Automated image segmentation for breast analysis using infrared images

Scales, N.; Kerry, C.; Prize, M.

doi:10.1109/iembs.2004.1403521

Cited by 40 publications

(20 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although only unaltered images and professional radiologists were utilized in the evaluation of the performance, the highest reported accuracy within the literature was surpassed. Unlike other published papers that had their results evaluated by the same research team [15,13,30], all segmentation results obtained with this method were evaluated independently by a board of three professional radiologists, whose evaluation results were combined to produce the final accuracy.…”

Section: Resultsmentioning

confidence: 99%

“…In breast thermography, the lack of defined boundaries on the upper regions of the breast and the fact that any individual person's breasts can show variations in shape, size, and appearance make the automatic segmentation of breasts from thermography a difficult task for many segmentation concepts. Many methods have been proposed for automatic segmentation of breasts in thermography, such as snakes and active contours [12], Hough transform-based segmentation [13], morphological image segmentation [14], and curvature-based segmentation [15]. However, these methods require images that are captured based on very restrictive imaging protocols and require a fair amount of manual alteration to certain parts of the image to make them usable.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Breast boundary segmentation in thermography images based on random walkers

Moghbel¹,

Mashohor²,

Mahmud³

et al. 2017

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

Breast and areola boundary detection and segmentation present the biggest challenge in breast segmentation from thermography images, as breast boundaries, especially in the upper quadrants of the breast, are nonexistent. Many segmentation approaches have been proposed for breast segmentation, such as active contours and snakes, circular Hough transforms, and live wires, but these methods often fail to achieve satisfactory results. With recent advances in image processing techniques, new segmentation concepts are being developed, such as random walkers, which have received high interest from the medical imaging community. In this study, 91 images acquired utilizing a FLIR A320 thermal camera are used for developing an automatic breast segmentation from thermography images based on the random walker algorithm. A series of enhancement filters are applied to the image to make areola detection more accurate.Afterwards, the areola is detected using a series of circular Hough transforms. The detected areola region is then utilized for automatic seed placement for the random walker algorithm. Based on expert radiologist evaluation, the proposed segmentation algorithm was able to achieve 81.3% success rate in breast segmentation, while there was a 37.5% increase in the detection of breast cancer-related abnormalities by radiologists utilizing the segmented images, compared to utilizing original images.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Breast boundary segmentation in thermography images based on random walkers

Moghbel¹,

Mashohor²,

Mahmud³

et al. 2017

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

show abstract

“…Prior to experiments, breast regions in all frames are seg− mented from the background area, using a semi−automated algorithm developed by N. Scales et al [10] that employs Canny edge detection and the Hough transform to detect the symmetric breast boundaries and isolates the region of inte− rest. Breast segmentation was checked and altered where necessary using an algorithm embedded in the software 3D Slicer [11], and based on multiple label mapping and Otsu's thresholding.…”

Section: Methodsmentioning

confidence: 99%

Comparison of time-series registration methods in breast dynamic infrared imaging

Riyahi-Alam

Agostini

Molinari

et al. 2015

Opto-Electronics Review

View full text Add to dashboard Cite

Automated motion reduction in dynamic infrared imaging is on demand in clinical applications, since movement disarranges time−temperature series of each pixel, thus originating thermal artifacts that might bias the clinical decision. All previously proposed registration methods are feature based algorithms requiring manual intervention. The aim of this work is to optimize the registration strategy specifically for Breast Dynamic Infrared Imaging and to make it user−independent. We implemented and evaluated 3 different 3D time−series registration methods: 1. Linear affine, 2. Non−linear Bspline, 3. Demons applied to 12 datasets of healthy breast thermal images. The results are evaluated through normalized mutual information with average values of 0.70 ±0.03, 0.74 ±0.03 and 0.81 ±0.09 (out of 1) for Affine, Bspline and Demons registration, respectively, as well as breast boundary overlap and Jacobian determinant of the deformation field. The statistical analysis of the results showed that symmetric diffeomorphic Demons’ registration method outperforms also with the best breast alignment and non−negative Jacobian values which guarantee image similarity and anatomical consistency of the transformation, due to homologous forces enforcing the pixel geometric disparities to be shortened on all the frames. We propose Demons’ registration as an effective technique for time−series dynamic infrared registration, to stabilize the local temperature oscillation.

show abstract

“…It also has applications in thermal medical imaging [1][2][3]5]. This evolution of night vision cameras has encouraged the research in infrared image enhancement for information extraction from these images.…”

Section: Introductionmentioning

confidence: 99%

“…Enhancement aims at improving the visual quality of an image by reinforcing edges and smoothing flat areas. Several researchers have evaded this field using different approaches such as simple filtering, adaptive filtering, wavelet denoising, homomorphic enhancement etc., [1][2][3][4][5][6][7]. All these approaches concentrate on reinforcing the details of the image to be enhanced.…”

Section: Introductionmentioning

confidence: 99%

Homomorphic Enhancement of Infrared Images Using the Additive Wavelet Transform

Ashiba¹,

Awadalla²,

El-Halfawy³

et al. 2008

PIER C

View full text Add to dashboard Cite

Abstract-This paper presents a new enhancement technique for infrared images. This technique combines the benefits of homomorphic image processing and the additive wavelet transform. The idea behind this technique is based on decomposing the image into subbands in an additive fashion using the additive wavelet transform. This transform gives the image as an addition of subbands of the same resolution. The homomorphic processing is performed on each subband, separately. It is known that the homomorphic processing on images is performed in the log domain which transforms the image into illumination and reflectance components. Enhancement of the reflectance reinforces details in the image. So, applying this process in each subband enhances the details of the image in each subband. Finally, an inverse additive wavelet transform is performed on the homomorphic enhanced subbands to get an infrared image with better visual details.

show abstract

Automated image segmentation for breast analysis using infrared images

Cited by 40 publications

References 6 publications

Breast boundary segmentation in thermography images based on random walkers

Breast boundary segmentation in thermography images based on random walkers

Comparison of time-series registration methods in breast dynamic infrared imaging

Homomorphic Enhancement of Infrared Images Using the Additive Wavelet Transform

Contact Info

Product

Resources

About