Papers by farida khursheed
Multimedia Tools and Applications, Mar 29, 2023
Today breast cancer is the leading type of cancer among women undergoing cancer screening. A slig... more Today breast cancer is the leading type of cancer among women undergoing cancer screening. A slight delay in detecting and diagnosing this disease may result in irreversible convolutions. Histopathological images from the biopsy examination present a large amount of structural information that can signi cantly improve the prognosis for breast cancer. The pathological analysis, which involves the microscopic examination of the histopathological slides, is a challenging task. An automated computer-aided detection (CAD) procedure is inevitable, as it may decrease the pathologist examination time and help detect the disease at an early stage. Lately, deep learning methods using arti cial neural networks are consistently in use to improve the performance of CAD methods. A common practice among recent studies is to use the transfer learning approach of training deep neural network architectures. Transfer Learning is an established learning approach that facilitates a deep neural network to train quickly on a speci c dataset and resolve an interdisciplinary problem. Deep Learning methods employing the transfer learning approach have provided highly competitive results on the datasets consisting of the whole slide images, which are captured generally at high resolutions. However, the performance is not remarkably appreciable on the small and low-resolution image datasets, in particular the datasets that include patch samples. In this direction, the present study proposes a novel domain-speci c learning strategy, Breast Histo-Fusion, which aims to detect breast cancer even from images of low resolution and small size. Further, four state-of-the-art deep CNNs (AlexNet, VGG19, ResNet, and DenseNet) are trained using both learning approaches: Transfer Learning and Histo-Fusion on the IDC dataset. The proposed Histo-Fusion learning approach has improved the discriminating abilities and performance of each deep CNN by providing better results of (AlexNet-95.75%, VGG19-95.96%, ResNet34-96.17%, ResNet50-96.67%, and DenseNet121-97.49%) compared to (AlexNet-90.41, VGG19-90.51%, ResNet34-90.83%, ResNet50-92.27%, and DenseNet121-93.05%) using the transfer learning strategy. As a result, the procedure can help expert pathologists to perform accurate diagnoses and reduce false-positive rates.
Neural Computing and Applications
International Journal on Document Analysis and Recognition (IJDAR)
Today breast cancer is the leading type of cancer among women undergoing cancer screening. A slig... more Today breast cancer is the leading type of cancer among women undergoing cancer screening. A slight delay in detecting and diagnosing this disease may result in irreversible convolutions. Histopathological images from the biopsy examination present a large amount of structural information that can significantly improve the prognosis for breast cancer. The pathological analysis, which involves the microscopic examination of the histopathological slides, is a challenging task. An automated computer-aided detection (CAD) procedure is inevitable, as it may decrease the pathologist examination time and help detect the disease at an early stage. Lately, deep learning methods using artificial neural networks are consistently in use to improve the performance of CAD methods. A common practice among recent studies is to use the transfer learning approach of training deep neural network architectures. Transfer Learning is an established learning approach that facilitates a deep neural network...
Concurrency and Computation: Practice and Experience, Aug 29, 2022
Concurrency and Computation: Practice and Experience
Over the years passed has witnessed great interest in research on content-based image retrieval. ... more Over the years passed has witnessed great interest in research on content-based image retrieval. This has paved the way for a large number of new techniques and systems, and a growing interest in associated fields to support such systems. Similarly, digital image retrieval has expanded in many directions that are resulting into explosion in the volume of image data required to be organized. This paper presents a framework for image retrieval based on chain code and auto regression that helps to achieve higher retrieval efficiency. In this paper, we discuss about the key contributions of the methodology that is followed while performing experiment for image retrieval based on chain code and auto regression. Here comparative study of results and also efficiency of both these image retrieval techniques are discussed which are obtained while experimentation.
Algorithms for Intelligent Systems, 2021
Digital images are used to convey information as we human tend to trust what we perceive. Owing t... more Digital images are used to convey information as we human tend to trust what we perceive. Owing to the development in tools and technology, manipulation of digital image is becoming drastically easy and more frequent. Digital images are forged beyond visual comprehension for the ulterior motives. Forgery detection in digital images is necessary to unravel the truth. There are several methods that have been discussed in great detail; basically, an overview of different types of image forgeries and their corresponding detection has been provided, and we are dedicated toward pixel-based methods of passive\blind detection.
Biomedical Signal Processing and Control
8th International Conference one-Health Networking, Applications and Services, 2006. HEALTHCOM 2006.
ABSTRACT In Computed Tomography (CT) images a gap exists between the information these images con... more ABSTRACT In Computed Tomography (CT) images a gap exists between the information these images contain and the information that can be retrieved on the basis of visual inspection. In this paper an attempt is made to reduce this gap with an objective of identifying the disease at an early stage from CT images. This is achieved by combining the identified spatial domain statistical texture parameters with fuzzy logic image enhancement technique. The texture parameters are used to separate out the scans containing information about the disease that can not be determined on the basis of visual perception. Fuzzy logic based image enhancement technique is applied on these scans to identify the underlying pathology.
Journal of Biomedical Informatics, 2022
International Journal of Electronic Security and Digital Forensics, 2022
With the improvement in cryptanalysis techniques, need of more secure techniques evolved. Accordi... more With the improvement in cryptanalysis techniques, need of more secure techniques evolved. Accordingly, people thought of using covers for hiding the sensitive information. This hiding of information in covers is known as Steganography. In Steganography that uses images as covers, main goal is to embed information in such a way so as to make information imperceptible. A number of techniques were developed to embed the information in spatial and frequency domain .For embedding of information in transform domain the images are first transformed to frequency domain to save it from Common cover attacks. However, transformation of image into frequency domain for embedding and then retrieval of information is computationally expensive. In this work, therefore, attempt has been made to embed the information in fuzzy logic domain. Results reveal the advantages of embedding not only in terms of computational ease but it also in terms of embedding versatility; since it embeds text, color and g...
2019 Fifth International Conference on Image Information Processing (ICIIP), 2019
Language Modeling is defined as the operation of predicting next word. It is considered as one of... more Language Modeling is defined as the operation of predicting next word. It is considered as one of the basic tasks of Natural Language Processing(NLP) and Language Modeling has several applications. In this research paper, the assorted potentialities for the efficient utilization of language models in structured document retrieval are mentioned. A tree-based generative language model for ranking documents and parts has been used here. Nodes within the tree correspond to different document parts like titles, paragraphs and sections. At every node within the document tree, there's a well-defined language model. The language model for a leaf node is predictable directly from the text within the document part related to the node. Inner nodes within the tree are predictable employing a linear interpolation among the various youngster nodes. The paper additionally describes how some common structural queries would be satisfactorily described inside this model.
Over last two decades, due to hostilities of environment over the internet the concerns about con... more Over last two decades, due to hostilities of environment over the internet the concerns about confidentiality of information have increased at phenomenal rate. Therefore to safeguard the information from attacks, number of data/information hiding methods have evolved in spatial and transformation domain. Literature survey revealed that embedding in fuzzy logic domain has not been attempted. In this paper we have made an attempt to embed information in fuzzy logic domain. Results reveal the advantages of embedding in fuzzy domain not only in terms of computational ease but also in terms of embedding versatility.
Multimedia Tools and Applications
Advances in Healthcare Information Systems and Administration
The expeditious progress of machine learning, especially the deep learning techniques, keep prope... more The expeditious progress of machine learning, especially the deep learning techniques, keep propelling the medical imaging community's heed in applying these techniques in improving the accuracy of cancer screening. Among various types of cancers, breast cancer is the most detrimental disease affecting women today. The prognosis of such types of disease becomes a very challenging task for radiologists due the huge number of cases together with careful and thorough examination it demands. The constraints of present CAD open up a need for new and accurate detection procedures. Deep learning approaches have gained a tremendous recognition in the areas of object detection, segmentation, image recognition, and computer vision. Precise and premature detection and classification of lesions is very critical for increasing the survival rates of patients. Recent CNN models are designed to enhance radiologists' understandings to identify even the least possible lesions at the very earl...
International Journal of Computing and Digital Systemss
Increased threats to conventional personnel verification methods, have given rise to verification... more Increased threats to conventional personnel verification methods, have given rise to verification methods based on biometrics. This paper presents a novel approach for personal verification based on fusion of two biometric modalities: Ear and Iris. Fusion has been done in two levels. In level one right iris and left iris features have been fused using texture based Grey Level Concurrence Matrix (GLCM) and in level two these features have been fused with AR coefficients taken from ear substructure. It has been found that this two level fusion of features improves the recognition rate of a person to the extent of 100%. The biometric modalities of ear and iris have been chosen on the basis of their desirable properties like uniqueness, universality, permanency and acceptability.
![International Journals for Researchers [ER Publication, WOAR Journals, IJEAS and IJEART]](https://0.academia-photos.com/7031831/2656864/151250930/s200_international_journals_for_researchers._er_publication_woar_journals_ijeas_and_ijeart_.png)
Uploads
Papers by farida khursheed