scholarly journals Nuclear Morphology Optimized Deep Hybrid Learning (NUMODRIL): A novel architecture for accurate diagnosis/prognosis of Ovarian Cancer

2020 ◽  
Author(s):  
Duhita Sengupta ◽  
Sk Nishan Ali ◽  
Aditya Bhattacharya ◽  
Joy Mustafi ◽  
Asima Mukhopadhyay ◽  
...  
Keyword(s):  
Ovarian Cancer ◽  
Deep Learning ◽  
Cancer Cells ◽  
Hybrid Learning ◽  
Nuclear Morphology ◽  
Learning Approach ◽  
Diagnosis And Prognosis ◽  
Learning Techniques ◽  
Wide Range ◽  
Cancer Tissues

AbstractNuclear morphological features are potent determining factors for clinical diagnostic approaches adopted by pathologists to analyse the malignant potential of cancer cells. Considering the structural alteration of nucleus in cancer cells, various groups have developed machine learning techniques based on variation in nuclear morphometric information like nuclear shape, size, nucleus-cytoplasm ratio and various non-parametric methods like deep learning have also been tested for analysing immunohistochemistry images of tissue samples for diagnosing various cancers. Our aim is to study the morphometric distribution of nuclear lamin proteins as a specific parameter in ovarian cancer tissues. Besides being the principal mechanical component of the nucleus, lamins also present a platform for binding of proteins and chromatin thereby serving a wide range of nuclear functions like maintenance of genome stability, chromatin regulation. Altered expression of lamins in different subtypes of cancer is now evident from data across the world. It has already been elucidated that in ovarian cancer, extent of alteration in nuclear shape and morphology can determine degree of genetic changes and thus can be utilized to predict the outcome of low to high form of serous carcinoma. In this work, we have performed exhaustive imaging of ovarian cancer versus normal tissue and introduced a novel Deep Hybrid Learning approach on the basis of the distribution of lamin proteins. Although developed with ovarian cancer datasets in view, this architecture would be of immense importance in accurate and fast diagnosis and prognosis of all types of cancer associated with lamin induced morphological changes and would perform across small/medium to large datasets with equal efficiency.Significance StatementWe have developed a novel Deep Hybrid Learning approach based on nuclear morphology to classify normal and ovarian cancer tissues with highest possible accuracy and speed. Ovarian cancer cells can be easily distinguished from their enlarged nuclear morphology as is evident from lamin A & B distribution pattern. This is the first report to invoke specific nuclear markers like lamin A & B instead of classical haematoxylin-eosin staining in an effort to build parametric datasets. Our approach has been shown to outperform the existing deep learning techniques in training and validation of datasets over a wide range. Therefore this method could be used as a robust model to predict malignant transformations of benign nuclei and thus be implemented in the diagnosis and prognosis of ovarian cancer in future. Most importantly, this method can be perceived as a generalized approach in the diagnosis for all types of cancer.

scholarly journals Nuclear Morphology Optimized Deep Hybrid Learning (NUMODRIL) For Accurate Diagnosis and Prognosis of Ovarian Cancer

2021 ◽  
Author(s):  
Duhita Sengupta ◽  
Sk Nishan Ali ◽  
Aditya Bhattacharya ◽  
Joy Mustafi ◽  
Asima Mukhopadhyay ◽  
...  
Keyword(s):  
Ovarian Cancer ◽  
Cancer Cells ◽  
Morphological Changes ◽  
Hybrid Learning ◽  
Nuclear Shape ◽  
Machine Learning Techniques ◽  
Serous Carcinoma ◽  
Altered Expression ◽  
Diagnosis And Prognosis ◽  
Wide Range

Abstract Nuclear morphological features are potent determining factors for clinical diagnostic approaches adopted by pathologists to analyse the malignant potential of cancer cells. Considering the structural alteration of nucleus in cancer cells, various groups have developed machine learning techniques based on variation in nuclear morphometric information like nuclear shape, size, nucleus-cytoplasm ratio and various non-parametric methods like deep learning have also been tested for analysing immunohistochemistry images of tissue samples for diagnosing various cancers. Our aim is to study the morphometric distribution of nuclear lamin proteins as a specific parameter in ovarian cancer tissues. Besides being the principal mechanical component of the nucleus, lamins also present a platform for binding of proteins and chromatin thereby serving a wide range of nuclear functions like maintenance of genome stability, chromatin regulation. Altered expression of lamins in different subtypes of cancer is now evident from data across the world. It has already been elucidated that in ovarian cancer, extent of alteration in nuclear shape and morphology can determine degree of genetic changes and thus can be utilized to predict the outcome of low to high form of serous carcinoma. In this work, we have performed exhaustive imaging of ovarian cancer versus normal tissue and introduced a novel Deep Hybrid Learning approach on the basis of the distribution of lamin proteins. Although developed with ovarian cancer datasets in view, this architecture would be of immense importance in accurate and fast diagnosis and prognosis of all types of cancer associated with lamin induced morphological changes and would perform across small/medium to large datasets with equal efficiency.

scholarly journals A deep hybrid learning pipeline for accurate diagnosis of ovarian cancer based on nuclear morphology

PLoS ONE ◽  
2022 ◽  
Vol 17 (1) ◽  
pp. e0261181
Author(s):  
Duhita Sengupta ◽  
Sk Nishan Ali ◽  
Aditya Bhattacharya ◽  
Joy Mustafi ◽  
Asima Mukhopadhyay ◽  
...  
Keyword(s):  
Machine Learning ◽  
Ovarian Cancer ◽  
Deep Learning ◽  
Cancer Cells ◽  
Hybrid Learning ◽  
Machine Learning Algorithms ◽  
Nuclear Shape ◽  
Machine Learning Techniques ◽  
Serous Carcinoma ◽  
Learning Techniques

Nuclear morphological features are potent determining factors for clinical diagnostic approaches adopted by pathologists to analyze the malignant potential of cancer cells. Considering the structural alteration of the nucleus in cancer cells, various groups have developed machine learning techniques based on variation in nuclear morphometric information like nuclear shape, size, nucleus-cytoplasm ratio and various non-parametric methods like deep learning have also been tested for analyzing immunohistochemistry images of tissue samples for diagnosing various cancers. We aim to correlate the morphometric features of the nucleus along with the distribution of nuclear lamin proteins with classical machine learning to differentiate between normal and ovarian cancer tissues. It has already been elucidated that in ovarian cancer, the extent of alteration in nuclear shape and morphology can modulate genetic changes and thus can be utilized to predict the outcome of low to a high form of serous carcinoma. In this work, we have performed exhaustive imaging of ovarian cancer versus normal tissue and developed a dual pipeline architecture that combines the matrices of morphometric parameters with deep learning techniques of auto feature extraction from pre-processed images. This novel Deep Hybrid Learning model, though derived from classical machine learning algorithms and standard CNN, showed a training and validation AUC score of 0.99 whereas the test AUC score turned out to be 1.00. The improved feature engineering enabled us to differentiate between cancerous and non-cancerous samples successfully from this pilot study.

scholarly journals A Generalization Performance Study Using Deep Learning Networks in Embedded Systems

Sensors ◽  
2021 ◽  
Vol 21 (4) ◽  
pp. 1031
Author(s):  
Joseba Gorospe ◽  
Rubén Mulero ◽  
Olatz Arbelaitz ◽  
Javier Muguerza ◽  
Miguel Ángel Antón
Keyword(s):  
Deep Learning ◽  
Embedded Systems ◽  
Embedded System ◽  
General Purpose ◽  
Learning Networks ◽  
Performance Study ◽  
Learning Techniques ◽  
Wide Range ◽  
Learning Architectures

Deep learning techniques are being increasingly used in the scientific community as a consequence of the high computational capacity of current systems and the increase in the amount of data available as a result of the digitalisation of society in general and the industrial world in particular. In addition, the immersion of the field of edge computing, which focuses on integrating artificial intelligence as close as possible to the client, makes it possible to implement systems that act in real time without the need to transfer all of the data to centralised servers. The combination of these two concepts can lead to systems with the capacity to make correct decisions and act based on them immediately and in situ. Despite this, the low capacity of embedded systems greatly hinders this integration, so the possibility of being able to integrate them into a wide range of micro-controllers can be a great advantage. This paper contributes with the generation of an environment based on Mbed OS and TensorFlow Lite to be embedded in any general purpose embedded system, allowing the introduction of deep learning architectures. The experiments herein prove that the proposed system is competitive if compared to other commercial systems.

scholarly journals Forecast Breast Cancer Cells from Microscopic Biopsy Images using Big Transfer (BiT): A Deep Learning Approach

2021 ◽  
Vol 12 (10) ◽  
Author(s):  
Md. Ashiqul Islam ◽  
Dhonita Tripura ◽  
Mithun Dutta ◽  
Md. Nymur Rahman Shuvo ◽  
Wasik Ahmmed Fahim ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Deep Learning ◽  
Cancer Cells ◽  
Breast Cancer Cells ◽  
Learning Approach

scholarly journals A Study on the Evolution of Ransomware Detection Using Machine Learning and Deep Learning Techniques

IoT ◽  
2020 ◽  
Vol 1 (2) ◽  
pp. 551-604
Author(s):  
Damien Warren Fernando ◽  
Nikos Komninos ◽  
Thomas Chen
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Learning Approach ◽  
Learning Approaches ◽  
Learning Techniques ◽  
New Directions ◽  
Encryption Schemes ◽  
Malware Evolution ◽  
Number Of Citations ◽  
Research Studies

This survey investigates the contributions of research into the detection of ransomware malware using machine learning and deep learning algorithms. The main motivations for this study are the destructive nature of ransomware, the difficulty of reversing a ransomware infection, and how important it is to detect it before infecting a system. Machine learning is coming to the forefront of combatting ransomware, so we attempted to identify weaknesses in machine learning approaches and how they can be strengthened. The threat posed by ransomware is exceptionally high, with new variants and families continually being found on the internet and dark web. Recovering from ransomware infections is difficult, given the nature of the encryption schemes used by them. The increase in the use of artificial intelligence also coincides with this boom in ransomware. The exploration into machine learning and deep learning approaches when it comes to detecting ransomware poses high interest because machine learning and deep learning can detect zero-day threats. These techniques can generate predictive models that can learn the behaviour of ransomware and use this knowledge to detect variants and families which have not yet been seen. In this survey, we review prominent research studies which all showcase a machine learning or deep learning approach when detecting ransomware malware. These studies were chosen based on the number of citations they had by other research. We carried out experiments to investigate how the discussed research studies are impacted by malware evolution. We also explored the new directions of ransomware and how we expect it to evolve in the coming years, such as expansion into IoT (Internet of Things), with IoT being integrated more into infrastructures and into homes.

scholarly journals Breaking the scale: how disrupting the karyoplasmic ratio gives cancer cells an advantage for metastatic invasion

2017 ◽  
Vol 45 (6) ◽  
pp. 1333-1344
Author(s):  
Andrea Rizzotto ◽  
Eric C. Schirmer
Keyword(s):  
Breast Cancer ◽  
Lung Cancer ◽  
Cancer Cells ◽  
Nuclear Size ◽  
Nuclear Morphology ◽  
Metastatic Tumors ◽  
Size Regulation ◽  
Diagnosis And Prognosis ◽  
Ploidy Changes ◽  
Tumor Types

Nuclear size normally scales with the size of the cell, but in cancer this ‘karyoplasmic ratio’ is disrupted. This is particularly so in more metastatic tumors where changes in the karyoplasmic ratio are used in both diagnosis and prognosis for several tumor types. However, the direction of nuclear size changes differs for particular tumor types: for example in breast cancer, larger nuclear size correlates with increased metastasis, while for lung cancer smaller nuclear size correlates with increased metastasis. Thus, there must be tissue-specific drivers of the nuclear size changes, but proteins thus far linked to nuclear size regulation are widely expressed. Notably, for these tumor types, ploidy changes have been excluded as the basis for nuclear size changes, and so, the increased metastasis is more likely to have a basis in the nuclear morphology change itself. We review what is known about nuclear size regulation and postulate how such nuclear size changes can increase metastasis and why the directionality can differ for particular tumor types.

scholarly journals Demethoxycurcumin inhibited human epithelia ovarian cancer cells’ growth via up-regulating miR-551a

Tumor Biology ◽  
2017 ◽  
Vol 39 (3) ◽  
pp. 101042831769430 ◽  
Author(s):  
Zhenhua Du ◽  
Xianqun Sha
Keyword(s):  
Ovarian Cancer ◽  
Cell Proliferation ◽  
Cancer Cells ◽  
Ovarian Cancer Cells ◽  
Natural Form ◽  
Ovarian Cancer Cell Lines ◽  
Induced Apoptosis ◽  
Tumor Suppressive Function ◽  
Cancer Tissues

Curcumin is a natural agent that has ability to dampen tumor cells’ growth. However, the natural form of curcumin is prone to degrade and unstable in vitro. Here, we demonstrated that demethoxycurcumin (a curcumin-related demethoxy compound) could inhibit cell proliferation and induce apoptosis of ovarian cancer cells. Moreover, IRS2/PI3K/Akt axis was inactivated in cells treated with demethoxycurcumin. Quantitative real-time reverse transcription polymerase chain reaction demonstrated that miR-551a was down-regulated in ovarian cancer tissues and ovarian cancer cell lines. Over-expression of miR-551a inhibited cell proliferation and induced apoptosis of ovarian cancer cells, whereas down-regulation of miR-551a exerted the opposite function. Luciferase assays confirmed that there was a binding site of miR-551a in IRS2, and we found that miR-551a exerted tumor-suppressive function by targeting IRS2 in ovarian cancer cells. Remarkably, miR-551a was up-regulated in the cells treated with demethoxycurcumin, and demethoxycurcumin suppressed IRS2 by restoration of miR-551a. In conclusion, demethoxycurcumin hindered ovarian cancer cells’ malignant progress via up-regulating miR-551a.

scholarly journals Full color generation with Fano-type resonant HfO2 nanopillars designed by a deep-learning approach

Nanoscale ◽  
2019 ◽  
Vol 11 (44) ◽  
pp. 21266-21274 ◽  
Author(s):  
Omid Hemmatyar ◽  
Sajjad Abdollahramezani ◽  
Yashar Kiarashinejad ◽  
Mohammadreza Zandehshahvar ◽  
Ali Adibi
Keyword(s):  
Deep Learning ◽  
Fano Resonance ◽  
Learning Approach ◽  
Full Color ◽  
Wide Range ◽  
Learning Technique ◽  
First Time

Here, for the first time to our knowledge, a Fano resonance metasurface made of HfO2 is experimentally demonstrated to generate a wide range of colors. We use a novel deep-learning technique to design and optimize the metasurface.

Sign in / Sign up

Export Citation Format

Share Document