E Likaki

University of Patras, Patrís, Kentriki Makedonia, Greece

Are you E Likaki?

Claim your profile

Publications (13)22.42 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: The purpose of this study is size-adapted segmentation of individual microcalcifications in mammography, based on microcalcification scale-space signature estimation, enabling robust scale selection for initialization of multiscale active contours. Segmentation accuracy was evaluated by the area overlap measure, by comparing the proposed method and two recently proposed ones to expert manual delineations. The method achieved area overlap of 0.61+/-0.15 outperforming statistically (p<0.001) the other two methods (0.53+/-0.18, 0.42+/-0.16). Only the proposed method performed equally for both small (< 460 microm) and large (>/= 460 microm) microcalcifications. Results indicate an accurate method, which could be utilized in computer-aided diagnosis schemes of microcalcification clusters.
    Computerized medical imaging and graphics: the official journal of the Computerized Medical Imaging Society 09/2010; 34(6):487-93. · 1.04 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Morphology of individual microcalcifications is an important clinical factor in microcalcification clusters diagnosis. Accurate segmentation remains a difficult task due to microcalcifications small size, low contrast, fuzzy nature and low distinguishability from surrounding tissue. A novel application of active rays (polar transformed active contours) on B-spline wavelet representation is employed, to provide initial estimates of microcalcification boundary. Then, a region growing method is used with pixel aggregation constrained by the microcalcification boundary estimates, to obtain the final microcalcification boundary. The method was tested on dataset of 49 microcalcification clusters (30 benign, 19 malignant), originating from the DDSM database. An observer study was conducted to evaluate segmentation accuracy of the proposed method, on a 5-point rating scale (from 5:excellent to 1:very poor). The average accuracy rating was 3.98±0.81 when multiscale active rays were combined to region growing and 2.93±0.92 when combined to linear polynomial fitting, while the difference in rating of segmentation accuracy was statistically significant (p < 0.05).
    Journal of Instrumentation 07/2009; 4(07):P07009. · 1.66 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The current study investigates texture properties of the tissue surrounding microcalcification (MC) clusters on mammograms for breast cancer diagnosis. The case sample analyzed consists of 85 dense mammographic images, originating from the Digital Database for Screening Mammography. Mammograms analyzed contain 100 subtle MC clusters (46 benign and 54 malignant). The tissue surrounding MCs is defined on original and wavelet decomposed images, based on a redundant discrete wavelet transform. Gray-level texture and wavelet coefficient texture features at three decomposition levels are extracted from surrounding tissue regions of interest (ST-ROIs). Specifically, gray-level first-order statistics, gray-level cooccurrence matrices features, and Laws' texture energy measures are extracted from original image ST-ROIs. Wavelet coefficient first-order statistics and wavelet coefficient cooccurrence matrices features are extracted from subimages ST-ROIs. The ability of each feature set in differentiating malignant from benign tissue is investigated using a probabilistic neural network. Classification outputs of most discriminating feature sets are combined using a majority voting rule. The proposed combined scheme achieved an area under receiver operating characteristic curve ( A(z)) of 0.989. Results suggest that MCs' ST texture analysis can contribute to computer-aided diagnosis of breast cancer.
    IEEE transactions on information technology in biomedicine: a publication of the IEEE Engineering in Medicine and Biology Society 12/2008; 12(6):731-8. · 1.69 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Accurate Microcalcification (MC) segmentation is a crucial first step in morphology based computer aided diagnosis systems for microcalcifications in mammography. In this article we present an automated segmentation method of individual MCs adaptive to both size and shape variations. Size is estimated by active rays (polar-transformed active contours) on continuous wavelet representation while shape adaptivity is achieved by a subsequent region growing step. Following MC seed point annotation, contour point estimates are obtained by implementing active rays on an analytic scale-space representation in a coarse-to-fine strategy. Initial coarsest scale is automatically defined by analyzing MC responses across scales. A region growing method is used to delineate the final MC contour curve, with pixel aggregation constrained by the MC contour point estimates. The segmentation accuracy of the proposed method was quantitatively evaluated by means of area overlap by comparing automatically derived borders with manually traced ones provided by an expert radiologist. The proposed method achieved an area overlap of 0.68plusmn0.13 on a dataset of 67 individual microcalcifications, originating from pleomorphic clusters.
    BioInformatics and BioEngineering, 2008. BIBE 2008. 8th IEEE International Conference on; 11/2008
  • [Show abstract] [Hide abstract]
    ABSTRACT: Accurate segmentation of microcalcifications in mammography is crucial for the quantification of morphologic properties by features incorporated in computer-aided diagnosis schemes. A novel segmentation method is proposed implementing active rays (polar-transformed active contours) on B-spline wavelet representation to identify microcalcification contour point estimates in a coarse-to-fine strategy at two levels of analysis. An iterative region growing method is used to delineate the final microcalcification contour curve, with pixel aggregation constrained by the microcalcification contour point estimates. A radial gradient-based method was also implemented for comparative purposes. The methods were tested on a dataset consisting of 149 mainly pleomorphic microcalcification clusters originating from 130 mammograms of the DDSM database. Segmentation accuracy of both methods was evaluated by three radiologists, based on a five-point rating scale. The radiologists' average accuracy ratings were 3.96 +/- 0.77, 3.97 +/- 0.80, and 3.83 +/- 0.89 for the proposed method, and 2.91 +/- 0.86, 2.10 +/- 0.94, and 2.56 +/- 0.76 for the radial gradient-based method, respectively, while the differences in accuracy ratings between the two segmentation methods were statistically significant (Wilcoxon signed-ranks test, p < 0.05). The effect of the two segmentation methods in the classification of benign from malignant microcalcification clusters was also investigated. A least square minimum distance classifier was employed based on cluster features reflecting three morphological properties of individual microcalcifications (area, length, and relative contrast). Classification performance was evaluated by means of the area under ROC curve (Az). The area and length morphologic features demonstrated a statistically significant (Mann-Whitney U-test, p < 0.05) higher patient-based classification performance when extracted from microcalcifications segmented by the proposed method (0.82 +/- 0.06 and 0.86 +/- .05, respectively), as compared to segmentation by the radial gradient-based method (0.71 +/- 0.08 and 0.75 +/- 0.08). The proposed method demonstrates improved segmentation accuracy, fulfilling human visual criteria, and enhances the ability of morphologic features to characterize microcalcification clusters.
    Medical Physics 11/2008; 35(11):5161-71. · 2.91 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Diagnosis of microcalcifications (MCs) is challenged by the presence of dense breast parenchyma, resulting in low specificity values and thus in unnecessary biopsies. The current study investigates whether texture properties of the tissue surrounding MCs can contribute to breast cancer diagnosis. A case sample of 100 biopsy-proved MC clusters (46 benign, 54 malignant) from 85 dense mammographic images, included in the Digital Database for Screening Mammography, was analysed. Regions of interest (ROIs) containing the MCs were pre-processed using a wavelet-based contrast enhancement method, followed by local thresholding to segment MCs; the segmented MCs were excluded from original image ROIs, and the remaining area (surrounding tissue) was subjected to texture analysis. Four categories of textural features (first order statistics, co-occurrence matrices features, run length matrices features and Laws' texture energy measures) were extracted from the surrounding tissue. The ability of each feature category in discriminating malignant from benign tissue was investigated using a k-nearest neighbour (kNN) classifier. An additional classification scheme was performed by combining classification outputs of three textural feature categories (the most discriminating ones) with a majority voting rule. Receiver operating characteristic (ROC) analysis was conducted for classifier performance evaluation of the individual textural feature categories and of the combined classification scheme. The best performance was achieved by the combined classification scheme yielding an area under the ROC curve (A(z)) of 0.96 (sensitivity 94.4%, specificity 80.0%). Texture analysis of tissue surrounding MCs shows promising results in computer-aided diagnosis of breast cancer and may contribute to the reduction of unnecessary biopsies.
    The British journal of radiology 09/2007; 80(956):648-56. · 2.11 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The current study investigates whether texture properties of the tissue surrounding microcalcification (MC) clusters can contribute to breast cancer diagnosis. The case sample analyzed consists of 100 mammographic images, originating from the Digital Database for Screening Mammography (DDSM). All mammograms selected correspond to heterogeneously and extremely dense breast parenchyma and contain subtle MC clusters (46 benign and 54 malignant, according to database ground truth tables). Regions of interest (ROIs) of 128x128 pixels, containing the MCs are used for the subsequent texture analysis. ROIs are preprocessed using a wavelet-based locally adapted contrast enhancement method and a thresholding technique is applied to exclude MCs. Texture features are extracted from the remaining ROI area (surrounding tissue) employing first and second order statistics algorithms, grey level run length matrices and Laws' texture energy measures. Differentiation between malignant and benign MCs is performed using a k-nearest neighbour (kNN) classifier and employing the leave-one-out training-testing methodology. The Laws' texture energy measures demonstrated the highest performance achieving an overall accuracy of 89%, sensitivity 90.74% (49 of 54 malignant cases classified correctly) and specificity 86.96% (40 of the 46 benign cases classified correctly). Texture analysis of the tissue surrounding MCs shows promising results in computer-aided diagnosis of breast cancer and may contribute to the reduction of benign biopsies.
    01/2006;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Presence of dense parenchyma in mammographic images masks lesions resulting in either missed detections or mischaracterizations, thus decreasing mammographic sensitivity and specificity. The aim of this study is evaluating the effect of a wavelet enhancement method on dense parenchyma for a lesion contour characterization task, using simulated lesions. The method is recently introduced, based on a two-stage process, locally adaptive denoising by soft-thresholding and enhancement by linear stretching. Sixty simulated low-contrast lesions of known image characteristics were generated and embedded in dense breast areas of normal mammographic images selected from the DDSM database. Evaluation was carried out by an observer performance comparative study between the processed and initial images. The task for four radiologists was to classify each simulated lesion with respect to contour sharpness/unsharpness. ROC analysis was performed. Combining radiologists' responses, values of the area under ROC curve (Az) were 0.93 (95% CI 0.89, 0.96) and 0.81 (CI 0.75, 0.86) for processed and initial images, respectively. This difference in Az values was statistically significant (Student's t-test, P<0.05), indicating the effectiveness of the enhancement method. The specific wavelet enhancement method should be tested for lesion contour characterization tasks in softcopy-based mammographic display environment using naturally occurring pathological lesions and normal cases.
    European Radiology 09/2005; 15(8):1615-22. · 4.34 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The purpose of this study is the use of simulated low contrast radiopaque lesions embedded in dense parenchyma for detection accuracy assessment of two contrast enhancement methods in mammographic imaging. Fifty (50) simulated lesions of known image characteristics, such as contrast and contour, were generated using a lesion simulation tool developed in our department and superimposed at dense parenchyma of mammographic images selected from the DDSM database. The entire sample, 50 with lesions and 50 normal, was processed by an adaptive wavelet (AW) method and an intensity windowing (IW) one. The task for two radiologists was to detect the simulated lesions using a five-point rating scale. A ROC curve was individually fitted to the scores of each radiologist and the area under the ROC curve (A z) was calculated for each method. Combining radiologists' responses, the A z values were 0.92 and 0.89 and their corresponding confidence intervals were (0.85, 0.95) and (0.83, 0.93) for AW and IW methods, respectively. This difference in A z values is not statistically significant (Students' t-test, p>0.05), indicating similar detection accuracy for both methods. Use of simulated lesions in difficult case samples servers verification purposes required to assess the performance of various digital image post-processing methods.
    01/2004;
  • [Show abstract] [Hide abstract]
    ABSTRACT: The aim of this study was to determine the visually lossless threshold of a wavelet-based compression algorithm in case of microcalcification cluster detection in mammography. The threshold was determined by means of observer performance using a set of digitized mammograms. In addition, the transfer characteristics of the compression algorithm were assessed by means of image-quality parameters using computer-generated test images. The observer performance study was based on rating performed by four independent radiologists, who reviewed 68 mammograms, from the Digital Database for Screening Mammography (DDSM), at six different compression ratios. Receiver operating characteristics (ROC) analysis was performed on observers' responses and the area under ROC curve (A(z)) was calculated at each compression ratio for each observer. The parameters used for assessment of transfer characteristics of the compression algorithm were input/output response, noise, high-contrast response, and low-contrast-detail response. The computer-generated test image, used for this assessment, mimicked mammographic image characteristics (pixel size, pixel depth, and noise) as well as microcalcification characteristics (size and contrast). The ROC analysis for microcalcification cluster detection indicated a threshold at compression ratio 40:1, as Student's t-test shows statistically significant differences in A(z) values (p<0.05) for compression ratios 70:1 and 100:1. Observers' grading of mammogram quality lowers this threshold at 25:1. Low-contrast-detail detectability in the transfer characteristics study indicate a threshold of 35:1, whereas non-perceptibility of image-quality-parameters degradation lowers this threshold to 30:1. The ROC and transfer characteristics analysis provided comparable thresholds, indicating the potential use of the latter in limiting the target range of compression ratios for subsequent observer studies.
    European Radiology 10/2003; 13(10):2390-6. · 4.34 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Optimization performance of digital image post-processing techniques in mammography requires controlled conditions of data sets permitting quantitative representation of image characteristics of pathological findings. Digital test objects, although objective and quantitative, do not mimic mammographic appearance and clinical data sets do not provide adequate sets of values of the various pathological finding characteristics. This can be overcome by digital simulation of pathological findings and superimposition on mammographic images. A simple method for simulation of mammographic appearance of radiopaque and/or radiolucent circumscribed lesions is presented. Circumscribed lesions are simulated using grey-level transformation functions which shift and compress the range of the initial pixel grey-level values in a region of interest (ROI) of a digitized mammographic image, according to grey-level analysis in 200 ROIs of real circumscribed lesions from digitized mammographic images. Simulation addresses lesion image characteristics, such as elliptical shape, orientation, halo sign for radiopaque lesions and capsule for radiolucent lesions, and is implemented in a user-driven PC-based interactive application. The appearance of the lesions is evaluated by six radiologists on a sample of 60 real and 60 simulated radiopaque lesions with the use of receiver operating characteristic (ROC) analysis. The area under the ROC curve, pooling the responses of the observers, was 0.55+/-0.03 indicating no statistically significant difference between real and simulated lesions (p>0.05). The method adequately simulates the mammographic appearance of circumscribed lesions and could be used to generate circumscribed lesion data sets for performance evaluation of image processing techniques, as well as education purposes.
    European Radiology 05/2003; 13(5):1137-47. · 4.34 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The aim of this study is to evaluate the performance of a wavelet-based image compression algorithm in mammography. As microcalcifications detection is selected as diagnostic task, the algorithm is evaluated with respect to its visually lossless threshold at low compression ratios and not its supra-threshold performance. The threshold is determined by means of observer performance using a set of digitized mammograms from the Digital Database for Screening Mammography and by means of an image quality study, related to transfer characteristics of the compression algorithm, using computer-generated test images. The image quality parameters used are input/output response, noise, high contrast response and low contrast-detail response. The computer-generated test images mimic mammographic image and microcalcification characteristics. Receiver operating characteristics analysis of pooled data for microcalcification detection in mammograms indicated a threshold at compression ratio 40:1. Image quality parameters assessment, with respect to low contrast-detail patterns, indicated a threshold at compression ratio 35:1. The two approaches provided comparable thresholds, indicating the potential use of image quality parameters.
    Digital Signal Processing, 2002. DSP 2002. 2002 14th International Conference on; 02/2002
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In mammographic imaging, characterization of lesions (microcalcifications and masses) is a demanding task for radiologists, especially in breasts with dense parenchyma. The aim of this study is the comparative evaluation between an edge enhancement (Adaptive Wavelet, AW) and a contrast enhancement (Intensity Windowing, IW) method, in characterization of lesions at presence of dense parenchyma. The sample for microcalcification characterization consists of 43 mammograms containing microcalcification clusters (29 malignant and 14 benign) and the sample for mass characterization consists of 50 mammograms containing masses (30 malignant and 20 benign). All mammograms correspond to heterogeneously or extremely dense breasts, originating from the DDSM database. The samples were processed with AW and IW enhancement methods and reviewed by two experienced radiologists. Comparative evaluation was performed between AW- processed and IW-processed images, based on imaging characteristics related to characterization (microcalcification cluster criteria: morphology, size, number and density; mass criteria: contrast, contour, shape, orientation and size). Similar performance (~70%) is achieved for both enhancement methods in microcalcification cluster characterization (Wilcoxon signed ranks test, p>0.05). The AW enhancement method demonstrates statistically significant improvement (ROC analysis, p