scholarly journals A Hybrid High Performance Intelligent Computing Approach of CACNN and RNN for Skin Cancer Image Grading

2021 ◽  
Author(s):  
Manimurugan S
Keyword(s):  
Skin Cancer ◽  
High Performance ◽  
Spatial Arrangement ◽  
The Cancer Genome Atlas ◽  
Intelligent Computing ◽  
Skin Cells ◽  
Skin Cancers ◽  
Cancer Genome Atlas ◽  
Histological Images ◽  
Computational Pathology

Abstract Skin cancer is characterized as the uncontrollable growth of skin cells caused by unrepairable DNA damage. Melanoma is the deadliest form of skin cancers caused by melanocyte and early diagnosis supports therapists in curing it. Computational pathology offers a one-of-a-kind ability to spatially dissect certain interfaces on digitized histology images. A hybrid context-aware convolutional neural networks with recurrent neural network (CA-CNN-RNN) based on skin cancer histological images is proposed in this research. The proposed model encodes a histology image's local representation into higher-dimensional features first, then aggregated the feature by consider their spatial arrangement to enable the final predictions. In this research, H&E-stained sectioned images from the Cancer Genome Atlas are used as the dataset for assessment. From 58 images, 37 images were used for training and 21 images are used for testing. The process on histology images of melanoma skin cancer was analyzed and validated with various classifiers such as VGG-19, Inception, ResNet50, and DarkNet-53 using the hybrid CA-CNN-RNN model. The dataset is used to generate the results, which are then analyzed based on criteria such as accuracy, recall, precision, and F-score. The performance analysis shows that the proposed CA-CNN-RNN with different classifiers has performed better and among the classifiers the DarkNet-53 model has the better performance in all the parameters.

Prediction of transcription factors binding events based on epigenetic modifications in different human cells

Epigenomics ◽  
2020 ◽  
Vol 12 (16) ◽  
pp. 1443-1456
Author(s):  
Yan Huang ◽  
Dianshuang Zhou ◽  
Yihan Wang ◽  
Xingda Zhang ◽  
Mu Su ◽  
...  
Keyword(s):  
Cell Lines ◽  
Predictive Models ◽  
High Performance ◽  
Epigenetic Modifications ◽  
The Cancer Genome Atlas ◽  
Encode Project ◽  
Atlas Data ◽  
Random Forest Method ◽  
Cancer Genome Atlas ◽  
Genome Atlas

Aim: We aim to predict transcription factor (TF) binding events from knowledge of gene expression and epigenetic modifications. Materials & methods: TF-binding events based on the Encode project and The Cancer Genome Atlas data were analyzed by the random forest method. Results: We showed the high performance of TF-binding predictive models in GM12878, HeLa, HepG2 and K562 cell lines and applied them to other cell lines and tissues. The genes bound by the top TFs ( MAX and MAZ) were significantly associated with cancer-related processes such as cell proliferation and DNA repair. Conclusion: We successfully constructed TF-binding predictive models in cell lines and applied them in tissues.

scholarly journals Local Delivery of Minocycline and Vorinostat Act Synergistically to Inhibit Recurrence of Gliomas

2020 ◽  
Author(s):  
Gang Zhao ◽  
Jun Jia ◽  
Lansheng Wang ◽  
Yongkang Zhang ◽  
Han Yang ◽  
...  
Keyword(s):  
High Performance ◽  
Postoperative Recurrence ◽  
The Cancer Genome Atlas ◽  
The Public ◽  
Continuous Release ◽  
Clinical Consequences ◽  
Cancer Genome Atlas ◽  
Genome Atlas

Abstract Background:Postoperative recurrence is the main reason of poor clinical consequences in glioma patients, so preventing recurrence of tumors is crucial in management of gliomas. Methods:In this study, the expression of matrix metalloproteinases (MMPs)in tissues from normal were detected by using RNA-seq analysis.Glioma cases from the public databases (The Cancer Genome Atlas (TCGA), The Chinese Glioma Genome Atlas(CGGA), Betastasis) were included in this study.The hydrogelcontains minocycline (Mino) and vorinostat (Vor)(G/Mino + Vor) was formed under 365 nm when photoinitiator was added in. High Performance Liquid Chromatography (HPLC) assay was used to assessed the release of drugs in G/Mino + Vor hydrogel. MTT assay was used to explore the biosecurity of GelMA. Immunohistochemistry assay, ELISA assay, Tunel assay were used to demonstrate the antitumor effect of G/Mino + Vor hydrogel.Results:We developed G/Mino + Vor hydrogel successfully. Thenthe experiment in vitro and in vivo confirmed MMPs-responsive delivery of minocycline and vorinostat in hydrogel and the anti-glioma effect on incomplete tumor operation model, which indicated that G/Mino + Vor hydrogel effectively inhibited the recurrence of glioma after surgery.Conclusions: In summary, G/Mino + Vor hydrogel could continuous release minocycline and vorinostat in surgical cavity for inhibiting local recurrence of glioma after operation.

scholarly journals Regulation of Apoptosis during Environmental Skin Tumor Initiation

2021 ◽  
Author(s):  
Bilal Bin Hafeez ◽  
Eunmi Park ◽  
Kyung-Soo Chun ◽  
Yong-Yeon Cho ◽  
Dae Joon Kim
Keyword(s):  
Skin Cancer ◽  
Signaling Pathways ◽  
The United States ◽  
Common Type ◽  
Skin Tumor ◽  
Tumor Initiation ◽  
Environmental Carcinogens ◽  
Skin Cells ◽  
Skin Cancers ◽  
The Impact

Skin cancer is more prevalent than any other cancer in the United States. Non-melanoma skin cancers are the more common forms of skin cancer that affect individuals. The development of squamous cell carcinoma, the second most common type of skin cancer, can be stimulated by exposure of environmental carcinogens, such as chemical toxicants or UVB. It is developed by three distinct stages: initiation, promotion, and progression. During the initiation, the fate of DNA-damaged skin cells is determined by the homeostatic regulation of pro-apoptotic and anti-apoptotic signaling pathways. The imbalance or disruption of either signaling will lead to the survival of initiated cells, resulting in the development of skin cancer. In this chapter, we will discuss signaling pathways that regulate apoptosis and the impact of their dysfunction during skin tumor initiation.

scholarly journals Comparative Analysis of Skin Cancer (Benign vs. Malignant) Detection Using Convolutional Neural Networks

2021 ◽  
Vol 2021 ◽  
pp. 1-17
Author(s):  
Mohammed Rakeibul Hasan ◽  
Mohammed Ishraaf Fatemi ◽  
Mohammad Monirujjaman Khan ◽  
Manjit Kaur ◽  
Atef Zaguia
Keyword(s):  
Skin Cancer ◽  
Mortality Rate ◽  
Support Vector ◽  
Work Processes ◽  
Abnormal Growth ◽  
Skin Cells ◽  
Skin Cancers ◽  
Malignant Skin ◽  
Common Skin ◽  
Cancer Spread

We live in a world where people are suffering from many diseases. Cancer is the most threatening of them all. Among all the variants of cancer, skin cancer is spreading rapidly. It happens because of the abnormal growth of skin cells. The increase in ultraviolet radiation on the Earth’s surface is also helping skin cancer spread in every corner of the world. Benign and malignant types are the most common skin cancers people suffer from. People go through expensive and time-consuming treatments to cure skin cancer but yet fail to lower the mortality rate. To reduce the mortality rate, early detection of skin cancer in its incipient phase is helpful. In today’s world, deep learning is being used to detect diseases. The convolutional neural network (CNN) helps to find skin cancer through image classification more accurately. This research contains information about many CNN models and a comparison of their working processes for finding the best results. Pretrained models like VGG16, Support Vector Machine (SVM), ResNet50, and self-built models (sequential) are used to analyze the process of CNN models. These models work differently as there are variations in their layer numbers. Depending on their layers and work processes, some models work better than others. An image dataset of benign and malignant data has been taken from Kaggle. In this dataset, there are 6594 images of benign and malignant skin cancer. Using different approaches, we have gained accurate results for VGG16 (93.18%), SVM (83.48%), ResNet50 (84.39%), Sequential_Model_1 (74.24%), Sequential_Model_2 (77.00%), and Sequential_Model_3 (84.09%). This research compares these outcomes based on the model’s work process. Our comparison includes model layer numbers, working process, and precision. The VGG16 model has given us the highest accuracy of 93.18%.

scholarly journals Deep learning-based cross-classifications reveal conserved spatial behaviors within tumor histological images

2019 ◽  
Author(s):  
Javad Noorbakhsh ◽  
Saman Farahmand ◽  
Ali Foroughi pour ◽  
Sandeep Namburi ◽  
Dennis Caruana ◽  
...  
Keyword(s):  
Image Data ◽  
The Cancer Genome Atlas ◽  
Breast Cancers ◽  
Histopathological Classification ◽  
Histopathological Images ◽  
Cancer Genome Atlas ◽  
Histological Images ◽  
Cancer Types ◽  
Uterine Cancers ◽  
Spatial Behaviors

AbstractHistopathological images are a rich but incompletely explored data type for studying cancer. Manual inspection is time consuming, making it challenging to use for image data mining. Here we show that convolutional neural networks (CNNs) can be systematically applied across cancer types, enabling comparisons to reveal shared spatial behaviors. We develop CNN architectures to analyze 27,815 hematoxylin and eosin slides from The Cancer Genome Atlas for tumor/normal, cancer subtype, and mutation classification. Our CNNs are able to classify tumor/normal status of whole slide images (WSIs) in 19 cancer types with consistently high AUCs (0.995±0.008), as well as subtypes with lower but significant accuracy (AUC 0.87±0.1). Remarkably, tumor/normal CNNs trained on one tissue are effective in others (AUC 0.88±0.11), with classifier relationships also recapitulating known adenocarcinoma, carcinoma, and developmental biology. Moreover, classifier comparisons reveal intra-slide spatial similarities, with average tile-level correlation of 0.45±0.16 between classifier pairs. Breast cancers, bladder cancers, and uterine cancers have spatial patterns that are particularly easy to detect, suggesting these cancers can be canonical types for image analysis. Patterns for TP53 mutations can also be detected, with WSI self- and cross-tissue AUCs ranging from 0.65-0.80. Finally, we comparatively evaluate CNNs on 170 breast and colon cancer images with pathologist-annotated nuclei, finding that both cellular and intercellular regions contribute to CNN accuracy. These results demonstrate the power of CNNs not only for histopathological classification, but also for cross-comparisons to reveal conserved spatial biology.

scholarly journals Deep Texture Representations as a Universal Encoder for Pan-cancer Histology

2020 ◽  
Author(s):  
Daisuke Komura ◽  
Akihiro Kawabe ◽  
Keisuke Fukuta ◽  
Kyohei Sano ◽  
Toshikazu Umezaki ◽  
...  
Keyword(s):  
Large Scale ◽  
Gene Mutations ◽  
Clinical Information ◽  
The Cancer Genome Atlas ◽  
Strong Impact ◽  
Cancer Genome Atlas ◽  
Comprehensive Comparison ◽  
Similar Images ◽  
Histological Images ◽  
Cancer Types

SummaryCancer histological images contain rich biological and clinical information, but quantitative representation can be problematic and has prevented direct comparison and accumulation of large-scale datasets. Here we show that deep texture representations (DTRs) produced by a bilinear Convolutional Neural Network, express cancer morphology well in an unsupervised manner, and work as a universal encoder for cancer histology. DTRs are useful for content-based image retrieval, enabling quick retrieval of histologically similar images from optimally area selected datasets of 7,175 cases from The Cancer Genome Atlas. Via comprehensive comparison with driver and clinically actionable gene mutations, we have successfully predicted 309 combinations of genomic features and cancer types from hematoxylin and eosin-stained images at high accuracy (AUC > 0.70 and q < 0.02). With its mounting capabilities on accessible devices such as smartphones, DTR-based encoding for cancer histology has a potentially strong impact on global equalization for cancer diagnosis and targeted therapies.

scholarly journals Omics Pipe: A Computational Framework for Reproducible Multi-Omics Data Analysis

2014 ◽  
Author(s):  
Kathleen M Fisch ◽  
Tobias Meißner ◽  
Louis Gioia ◽  
Jean-Christophe Ducom ◽  
Tristan Carland ◽  
...  
Keyword(s):  
Data Analysis ◽  
High Performance ◽  
Best Practice ◽  
The Cancer Genome Atlas ◽  
Sequencing Analysis ◽  
Omics Data ◽  
Next Generation Sequencing Analysis ◽  
Cancer Genome Atlas ◽  
Computational Platform ◽  
Omics Data Analysis

Omics Pipe (https://bitbucket.org/sulab/omics_pipe) is a computational platform that automates multi-omics data analysis pipelines on high performance compute clusters and in the cloud. It supports best practice published pipelines for RNA-seq, miRNA-seq, Exome-seq, Whole Genome sequencing, ChIP-seq analyses and automatic processing of data from The Cancer Genome Atlas. Omics Pipe provides researchers with a tool for reproducible, open source and extensible next generation sequencing analysis.

scholarly journals Deep learning-based cross-classifications reveal conserved spatial behaviors within tumor histological images

2020 ◽  
Vol 11 (1) ◽  
Author(s):  
Javad Noorbakhsh ◽  
Saman Farahmand ◽  
Ali Foroughi pour ◽  
Sandeep Namburi ◽  
Dennis Caruana ◽  
...  
Keyword(s):  
Image Data ◽  
The Cancer Genome Atlas ◽  
Breast Cancers ◽  
Histopathological Classification ◽  
Histopathological Images ◽  
Cancer Genome Atlas ◽  
Histological Images ◽  
Cancer Types ◽  
Scanned Images ◽  
Spatial Behaviors

AbstractHistopathological images are a rich but incompletely explored data type for studying cancer. Manual inspection is time consuming, making it challenging to use for image data mining. Here we show that convolutional neural networks (CNNs) can be systematically applied across cancer types, enabling comparisons to reveal shared spatial behaviors. We develop CNN architectures to analyze 27,815 hematoxylin and eosin scanned images from The Cancer Genome Atlas for tumor/normal, cancer subtype, and mutation classification. Our CNNs are able to classify TCGA pathologist-annotated tumor/normal status of whole slide images (WSIs) in 19 cancer types with consistently high AUCs (0.995 ± 0.008), as well as subtypes with lower but significant accuracy (AUC 0.87 ± 0.1). Remarkably, tumor/normal CNNs trained on one tissue are effective in others (AUC 0.88 ± 0.11), with classifier relationships also recapitulating known adenocarcinoma, carcinoma, and developmental biology. Moreover, classifier comparisons reveal intra-slide spatial similarities, with an average tile-level correlation of 0.45 ± 0.16 between classifier pairs. Breast cancers, bladder cancers, and uterine cancers have spatial patterns that are particularly easy to detect, suggesting these cancers can be canonical types for image analysis. Patterns for TP53 mutations can also be detected, with WSI self- and cross-tissue AUCs ranging from 0.65-0.80. Finally, we comparatively evaluate CNNs on 170 breast and colon cancer images with pathologist-annotated nuclei, finding that both cellular and intercellular regions contribute to CNN accuracy. These results demonstrate the power of CNNs not only for histopathological classification, but also for cross-comparisons to reveal conserved spatial behaviors across tumors.

Sign in / Sign up

Export Citation Format

Share Document