Rejected images represent both unnecessary radiation exposure to patients and inefficiency in the imaging operation. Rejected images are inherent to projection radiography, where patient positioning and alignment are integral components of image quality. Patient motion and artifacts unique to digital image receptor technology can result in rejected images also. We present a centralized, server-based solution for the collection, archival, and distribution of rejected image and exposure indicator data that automates the data collection process. Reject analysis program (RAP) and exposure indicator data were collected and analyzed during a 1-year period. RAP data were sorted both by reason for repetition and body part examined. Data were also stratified by clinical area for further investigation. The monthly composite reject rate for our institution fluctuated between 8% and 10%. Positioning errors were the main cause of repeated images (77.3%). Stratification of data by clinical area revealed that areas where computed radiography (CR) is seldom used suffer from higher reject rates than areas where it is used frequently. S values were log-normally distributed for examinations performed under either manual or automatic exposure control. The distributions were positively skewed and leptokurtic. S value decreases due to radiologic technology student rotations, and CR plate reader calibrations were observed. Our data demonstrate that reject analysis is still necessary and useful in the era of digital imaging. It is vital though that analysis be combined with exposure indicator analysis, as digital radiography is not self-policing in terms of exposure. When combined, the two programs are a powerful tool for quality assurance.

Multidetector row computed tomography (MDCT) creates massive amounts of data, which can overload a picture archiving and communication system (PACS). To solve this problem, we designed a new data storage and image interpretation system in an existing PACS. Two MDCT image datasets, a thick- and a thin-section dataset, and a single-detector CT thick-section dataset were reconstructed. The thin-section dataset was archived in existing PACS disk space reserved for temporary storage, and the system overwrote the source data to preserve available disk space. The thick-section datasets were archived permanently. Multiplanar reformation (MPR) images were reconstructed from the stored thin-section datasets on the PACS workstation. In regular interpretations by eight radiologists during the same week, the volume of images and the times taken for interpretation of thick-section images with (246 CT examinations) or without (170 CT examinations) thin-section images were recorded, and the diagnostic usefulness of the thin-section images was evaluated. Thin-section datasets and MPR images were used in 79% and 18% of cases, respectively. The radiologists’ assessments of this system were useful, though the volume of images and times taken to archive, retrieve, and interpret thick-section images together with thin-section images were significantly greater than the times taken without thin-section images. The limitations were compensated for by the usefulness of thin-section images. This data storage and image interpretation system improves the storage and availability of the thin-section datasets of MDCT and can prevent overloading problems in an existing PACS for the moment.

Intracranial aneurysms represent a significant cause of morbidity and mortality. While the risk factors for aneurysm formation are known, the detection of aneurysms remains challenging. Magnetic resonance angiography (MRA) has recently emerged as a useful non-invasive method for aneurysm detection. However, even for experienced neuroradiologists, the sensitivity to small (<5 mm) aneurysms in MRA images is poor, on the order of 30~60% in recent, large series. We describe a fully automated computer-aided detection (CAD) scheme for detecting aneurysms on 3D time-of-flight (TOF) MRA images. The scheme locates points of interest (POIs) on individual MRA datasets by combining two complementary techniques. The first technique segments the intracranial arteries automatically and finds POIs from the segmented vessels. The second technique identifies POIs directly from the raw, unsegmented image dataset. This latter technique is useful in cases of incomplete segmentation. Following a series of feature calculations, a small fraction of POIs are retained as candidate aneurysms from the collected POIs according to predetermined rules. The CAD scheme was evaluated on 287 datasets containing 147 aneurysms that were verified with digital subtraction angiography, the accepted standard of reference for aneurysm detection. For two different operating points, the CAD scheme achieved a sensitivity of 80% (71% for aneurysms less than 5 mm) with three mean false positives per case, and 95% (91% for aneurysms less than 5 mm) with nine mean false positives per case. In conclusion, the CAD scheme showed good accuracy and may have application in improving the sensitivity of aneurysm detection on MR images.

A fully automated and three-dimensional (3D) segmentation method for the identification of the pulmonary parenchyma in thorax X-ray computed tomography (CT) datasets is proposed. It is meant to be used as pre-processing step in the computer-assisted detection (CAD) system for malignant lung nodule detection that is being developed by the Medical Applications in a Grid Infrastructure Connection (MAGIC-5) Project. In this new approach the segmentation of the external airways (trachea and bronchi), is obtained by 3D region growing with wavefront simulation and suitable stop conditions, thus allowing an accurate handling of the hilar region, notoriously difficult to be segmented. Particular attention was also devoted to checking and solving the problem of the apparent ‘fusion’ between the lungs, caused by partial-volume effects, while 3D morphology operations ensure the accurate inclusion of all the nodules (internal, pleural, and vascular) in the segmented volume. The new algorithm was initially developed and tested on a dataset of 130 CT scans from the Italung-CT trial, and was then applied to the ANODE09-competition images (55 scans) and to the LIDC database (84 scans), giving very satisfactory results. In particular, the lung contour was adequately located in 96% of the CT scans, with incorrect segmentation of the external airways in the remaining cases. Segmentation metrics were calculated that quantitatively express the consistency between automatic and manual segmentations: the mean overlap degree of the segmentation masks is 0.96 ± 0.02, and the mean and the maximum distance between the mask borders (averaged on the whole dataset) are 0.74 ± 0.05 and 4.5 ± 1.5, respectively, which confirms that the automatic segmentations quite correctly reproduce the borders traced by the radiologist. Moreover, no tissue containing internal and pleural nodules was removed in the segmentation process, so that this method proved to be fit for the use in the framework of a CAD system. Finally, in the comparison with a two-dimensional segmentation procedure, inter-slice smoothness was calculated, showing that the masks created by the 3D algorithm are significantly smoother than those calculated by the 2D-only procedure.

The workflow in radiology departments has changed dramatically with the transition to digital PACS, especially with the shift from tile mode to stack mode display of volumetric images. With the increasing number of images in routinely captured datasets, the standard user interface devices (UIDs) become inadequate. One basic approach to improve the navigation of the stack mode datasets is to take advantage of alternative UIDs developed for other domains, such as the computer game industry. We evaluated three UIDs both in clinical practice and in a task-based experiment. After using the devices in the daily image interpretation work, the readers reported that both of the tested alternative UIDs were better in terms of ergonomics compared to the standard mouse and that both alternatives were more efficient when reviewing large CT datasets. In the task-based experiment, one of the tested devices was faster than the standard mouse, while the other alternative was not significantly faster. One of the tested alternative devices showed a larger number of traversed images during the task. The results indicate that alternative user interface devices can improve the navigation of stack mode datasets and that radiologists should consider the potential benefits of alternatives to the standard mouse.

Authenticating medical images using watermarking techniques has become a very popular area of research, and some works in this area have been reported worldwide recently. Besides authentication, many data-hiding techniques have been proposed to conceal patient’s data into medical images aiming to reduce the cost needed to store data and the time needed to transmit data when required. In this paper, we present a new hybrid watermarking scheme for DICOM images. In our scheme, two well-known techniques are combined to gain the advantages of both and fulfill the requirements of authentication and data hiding. The scheme divides the images into two parts, the region of interest (ROI) and the region of non-interest (RONI). Patient’s data are embedded into ROI using a reversible technique based on difference expansion, while tamper detection and recovery data are embedded into RONI using a robust technique based on discrete wavelet transform. The experimental results show the ability of hiding patient’s data with a very good visual quality, while ROI, the most important area for diagnosis, is retrieved exactly at the receiver side. The scheme also shows some robustness against certain levels of salt and pepper and cropping noise.

The objective of this study was to compare the diagnostic accuracy in the interpretation of chest nodules using original CT images versus enhanced CT images based on the wavelet transform. The CT images of 118 patients with cancers and 60 with benign nodules were used in this study. All images were enhanced through an algorithm based on the wavelet transform. Two experienced radiologists interpreted all the images in two reading sessions. The reading sessions were separated by a minimum of 1 month in order to minimize the effect of observer’s recall. The Mann–Whitney U nonparametric test was used to analyze the interpretation results between original and enhanced images. The Kruskal–Wallis H nonparametric test of K independent samples was used to investigate the related factors which could affect the diagnostic accuracy of observers. The area under the ROC curves for the original and enhanced images was 0.681 and 0.736, respectively. There is significant difference in diagnosing the malignant nodules between the original and enhanced images (z = 7.122, P < 0.001), whereas there is no significant difference in diagnosing the benign nodules (z = 0.894, P = 0.371). The results showed that there is significant difference between original and enhancement images when the size of nodules was larger than 2 cm (Z = −2.509, P = 0.012, indicating the size of the nodules is a critical evaluating factor of the diagnostic accuracy of observers). This study indicated that the image enhancement based on wavelet transform could improve the diagnostic accuracy of radiologists for the malignant chest nodules.

In current radiologists’ workstations, a scroll mouse is typically used as the primary input device for navigating image slices and conducting operations on an image. Radiological analysis and diagnosis rely on careful observation and annotation of medical images. During analysis of 3D MRI and CT volumes, thousands of mouse clicks are performed everyday, which can cause wrist fatigue. This paper presents a dynamic control-to-display (C-D) gain mouse movement method, controlled by an eyegaze tracker as the target predictor. By adjusting the C-D gain according to the distance to the target, the mouse click targeting time is reduced. Our theoretical and experimental studies show that the mouse movement time to a known target can be reduced by up to 15%. We also present an experiment with 12 participants to evaluate the role of eyegaze targeting in the realistic situation of unknown target positions. These results indicate that using eyegaze to predict the target position, the dynamic C-D gain method can improve pointing performance by 8% and reduce the error rate over traditional mouse movement.

This paper addresses the need to quantify tumor growth and detect changes as this information is relevant to manage the patient treatment and to aid biotechnological efforts to cure cancer (Silva et al. 2008). An interactive tumor segmentation technique is used to recover the shape and size of tumors without imposing shape constraints. This segmentation algorithm provides good convergence, is robust to the initialization conditions, and requires simple and intuitive user interactions. A parametric approach to model tumor growth analytically is proposed in this paper. The preliminary experimental results are encouraging. The segmentation method is shown to be robust and simple to use, even in situations where the tumor boundary definition is challenging. Also, the experiments indicate that the proposed model potentially can be used to extrapolate the available data and help predict the tumor size (assuming unconstrained growth). Additionally, the proposed method potentially can provide a quantitative reference to compare the tumor shrinkage rate in cancer treatments.

We evaluated the use of a stylus as a computer interface for radiographic image annotation. Our case study concerned the annotation of spiculated lesions on mammograms. Three experienced radiologists annotated 20 mammograms depicting spiculated lesions. We evaluated the interobserver agreement in annotations marked with a stylus versus those marked with a mouse using the intraclass correlation coefficient. Better agreement in annotating spicule width was observed with the stylus, suggesting that it is easier to accurately annotate subtle regions on an image using a stylus.

A virtual medical imaging department is an innovative and demanding organizational model, to the extent that the underlying goal is to achieve a continuous and advanced organizational integration of human and physical resources, clinical data, and clienteles. To better understand the kind of benefits offered, we conducted a survey of three groups of users—radiologists, radiological technologists, and medical specialists—working in a five-site virtual organization. We received 127 valid questionnaires, for an overall response rate of 66%. The assessments vary according to the use made of the system. The scores for system quality and the quality of the data produced were markedly higher for intra-hospital use (respectively 7.9 and 8.7 out of 10) than for inter-hospital use (5.4 and 7.0). Despite the negative assessments they made of inter-hospital use, users maintained a positive attitude toward some type of virtual organization of medical imaging. Indeed, the score for Overall satisfaction with the system was very high, 8.9 out of 10. Moreover, the scores for Intended future use of the system were very high for both intra-hospital use (8.9) and inter-hospital use (8.7). We also found significant differences in perceptions among user groups.

To address potential concern for cumulative radiation exposure with serial spiral chest computed tomography (CT) scans in children with chronic lung disease, we developed an approach to match bronchial airways on low-dose spiral and low-dose high-resolution CT (HRCT) chest images to allow serial comparisons. An automated algorithm matches the position and orientation of bronchial airways obtained from HRCT slices with those in the spiral CT scan. To validate this algorithm, we compared manual matching vs automatic matching of bronchial airways in three pediatric patients. The mean absolute percentage difference between the manually matched spiral CT airway and the index HRCT airways were 9.4 ± 8.5% for the internal diameter measurements, 6.0 ± 4.1% for the outer diameter measurements, and 10.1 ± 9.3% for the wall thickness measurements. The mean absolute percentage difference between the automatically matched spiral CT airway measurements and index HRCT airway measurements were 9.2 ± 8.6% for the inner diameter, 5.8 ± 4.5% for the outer diameter, and 9.9 ± 9.5% for the wall thickness. The overall difference between manual and automated methods was 2.1 ± 1.2%, which was significantly less than the interuser variability of 5.1 ± 4.6% (p < 0.05). Tests of equivalence had p < 0.05, demonstrating no significant difference between the two methods. The time required for matching was significantly reduced in the automated method (p < 0.01) and was as accurate as manual matching, allowing efficient comparison of airways obtained on low-dose spiral CT imaging with low-dose HRCT scans.

Electronic supplementary material

The online version of this article (doi:10.1007/s10278-009-9199-3) contains supplementary material, which is available to authorized users.

Single standard anteroposterior radiograph-based methods for measuring cup orientation following total hip arthroplasty (THA) are subject to substantial errors if the individual pelvic orientation with respect to X-ray plate is not taken into consideration. Previously, we proposed to use a hybrid 2D–3D registration scheme to determine the postoperative acetabular cup orientation and developed an object-oriented cross-program called “HipMatch.” However, its accuracy and robustness have not been fully investigated. To assess the potential factors that may affect the accuracy and robustness of the hybrid 2D–3D registration scheme in determining the postoperative acetabular cup orientation, a comprehensive validation study using a cadaver pelvis was performed. Nine X-ray radiographs taken from different pelvic positions relative to the X-ray plate and two computed tomography volumes of the pelvis with one acquired before the cup implantation and the other acquired after the cup implantation were used in the validation study. Potential factors that may affect the accuracy and robustness of the hybrid 2D–3D registration scheme were experimentally determined. Our experimental results demonstrate that (1) the plain radiograph-based method is not accurate; (2) the hybrid 2D–3D registration scheme helps to improve the estimation accuracy; (3) the hybrid 2D–3D registration scheme can robustly and accurately estimate the cup orientation even when a big portion of the radiograph is occluded; and (4) image resolution has minor effect on the estimation accuracy. The hybrid 2D–3D registration scheme is an accurate and robust method to measure exact cup orientation in THA. It holds the promise to be a valuable tool for clinical routine usage for providing evidence-based information.

Automatic bone segmentation of computed tomography (CT) images is an important step in image-guided surgery that requires both high accuracy and minimal user interaction. Previous attempts include global thresholding, region growing, region competition, watershed segmentation, and parametric active contour (AC) approaches, but none claim fully satisfactory performance. Recently, geometric or level-set-based AC models have been developed and appear to have characteristics suitable for automatic bone segmentation such as initialization insensitivity and topology adaptability. In this study, we have tested the feasibility of five level-set-based AC approaches for automatic CT bone segmentation with both synthetic and real CT images: namely, the geometric AC, geodesic AC, gradient vector flow fast geometric AC, Chan–Vese (CV) AC, and our proposed density distance augmented CV AC (Aug. CV AC). Qualitative and quantitative evaluations have been made in comparison with the segmentation results from standard commercial software and a medical expert. The first three models showed their robustness to various image contrasts, but their performances decreased much when noise level increased. On the contrary, the CV AC’s performance was more robust to noise, yet dependent on image contrast. On the other hand, the Aug. CV AC demonstrated its robustness to both noise and contrast levels and yielded improved performances on a set of real CT data compared with the commercial software, proving its suitability for automatic bone segmentation from CT images.

Workflow efficiency is a crucial factor in selecting computed radiography (CR) versus digital radiography (DR) systems for digital projection radiography operations. DR systems can be more efficient, but present higher costs and limitations in performing some radiographic exams. A newly developed CR system presents a good alternative with its faster line-by-line instead of pixel-by-pixel image plate-scanning technology and a more efficient workstation. To evaluate workflow characteristics, a time–motion study was conducted to compare radiographic exam times of the new CR system with traditional CR and DR systems in a high-volume orthopedic operation. Approximately 200 exams for each modality were documented from the moment when a patient entered the X-ray room to the moment when all images were sent to the PACS archive using a timer and speech-recognition software. Applying Welch ANOVA and Tamhane’s T2 tests, average exam times for the new CR system were significantly faster (18–42%; P ≤ 0.025) than for the traditional CR system. Average exam times for the DR system were also faster than for the new CR system by 22–36% (P < 0.001) with one exception. In the case where the new CR system was located outside the X-ray room, using a one-technologist workflow model, average single-study exam times were not significantly different from those found when using DR. Therefore, the new CR system may be comparable in efficiency with the DR system for this particular setting and operation.

Medical Imaging has been fortunate to see an avalanche of free and open source software become available in the last several years. Applications have been written to enable image viewing/storage/analysis/processing, DICOM and HL7 message parsing, results aggregation, anonymization, and more. While robust, many of these packages are difficult to install and configure. Our group desired an approach that would mitigate the efforts required to use these packages across different projects. We found such a solution in the context of using virtual machines.

Recently, several types of post-processing image filter which was designed to reduce noise allowing a corresponding dose reduction in CT images have been proposed and these were reported to be useful for noise reduction of CT images of adult patients. However, these have not been reported on adaptation for pediatric patients. Because they are not very effective with small (<20 cm) display fields of view, they could not be used for pediatric (e.g., premature babies and infants) body CT images. In order to solve this restriction, we have developed a new noise reduction filter algorithm which can be applicable for pediatric body CT images. This algorithm is based on a three-dimensional post processing, in which output pixel values are calculated by multi-directional, one-dimensional median filters on original volumetric datasets. The processed directions were selected except in in-plane (axial plane) direction, and consequently the in-plane spatial resolution was not affected by the filter. Also, in other directions, the spatial resolutions including slice thickness were almost maintained due to a characteristic of non-linear filtering of the median filter. From the results of phantom studies, the proposed algorithm could reduce standard deviation values as a noise index by up to 30% without affecting the spatial resolution of all directions, and therefore, contrast-to-noise ratio was improved by up to 30%. This newly developed filter algorithm will be useful for the diagnosis and radiation dose reduction of pediatric body CT images.

 

This study evaluates the accuracy of augmenting initial intraprocedural computed tomography (CT) during radiofrequency ablation (RFA) of hepatic metastases with preprocedural positron emission tomography (PET) through a hardware-accelerated implementation of an automatic nonrigid PET–CT registration algorithm. The feasibility of augmenting intraprocedural CT with preprocedural PET to improve localization of CT-invisible but PET-positive tumors with images from actual RFA was explored. Preprocedural PET and intraprocedural CT images from 18 cases of hepatic RFA were included. All PET images in the study originated from a hybrid PET/CT scanner, and PET–CT registration was performed in two ways: (1) direct registration of preprocedural PET with intraprocedural CT and (2) indirect registration of preprocedural CT (i.e., the CT of hybrid PET/CT scan) with intraprocedural CT. A hardware-accelerated registration took approximately 2 min. Calculated registration errors were 7.0 and 8.4 mm for the direct and indirect methods, respectively. Overall, the direct registration was found to be statistically not distinct from that performed by a group of clinical experts. The accuracy, execution speed, and compactness of our implementation of nonrigid image registration suggest that existing PET can be overlaid on intraprocedural CT, promising a novel, technically feasible, and clinically viable approach for PET augmentation of CT guidance of RFA.

We developed positron emission tomography (PET)/computed tomography (CT) viewing software (PETviewer) that can display co-registered PET and CT images obtained by PET/CT and stored on picture archiving and communication systems (PACS). PETviewer has tools for presetting windows for CT display; control bars for PET window level; zoom, pan, and pseudo-color functions; and allows the user to draw a rectangular region of interest (ROI) for standardized uptake value (SUV) measurement. SUV was calculated using PET DICOM header information and the pixel intensity in PETviewer. Reconstructed datasets of PET/CT and maximum intensity projection (MIP) of the PET images were transferred and archived in PACS. Phantom experiments were performed to evaluate the validity of image fusion. PET/CT images were displayed on an independent window in PACS. Transaxial PET images were reformatted as sagittal and coronal PET images, which were displayed with the corresponding CT and PET/CT fusion images with adjustable color and transparency. Transaxial, sagittal, and coronal PET images corresponding to the location of the cursor were shown using cine display of MIP images. All images were displayed in PETviewer within 20 s on a personal computer for PACS, which was equipped with a P4, 1.3-GHz CPU, and 512 Mb of RAM. We could measure maximum and mean SUV in a ROI using PETviewer. Transaxial fused images of patients and phantoms showed excellent registration and fusion of PET and CT images in the X and Y directions. PETviewer provided very useful clinical tools for assessing PET/CT images on PACS and should assist in maximizing the benefits derived from PET/CT imaging.

In this paper, a novel fast method for modeling mammograms by deterministic fractal coding approach to detect the presence of microcalcifications, which are early signs of breast cancer, is presented. The modeled mammogram obtained using fractal encoding method is visually similar to the original image containing microcalcifications, and therefore, when it is taken out from the original mammogram, the presence of microcalcifications can be enhanced. The limitation of fractal image modeling is the tremendous time required for encoding. In the present work, instead of searching for a matching domain in the entire domain pool of the image, three methods based on mean and variance, dynamic range of the image blocks, and mass center features are used. This reduced the encoding time by a factor of 3, 89, and 13, respectively, in the three methods with respect to the conventional fractal image coding method with quad tree partitioning. The mammograms obtained from The Mammographic Image Analysis Society database (ground truth available) gave a total detection score of 87.6%, 87.6%, 90.5%, and 87.6%, for the conventional and the proposed three methods, respectively.

The aim of this study was to assess the image display of a web-based teleradiology system that uses a common web browser and has no need of proprietary applets, plug-ins, or dedicated software for DICOM display. The teleradiology system (TS) is connected to the Internet by ADSL and to radiological modalities using the DICOM standard with TCP/IP. Images were displayed on a PC through Internet connection with the remote TS using a common web browser. MS lesion number and volume in T1- and T2-weighted images (T1w and T2w, respectively) of 30 brain MR studies were quantified using both the TS and a conventional software. Wilcoxon signed ranks test and intraclass correlation coefficient (ICC) were used to assess the variability and concordance between intra- and inter-observer and TS and conventional DICOM viewer, setting significance at p < 0.05. No significant differences in T1w and T2w volumes between the TS and the conventional software were found by either operator. The ICC results showed a high level of inter-operator agreement in volume estimation in T1w and T2w images using the two systems. Quantitative assessment of MS lesion volumes in T1w and T2w images with a user interface of a teleradiology system that allows the consultation by means of a common web browser, without the need for proprietary plug-ins, applets, or dedicated software for DICOM display showed no significant differences from, and almost complete agreement with, conventional DICOM viewers.

Studies reported in the literature indicate that the increase in the breast density is one of the strongest indicators of developing breast cancer. In this paper, we present an approach to automatically evaluate the density of a breast by segmenting its internal parenchyma in either fatty or dense class. Our approach is based on a statistical analysis of each pixel neighbourhood for modelling both tissue types. Therefore, we provide connected density clusters taking the spatial information of the breast into account. With the aim of showing the robustness of our approach, the experiments are performed using two different databases: the well-known Mammographic Image Analysis Society digitised database and a new full-field digital database of mammograms from which we have annotations provided by radiologists. Quantitative and qualitative results show that our approach is able to correctly detect dense breasts, segmenting the tissue type accordingly.

The effect of pixel resolution on texture features computed using the gray-level co-occurrence matrix (GLCM) was analyzed in the task of discriminating mammographic breast lesions as benign masses or malignant tumors. Regions in mammograms related to 111 breast masses, including 65 benign masses and 46 malignant tumors, were analyzed at pixel sizes of 50, 100, 200, 400, 600, 800, and 1,000 μm. Classification experiments using each texture feature individually provided accuracy, in terms of the area under the receiver operating characteristics curve (AUC), of up to 0.72. Using the Bayesian classifier and the leave-one-out method, the AUC obtained was in the range 0.73 to 0.75 for the pixel resolutions of 200 to 800 μm, with 14 GLCM-based texture features using adaptive ribbons of pixels around the boundaries of the masses. Texture features computed using the ribbons resulted in higher classification accuracy than the same features computed using the corresponding regions within the mass boundaries. The t test was applied to AUC values obtained using 100 repetitions of random splitting of the texture features from the ribbons of masses into the training and testing sets. The texture features computed with the pixel size of 200 μm provided the highest average AUC with statistically highly significant differences as compared to all of the other pixel sizes tested, except 100 μm.

Mammograms are X-ray images of human breast which are normally used to detect breast cancer. The presence of pectoral muscle in mammograms may disturb the detection of breast cancer as the pectoral muscle and mammographic parenchyma appear similar. So, the suppression or exclusion of the pectoral muscle from the mammograms is demanded for computer-aided analysis which requires the identification of the pectoral muscle. The main objective of this study is to propose an automated method to efficiently identify the pectoral muscle in medio-lateral oblique-view mammograms. This method uses a proposed graph cut-based image segmentation technique for identifying the pectoral muscle edge. The identified pectoral muscle edge is found to be ragged. Hence, the pectoral muscle is smoothly represented using Bezier curve which uses the control points obtained from the pectoral muscle edge. The proposed work was tested on a public dataset of medio-lateral oblique-view mammograms obtained from mammographic image analysis society database, and its performance was compared with the state-of-the-art methods reported in the literature. The mean false positive and false negative rates of the proposed method over randomly chosen 84 mammograms were calculated, respectively, as 0.64% and 5.58%. Also, with respect to the number of results with small error, the proposed method out performs existing methods. These results indicate that the proposed method can be used to accurately identify the pectoral muscle on medio-lateral oblique view mammograms.