A deep network designed for segmentation and classification of leukemia using fusion of the transfer learning models

White blood cells (WBCs) are a portion of the immune system which fights against germs. Leukemia is the most common blood cancer which may lead to death. It occurs due to the production of a large number of immature WBCs in the bone marrow that destroy healthy cells. To overcome the severity of this disease, it is necessary to diagnose the shapes of immature cells at an early stage that ultimately reduces the modality rate of the patients. Recently different types of segmentation and classification methods are presented based upon deep-learning (DL) models but still have some limitations. This research aims to propose a modified DL approach for the accurate segmentation of leukocytes and their classification. The proposed technique includes two core steps: preprocessing-based classification and segmentation. In preprocessing, synthetic images are generated using a generative adversarial network (GAN) and normalized by color transformation. The optimal deep features are extracted from each blood smear image using pretrained deep models i.e., DarkNet-53 and ShuffleNet. More informative features are selected by principal component analysis (PCA) and fused serially for classification. The morphological operations based on color thresholding with the deep semantic method are utilized for leukemia segmentation of classified cells. The classification accuracy achieved with ALL-IDB and LISC dataset is 100% and 99.70% for the classification of leukocytes i.e., blast, no blast, basophils, neutrophils, eosinophils, lymphocytes, and monocytes, respectively. Whereas semantic segmentation achieved 99.10% and 98.60% for average and global accuracy, respectively. The proposed method achieved outstanding outcomes as compared to the latest existing research works.

Deubiquitination of MITF-M Regulates Melanocytes Proliferation and Apoptosis

Microphthalmia-associated transcription factor-M (MITF-M) is the key gene in the proliferation and differentiation of melanocytes, which undergoes an array of post-translation modifications. As shown in our previous study, deubiquitinase USP13 is directly involved in melanogenesis. However, it is still ambiguous that the effect of USP13-mediated MITF-M expression on melanocytes proliferation and apoptosis. Herein, we found that MITF-M overexpressing melanocytes showed high cell proliferation, reduced apoptosis, and increased melanin levels. Besides, melanin-related genes, TYR, DCT, GPNMB, and PMEL, were significantly up-regulated in MITF-M overexpressing melanocytes. Furthermore, Exogenous USP13 significantly upregulated the endogenous MITF-M protein level, downregulated USP13 significantly inhibited MITF-M protein levels, without altering MITF-M mRNA expression. In addition, USP13 upregulation mitigated the MITF-M degradation and significantly increased the half-life of MITF-M. Also, USP13 stabilized the exogenous MITF protein levels. In conclusion, the MITF-M level was regulated by USP13 deubiquitinase in melanocytes, affecting melanocytes proliferation and apoptosis. This study provides the theoretical basis for coat color transformation that could be useful in the development of the new breed in fur animals.

Research on the classification of lymphoma pathological images based on deep residual neural network

BACKGROUND: Malignant lymphoma is a type of tumor that originated from the lymphohematopoietic system, with complex etiology, diverse pathological morphology, and classification. It takes a lot of time and energy for doctors to accurately determine the type of lymphoma by observing pathological images. OBJECTIVE: At present, an automatic classification technology is urgently needed to assist doctors in analyzing the type of lymphoma. METHODS: In this paper, by comparing the training results of the BP neural network and BP neural network optimized by genetic algorithm (GA-BP), adopts a deep residual neural network model (ResNet-50), with 374 lymphoma pathology images as the experimental data set. After preprocessing the dataset by image flipping, color transformation, and other data enhancement methods, the data set is input into the ResNet-50 network model, and finally classified by the softmax layer. RESULTS: The training results showed that the classification accuracy was 98.63%. By comparing the classification effect of GA-BP and BP neural network, the accuracy of the network model proposed in this paper is improved. CONCLUSIONS: The network model can provide an objective basis for doctors to diagnose lymphoma types.

AN EFFICIENT FPGA OVERLAY FOR COLOR TRANSFORMATION FUNCTION USING HIGH LEVEL SYNTHESIS

Image Processing is a significantly desirable in commercial, industrial, and medical applications. Processor based architectures are inappropriate for real time applications as Image processing algorithms are quite intensive in terms of computations. To reduce latency and limitation in performance due to limited amount of memory and fixed clock frequency for synthesis in processor-based architecture, FPGA can be used in smart devices for implementing real time image processing applications. To increase speed of real time image processing custom overlays (Hardware Library of programmable logic circuit) can be designed to run on FPGA fabric. The IP core generated by the HLS (High Level Synthesis) can be implemented on a reconfigurable platform which allows effective utilization of channel bandwidth and storage. In this paper we have presented FPGA overlay design for color transformation function using Xilinx’s python productivity board PYNQ-Z2 to get benefit in performance over a traditional processor. Performance comparison of custom overlay on FPGA and Processor based platform shows FPGA execution yields minimum computation time.

The Tungsten Oxide Within Phosphate Glasses to Investigate the Structural, Optical and Shielding Properties Variations

We prepared a series of sodium phosphate glasses by changing WO3/P2O5 content and investigated structure optical and radiation shielding features as a function of glass composition. The average density (ρexp) and was found to increase gradually from 2.49 to 3.07 g/cm3 while the average molar volume values reduced from 47.37 to 44.28 cm3/mol with WO3 addition. Also, the average field strength was also computed and found to increase with increasing WO3. The study of optical absorption spectra reveals that, the absorption peaks in the visible region become higher compared to the peaks In the UV region. This observation is accompanied with a color transformation of glasses from light to dark blue color, with more WO3 adding. The existence of pentavalent tungsten state (W5+) is identified by this blue color. with WO3 addition an absorption band at around at 350–390 nm is appeared. Moreover, this band is overlapped with Urbach edge, which regularly produces an artificial edge-like feature at ~400 nm. A detailed deconvolution protocol is required for an appropriate understanding of these spectra and unravelling the hidden Urbach edge. Our analysis shows that, with increasing WO3/P2O5 content, the optical band gap decreases. This behavior can be elucidated in terms of lower band gap of W (2.7 eV) than that of P2O5 (8.5 eV) and the high polarizing power W. Further, the radiation shielding parameters were investigated for the prepared glasses. WO3 addition improves these shielding parameters against radiation. Where, upon the increase of WO3 concentration, the LAC of glass material increases which leads to a decrease in HVL value. Then it is deducible that the amount of WO3 in this glass sample has an important impact on the shielding capability at lower energy values and has a slight impact at higher energy values.