. Moreover, these synthetic dyes deteriorate the water quality. Hence, for the health and environmental concerns, the removal of dyestuffs from coloured effluent is a primary process in textile wastewater treatment processes [12].
Classification and recognition method of white blood cells subclasses in batches based on phase characteristics with non-orthogonal phase imaging
Published in Journal of Modern Optics, 2022
Yuanyuan Xu, Hao Han, Yang Zou, Yawei Wang, Jingrong Liao
In order to illustrate the advantages and application of the proposed method, we compare it with some methods in theory. Compared with some conventional fast detection flow scattering techniques with fast detection speed, the proposed method can not only realize cell recognition and counting, but also understand the morphology and structure through phase map. Because the flow scattering technology uses the statistical method for calculation, it is difficult to accurately image of each cell. Compared with the microscopic examination technology, the proposed method is a labelling free method without staining and has the advantage of high-speed measurement. Because the microscopic examination usually requires staining, which is a time-consuming process, and the staining may affect the behaviour of cells. Compared with other phase imaging methods, the proposed method does not need refractive index tomography which needs a few phases, so its measurement rate is fast. Because the proposed method only requires two non-orthogonal phases and can know some information related to the morphology and structure of cells by calculating their phase characteristics. In short, the proposed method not only provides a new idea and theoretical basis for the detection and diagnosis of blood cells, but also provides a prototype for the replacement of the precision and high-end medical diagnostic instruments.
Dual Stage Normalization Approach Towards Classification of Breast Cancer
Published in IETE Journal of Research, 2022
Biopsy is a surgical process, where a tissue is removed from any part of the body and examined for the existence, reason, and scope of the cancer. This process is usually carried out in a histology laboratory by a pathologist [9]. Histopathology refers to the process of studying specimen of tissue seen with the help of microscope for the detection of cancer [10]. With the onset of digital pathology, the biomedical experts can help in the detection of cancer by applying image processing methods. The spectrum is very large that we can use automated techniques from the start to end of the biopsy process [11–13]. To anticipate features of the breast tissue, immunohistochemical stains are used. Different types of staining exist. Hematoxylin–Eosin (H&E) staining is a popular staining technique to visualize the structures in tissue samples. H&E staining represents histological structures in a logical and distinct way. Tissue nuclei will be colored as blue or violet and eosinophilic texture will be colored as red or pink [14]. These colored specimens are examined by the pathologist under microscope with a charge coupled device (CCD) camera.