A New Fast Fractal Modeling Approach for the Detection of Microcalcifications in Mammograms

Sankar, Deepa; Thomas, Tessamma

doi:10.1007/s10278-009-9224-6

Cited by 29 publications

(19 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…References Identification of masses [3], [5], [8], [14], [16], [22], or microcalcification [23], [26], [28], [29], [30], [31] Lesion segmentation [9], [17], [25], [32] Image Enhancement [4], [7], [12] Pectoral muscle identification [18], [20] relaxed shape constraint was proposed. Initial boundary on the smoothed mammogram was determined using contourbased level set method.…”

Section: Applicationsmentioning

confidence: 99%

Computational unfoldment of mammograms

Joshi

Bhale

2012

International Conference on Pattern Recognition, Informatics and Medical Engineering (PRIME-2012)

View full text Add to dashboard Cite

The importance of mammograms in early breast cancer detection is an accepted fact. Mammograms (either an analog x-ray film or a digital softcopy) are computationally empowered to extract significant information. Several computational techniques/algorithms process mammograms to highlight and reveal otherwise unseen features. Thus mammographic images are computationally unfolded to obtain appropriate information that can be used for further analysis. Computational analysis of mammograms is an essential tool, which is used by numerous specialists for various purposes. In this paper we review such research work reported in the literature in recent years . Our focus is in particular on computational preprocessing of mammograms. Preprocessing involves enhancement of mammographic images as well as extraction of relevant features from images. We grouped various image enhancement research approaches systematically. We also categorized various research techniques based on the types of features that are extracted and used to obtain intended results. Although mammograms are used mostly for breast cancer detection, the research is not confined to this aspect only. Several other areas that deal with mammograms are also explored by researchers including image compression, Content based Image Retrieval (CBIR) etc. Variety in these research applications is also discussed and presented in this paper.

show abstract

Section: Applicationsmentioning

confidence: 99%

Computational unfoldment of mammograms

Joshi

Bhale

2012

International Conference on Pattern Recognition, Informatics and Medical Engineering (PRIME-2012)

View full text Add to dashboard Cite

show abstract

“…A great deal of effort has gone into designing quality assessment methods that take advantage of known characteristics of the HVS. Natural image signals are highly structured [12]. The most basic principle to image quality assessment is that the HVS is highly adapted to extract structural information from the visual scene, and consequently a measurement of structural similarity (or distortion) provides a good approximation to perceptual image quality.…”

Section: The Comparative Proposed Approachmentioning

confidence: 99%

Assessment of Diverse Quality Metrics for Medical Images Including Mammography

VenkatNarayanaRao¹,

Govardhan²

2013

IJCA

View full text Add to dashboard Cite

This paper presents the comparative analy sis of various quality metrics for medical image processing. M easurement of image quality is vital for numerous image-processing applications. Image quality measurement is closely related to image resemblance and assessment in which quality is based on the differences (or similarity) between a degraded image and that of the original image i.e. unmodified image chiefly in mammographic images. We have e mp loy ed s imple verifiable techniques for representing the image quality rather than difficult mathematical procedures, which are costly, time consuming and observer dependent. In this paper, the images have been subjected to various degrees of blur, noise, co mp ression , contrast levels . Bas ed on thes e factors the quality has been measured in terms of metrics like M ean Squared Error (MSE), Peak Signal-toNoise Ratio (PSNR), M aximum difference (M D) including new metric of image qualities such as Structural Similarity Index M etrics (SSIM )for low cost medical image analysis. KeywordsImage quality analysis, M ean Square Error (M SE), Structural Similarity Index M etric(SSIM ), Peak Signal to Noise Ratio(PSNR), M ean Absolute Error(M AE)

show abstract

“…In the method by Eddaoudi et al [7], a coding scheme is used to reduce gray levels from original images to efficiently extract Haralick features from the gray level co-occurrence matrix. Sankar et al [8] suggest a new fast fractal coding approach where a modeled image similar to the original image is reconstructed. MCs are enhanced in the difference image between the original and reconstructed images.…”

Section: Introductionmentioning

confidence: 99%

“…The method of Quintanilla-Dominguez and Cortina-Januchs [5] where morphological filtering known as the 'top-hat' transform is applied to raw mammograms to discriminate the MCs and the method by Zhang et al [6] which adds wavelet decomposition to suppress non-target tissues such as vessels and mammary glands are examples of this approach. The third is coding based methods [7,8,9]. In the method by Eddaoudi et al [7], a coding scheme is used to reduce gray levels from original images to efficiently extract Haralick features from the gray level co-occurrence matrix.…”

Section: Introductionmentioning

confidence: 99%

Classification based micro-calcification detection using discriminative restricted Boltzmann machine in digitized mammograms

2014

View full text Add to dashboard Cite

We present a new method for automatic detection of micro-calcifications using the Discriminative Restricted Boltzmann Machine (DRBM). The DRBM is used to automatically learn the specific features which distinguish micro-calcifications from normal tissue as well as their morphological variations. Within the DRBM, low level image structures that are specific features of micro-calcifications are automatically captured without any appropriate feature selection based on expert knowledge or time-consuming hand-tuning, which was required for previous methods. Experimental evaluation conducted on a set of 33 mammograms gave a result of area under Receiver Operating Characteristics (ROC) curve 0.8294, which showed the effectiveness of the proposed method.

show abstract

A New Fast Fractal Modeling Approach for the Detection of Microcalcifications in Mammograms

Cited by 29 publications

References 13 publications

Computational unfoldment of mammograms

Computational unfoldment of mammograms

Assessment of Diverse Quality Metrics for Medical Images Including Mammography

Classification based micro-calcification detection using discriminative restricted Boltzmann machine in digitized mammograms

Contact Info

Product

Resources

About