Diabetic retinopathy is a major cause of vision loss in diabetic patients. Currently, there is a need for making decisions using intelligent computer algorithms when screening a large volume of data. This paper presents an expert decision-making system designed using a fuzzy support vector machine (FSVM) classifier to detect hard exudates in fundus images. The optic discs in the colour fundus images are segmented to avoid false alarms using morphological operations and based on circular Hough transform. To discriminate between the exudates and the non-exudates pixels, colour and texture features are extracted from the images. These features are given as input to the FSVM classifier. The classifier analysed 200 retinal images collected from diabetic retinopathy screening programmes. The tests made on the retinal images show that the proposed detection system has better discriminating power than the conventional support vector machine. With the best combination of FSVM and features sets, the area under the receiver operating characteristic curve reached 0.9606, which corresponds to a sensitivity of 94.1 % with a specificity of 90.0 %. The results suggest that detecting hard exudates using FSVM contribute to computer-assisted detection of diabetic retinopathy and as a decision support system for ophthalmologists.
Diabetic retinopathy; Colour fundus images; Hard exudates; Laws texture energy measures; Fuzzy support vector machine
An automatic method for cartilage segmentation using knee MRI images is described. Three binary classifiers with integral and partial pixel features are built using the Bayesian theorem to segment the femoral cartilage, tibial cartilage and patellar cartilage separately. First, an iterative procedure based on the feedback of the number of strong edges is designed to obtain an appropriate threshold for the Canny operator and to extract the bone-cartilage interface from MRI images. Second, the different edges are identified based on certain features, which allow for different cartilage to be distinguished synchronously. The cartilage is segmented preliminarily with minimum error Bayesian classifiers that have been previously trained. According to the cartilage edge and its anatomic location, the speed of segmentation is improved. Finally, morphological operations are used to improve the primary segmentation results. The cartilage edge is smooth in the automatic segmentation results and shows good consistency with manual segmentation results. The mean Dice similarity coefficient is 0.761.
Knee; Articular cartilage; Segmentation; MRI; Pattern recognition
Digital cardiovascular angiography accounts for a major portion of the radiation dose among the examinations performed at cardiovascular centres. However, dose-related information is neither monitored nor recorded systemically. This report concerns the construction of a radiation dose monitoring system based on digital imaging and communications in medicine (DICOM) data and its use at the cardiovascular centre of the University Hospitals in Korea. The dose information was analysed according to DICOM standards for a series of procedures, and the formulation of diagnostic reference levels (DRLs) at our cardiovascular centre represents the first of its kind in Korea. We determined a dose area product (DAP) DRL for coronary angiography of 75.6 Gy cm2 and a fluoroscopic time DRL of 318.0 s. The DAP DRL for percutaneous transluminal coronary intervention was 213.3 Gy cm2, and the DRL for fluoroscopic time was 1207.5 s.
DICOM MPPS; Radiation dose monitoring; Cardiovascular centre; Diagnostic reference levels; Order communication system
The production of medical imaging is a continuing trend in healthcare institutions. Quality assurance for planned radiation exposure situations (e.g. X-ray, computer tomography) requires examination-specific set-ups according to several parameters, such as patient’s age and weight, body region and clinical indication. These data are normally stored in several formats and with different nomenclatures, which hinder the continuous and automatic monitoring of these indicators and the comparison between several institutions and equipment. This article proposes a framework that aggregates, normalizes and provides different views over collected indicators. The developed tool can be used to improve the quality of radiologic procedures and also for benchmarking and auditing purposes. Finally, a case study and several experimental results related to radiation exposure and productivity are presented and discussed.
Medical imaging; PACS; ALARA; DICOM; Monitoring; Quality of service
We evaluated the image registration accuracy achieved using two deformable registration algorithms when radiation-induced normal tissue changes were present between serial computed tomography (CT) scans. Two thoracic CT scans were collected for each of 24 patients who underwent radiation therapy (RT) treatment for lung cancer, eight of whom experienced radiologically evident normal tissue damage between pre- and post-RT scan acquisition. For each patient, 100 landmark point pairs were manually placed in anatomically corresponding locations between each pre- and post-RT scan. Each post-RT scan was then registered to the pre-RT scan using (1) the Plastimatch demons algorithm and (2) the Fraunhofer MEVIS algorithm. The registration accuracy for each scan pair was evaluated by comparing the distance between landmark points that were manually placed in the post-RT scans and points that were automatically mapped from pre- to post-RT scans using the displacement vector fields output by the two registration algorithms. For both algorithms, the registration accuracy was significantly decreased when normal tissue damage was present in the post-RT scan. Using the Plastimatch algorithm, registration accuracy was 2.4 mm, on average, in the absence of radiation-induced damage and 4.6 mm, on average, in the presence of damage. When the Fraunhofer MEVIS algorithm was instead used, registration errors decreased to 1.3 mm, on average, in the absence of damage and 2.5 mm, on average, when damage was present. This work demonstrated that the presence of lung tissue changes introduced following RT treatment for lung cancer can significantly decrease the registration accuracy achieved using deformable registration.
Chest CT; Image registration; Lung; Radiotherapy
Image compression techniques aim at reducing the amount of data needed to accurately represent an image, such that the image can be economically transmitted or archived. This paper deals with employing symmetry as a parameter for compression of biomedical images. The approach presented in this paper offers great potential in complete lossless compression of the biomedical image under consideration, with the reconstructed image being mathematically identical to the original image. The method comprises getting rid of the redundant data and encoding the non-redundant data for the purpose of regenerating the image at the receiver section without any observable change in the image data.
Symmetry; Biomedical image compression; Redundancy; Diagnostic ability
The use of mobile devices for medical image capture has become increasingly popular given the widespread use of smartphone cameras. Prior studies have generally compared mobile phone capture images to digitized images. However, many underserved and rural areas without picture archiving and communication systems (PACS) still depend greatly on the use of film radiographs. Additionally, there is a scarcity of specialty-trained or formally licensed radiologists in many of these regions. Subsequently, there is great potential for the use of smartphone capture of plain radiograph films which would allow for increased access to economical and efficient consultation from board-certified radiologists abroad. The present study addresses the ability to diagnose a subset of radiographic findings identified on both the original film radiograph and the captured camera phone image.
Teleradiology; Diagnostic image quality; Image quality analysis; Digital image processing; ROC-based analysis
The paper is focused on a tiSsue-Based Standardization Technique (SBST) of magnetic resonance (MR) brain images. Magnetic Resonance Imaging intensities have no fixed tissue-specific numeric meaning, even within the same MRI protocol, for the same body region, or even for images of the same patient obtained on the same scanner in different moments. This affects postprocessing tasks such as automatic segmentation or unsupervised/supervised classification methods, which strictly depend on the observed image intensities, compromising the accuracy and efficiency of many image analyses algorithms. A large number of MR images from public databases, belonging to healthy people and to patients with different degrees of neurodegenerative pathology, were employed together with synthetic MRIs. Combining both histogram and tissue-specific intensity information, a correspondence is obtained for each tissue across images. The novelty consists of computing three standardizing transformations for the three main brain tissues, for each tissue class separately. In order to create a continuous intensity mapping, spline smoothing of the overall slightly discontinuous piecewise-linear intensity transformation is performed. The robustness of the technique is assessed in a post hoc manner, by verifying that automatic segmentation of images before and after standardization gives a high overlapping (Dice index >0.9) for each tissue class, even across images coming from different sources. Furthermore, SBST efficacy is tested by evaluating if and how much it increases intertissue discrimination and by assessing gaussianity of tissue gray-level distributions before and after standardization. Some quantitative comparisons to already existing different approaches available in the literature are performed.
General intensity scale; Magnetic Resonance Imaging; Nonlinear registration; Intensity standardization; Alzheimer’s Disease Neuroimaging Initiative
We analyze the importance of shape features for predicting spiculation ratings assigned by radiologists to lung nodules in computed tomography (CT) scans. Using the Lung Image Database Consortium (LIDC) data and classification models based on decision trees, we demonstrate that the importance of several shape features increases disproportionately relative to other image features with increasing size of the nodule. Our shaped-based classification results show an area under the receiver operating characteristic (ROC) curve of 0.65 when classifying spiculation for small nodules and an area of 0.91 for large nodules, resulting in a 26 % difference in classification performance using shape features. An analysis of the results illustrates that this change in performance is driven by features that measure boundary complexity, which perform well for large nodules but perform relatively poorly and do no better than other features for small nodules. For large nodules, the roughness of the segmented boundary maps well to the semantic concept of spiculation. For small nodules, measuring directly the complexity of hard segmentations does not yield good results for predicting spiculation due to limits imposed by spatial resolution and the uncertainty in boundary location. Therefore, a wider range of features, including shape, texture, and intensity features, are needed to predict spiculation ratings for small nodules. A further implication is that the efficacy of shape features for a particular classifier used to create computer-aided diagnosis systems depends on the distribution of nodule sizes in the training and testing sets, which may not be consistent across different research studies.
Biomedical image analysis; Computed tomography; Computer-aided diagnosis (CAD); Computer analysis; Decision support; Decision trees; Diagnostic imaging; Image interpretation; LIDC
Finding optimal compression levels for diagnostic imaging is not an easy task. Significant compressibility variations exist between modalities, but little is known about compressibility variations within modalities. Moreover, compressibility is affected by acquisition parameters. In this study, we evaluate the compressibility of thousands of computed tomography (CT) slices acquired with different slice thicknesses, exposures, reconstruction filters, slice collimations, and pitches. We demonstrate that exposure, slice thickness, and reconstruction filters have a significant impact on image compressibility due to an increased high frequency content and a lower acquisition signal-to-noise ratio. We also show that compression ratio is not a good fidelity measure. Therefore, guidelines based on compression ratio should ideally be replaced with other compression measures better correlated with image fidelity. Value-of-interest (VOI) transformations also affect the perception of quality. We have studied the effect of value-of-interest transformation and found significant masking of artifacts when window is widened.
Image compression; Image artifact; Image quality; Image visualization; JPEG2000; Computed tomography; Exposure; Slice thickness; Filter type
This study was conducted to determine whether facial photographs obtained simultaneously with radiographs improve radiologists’ detection rate of wrong-patient errors, when they are explicitly asked to include the photographs in their evaluation. Radiograph-photograph combinations were obtained from 28 patients at the time of portable chest radiography imaging. From these, pairs of radiographs were generated. Each unique pair consisted of one new and one old (comparison) radiograph. Twelve pairs of mismatched radiographs (i.e., pairs containing radiographs of different patients) were also generated. In phase 1 of the study, 5 blinded radiologist observers were asked to interpret 20 pairs of radiographs without the photographs. In phase 2, each radiologist interpreted another 20 pairs of radiographs with the photographs. Radiologist observers were not instructed about the purpose of the photographs but were asked to include the photographs in their review. The detection rate of mismatched errors was recorded along with the interpretation time for each session for each observer. The two-tailed Fisher exact test was used to evaluate differences in mismatch detection rates between the two phases. A p value of <0.05 was considered significant. The error detection rates without (0/20 = 0 %) and with (17/18 = 94.4 %) photographs were different (p = 0.0001). The average interpretation times for the set of 20 radiographs were 26.45 (SD 8.69) and 20.55 (SD 3.40) min, for phase 1 and phase 2, respectively (two-tailed Student t test, p = 0.1911). When radiologists include simultaneously obtained photographs in their review of portable chest radiographs, there is a significant improvement in the detection of labeling errors. No statistically significant difference in interpretation time was observed. This may lead to improved patient safety without affecting radiologists’ throughput.
Medical errors; Wrong-patient events
Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is a well-established technique for studying blood–brain barrier (BBB) permeability that allows measurements to be made for a wide range of brain pathologies, including multiple sclerosis and brain tumors (BT). This latter application is particularly interesting, because high-grade gliomas are characterized by increased microvascular permeability and a loss of BBB function due to the structural abnormalities of the endothelial layer. In this study, we compared the extended Tofts-Kety (ETK) model and an extended derivate class from phenomenological universalities called EU1 in 30 adult patients with different BT grades. A total of 75 regions of interest were manually drawn on the MRI and subsequently analyzed using the ETK and EU1 algorithms. Significant linear correlations were found among the parameters obtained by these two algorithms. The means of R2 obtained using ETK and EU1 models for high-grade tumors were 0.81 and 0.91, while those for low-grade tumors were 0.82 and 0.85, respectively; therefore, these two models are equivalent. In conclusion, we can confirm that the application of the EU1 model to the DCE-MRI experimental data might be a useful alternative to pharmacokinetic models in the study of BT, because the analytic results can be generated more quickly and easily than with the ETK model.
DCE-MRI; Brain tumors; Blood–brain barrier; Extended Tofts-Kety model; EU1 algorithm
This paper proposes a new method for simple, efficient, and robust removal of the non-brain tissues in MR images based on an irrational mask for filtration within a binary morphological operation framework. The proposed skull-stripping segmentation is based on two irrational 3 × 3 and 5 × 5 masks, having the sum of its weights equal to the transcendental number π value provided by the Gregory-Leibniz infinite series. It allows maintaining a lower rate of useful pixel loss. The proposed method has been tested in two ways. First, it has been validated as a binary method by comparing and contrasting with Otsu’s, Sauvola’s, Niblack’s, and Bernsen’s binary methods. Secondly, its accuracy has been verified against three state-of-the-art skull-stripping methods: the graph cuts method, the method based on Chan-Vese active contour model, and the simplex mesh and histogram analysis skull stripping. The performance of the proposed method has been assessed using the Dice scores, overlap and extra fractions, and sensitivity and specificity as statistical methods. The gold standard has been provided by two neurologist experts. The proposed method has been tested and validated on 26 image series which contain 216 images from two publicly available databases: the Whole Brain Atlas and the Internet Brain Segmentation Repository that include a highly variable sample population (with reference to age, sex, healthy/diseased). The approach performs accurately on both standardized databases. The main advantage of the proposed method is its robustness and speed.
Skull stripping; Irrational mask; Binarization; Similarity metrics; Magnetic resonance image
Additive manufacturing and bio-printing, with the potential for direct fabrication of complex patient-specific anatomies derived from medical scan data, are having an ever-increasing impact on the practice of medicine. Anatomic structures are typically derived from CT or MRI scans, and there are multiple steps in the model derivation process that influence the geometric accuracy of the printed constructs. In this work, we compare the dimensional accuracy of 3-D printed constructs of an L1 vertebra derived from CT data for an ex vivo cadaver T-L spine with the original vertebra. Processing of segmented structures using binary median filters and various surface extraction algorithms is evaluated for the effect on model dimensions. We investigate the effects of changing CT reconstruction kernels by scanning simple geometric objects and measuring the impact on the derived model dimensions. We also investigate if there are significant differences between physical and virtual model measurements. The 3-D models were printed using a commercial 3-D printer, the Replicator 2 (MakerBot, Brooklyn, NY) using polylactic acid (PLA) filament. We found that changing parameters during the scan reconstruction, segmentation, filtering, and surface extraction steps will have an effect on the dimensions of the final model. These effects need to be quantified for specific situations that rely on the accuracy of 3-D printed models used in medicine or tissue engineering applications.
Three-dimensional imaging (3-D imaging); 3-D reconstruction; 3-D segmentation; Computed tomography; Additive manufacturing; Orthopedic modeling; Dimensional accuracy
Breast density is a strong risk factor for breast cancer. In this paper, we present an automated approach for breast density segmentation in mammographic images based on a supervised pixel-based classification and using textural and morphological features. The objective of the paper is not only to show the feasibility of an automatic algorithm for breast density segmentation but also to prove its potential application to the study of breast density evolution in longitudinal studies. The database used here contains three complete screening examinations, acquired 2 years apart, of 130 different patients. The approach was validated by comparing manual expert annotations with automatically obtained estimations. Transversal analysis of the breast density analysis of craniocaudal (CC) and mediolateral oblique (MLO) views of both breasts acquired in the same study showed a correlation coefficient of ρ = 0.96 between the mammographic density percentage for left and right breasts, whereas a comparison of both mammographic views showed a correlation of ρ = 0.95. A longitudinal study of breast density confirmed the trend that dense tissue percentage decreases over time, although we noticed that the decrease in the ratio depends on the initial amount of breast density.
Breast tissue density; Segmentation; Mammography; Longitudinal studies; Computer-assisted image interpretation
Breast cancer screening is central to early breast cancer detection. Identifying and monitoring process measures for screening is a focus of the National Cancer Institute’s Population-based Research Optimizing Screening through Personalized Regimens (PROSPR) initiative, which requires participating centers to report structured data across the cancer screening continuum. We evaluate the accuracy of automated information extraction of imaging findings from radiology reports, which are available as unstructured text. We present prevalence estimates of imaging findings for breast imaging received by women who obtained care in a primary care network participating in PROSPR (n = 139,953 radiology reports) and compared automatically extracted data elements to a “gold standard” based on manual review for a validation sample of 941 randomly selected radiology reports, including mammograms, digital breast tomosynthesis, ultrasound, and magnetic resonance imaging (MRI). The prevalence of imaging findings vary by data element and modality (e.g., suspicious calcification noted in 2.6 % of screening mammograms, 12.1 % of diagnostic mammograms, and 9.4 % of tomosynthesis exams). In the validation sample, the accuracy of identifying imaging findings, including suspicious calcifications, masses, and architectural distortion (on mammogram and tomosynthesis); masses, cysts, non-mass enhancement, and enhancing foci (on MRI); and masses and cysts (on ultrasound), range from 0.8 to1.0 for recall, precision, and F-measure. Information extraction tools can be used for accurate documentation of imaging findings as structured data elements from text reports for a variety of breast imaging modalities. These data can be used to populate screening registries to help elucidate more effective breast cancer screening processes.
BI-RADS; Breast; Data extraction; Information storage and retrieval; Natural language processing
Medical image sharing is an important problem in modern radiology, with wide applications in Internet and mobile devices. Some important features need to be added and optimized to medical image sharing. In this paper, we present an extensible Web Access to DICOM Persistent Objects (WADO) middleware based on image cache and real-time Web monitor technology for regional medical image sharing. We first develop the extension method of WADO standard and workflow of extended WADO service. Then, we design a medical image cache method to improve the performance of medical image on-demand transmission. Using the real-time monitor can discover the performance bottlenecks and optimized critical points. The experimental results show that the middleware effectively delivers medical images and reports to Web clients over the Internet, regardless of the platform used for access. It can be deployed in one hospital to provide WADO service to medical workers and also can be applied to regional picture archiving and communication systems (PACS) to transmit medical images and reports to Internet users in a way that is transparent to end-user applications.
DICOM; WADO; Image cache; Web monitor; Regional medical image sharing
Log files of information retrieval systems that record user behavior have been used to improve the outcomes of retrieval systems, understand user behavior, and predict events. In this article, a log file of the ARRS GoldMiner search engine containing 222,005 consecutive queries is analyzed. Time stamps are available for each query, as well as masked IP addresses, which enables to identify queries from the same person. This article describes the ways in which physicians (or Internet searchers interested in medical images) search and proposes potential improvements by suggesting query modifications. For example, many queries contain only few terms and therefore are not specific; others contain spelling mistakes or non-medical terms that likely lead to poor or empty results. One of the goals of this report is to predict the number of results a query will have since such a model allows search engines to automatically propose query modifications in order to avoid result lists that are empty or too large. This prediction is made based on characteristics of the query terms themselves. Prediction of empty results has an accuracy above 88 %, and thus can be used to automatically modify the query to avoid empty result sets for a user. The semantic analysis and data of reformulations done by users in the past can aid the development of better search systems, particularly to improve results for novice users. Therefore, this paper gives important ideas to better understand how people search and how to use this knowledge to improve the performance of specialized medical search engines.
Image retrieval; Human-computer interaction; Machine learning; Statistic analysis; Information storage and retrieval; Medical image search; Log file analysis
Breast cancer is becoming a leading death of women all over the world; clinical experiments demonstrate that early detection and accurate diagnosis can increase the potential of treatment. In order to improve the breast cancer diagnosis precision, this paper presents a novel automated segmentation and classification method for mammograms. We conduct the experiment on both DDSM database and MIAS database, firstly extract the region of interests (ROIs) with chain codes and using the rough set (RS) method to enhance the ROIs, secondly segment the mass region from the location ROIs with an improved vector field convolution (VFC) snake and following extract features from the mass region and its surroundings, and then establish features database with 32 dimensions; finally, these features are used as input to several classification techniques. In our work, the random forest is used and compared with support vector machine (SVM), genetic algorithm support vector machine (GA-SVM), particle swarm optimization support vector machine (PSO-SVM), and decision tree. The effectiveness of our method is evaluated by a comprehensive and objective evaluation system; also, Matthew’s correlation coefficient (MCC) indicator is used. Among the state-of-the-art classifiers, our method achieves the best performance with best accuracy of 97.73 %, and the MCC value reaches 0.8668 and 0.8652 in unique DDSM database and both two databases, respectively. Experimental results prove that the proposed method outperforms the other methods; it could consider applying in CAD systems to assist the physicians for breast cancer diagnosis.
Computer-aided detection; Mammography; Automated mass segmentation; Classification; Random forest
A rapid and highly accurate diagnostic tool for distinguishing benign tumors from malignant ones is required owing to the high incidence of breast cancer. Although various computer-aided diagnosis (CAD) systems have been developed to interpret ultrasound images of breast tumors, feature selection and the setting of parameters are still essential to classification accuracy and the minimization of computational complexity. This work develops a highly accurate CAD system that is based on a support vector machine (SVM) and the artificial immune system (AIS) algorithm for evaluating breast tumors. Experiments demonstrate that the accuracy of the proposed CAD system for classifying breast tumors is 96.67 %. The sensitivity, specificity, PPV, and NPV of the proposed CAD system are 96.67, 96.67, 95.60, and 97.48 %, respectively. The receiver operator characteristic (ROC) area index Az is 0.9827. Hence, the proposed CAD system can reduce the number of biopsies and yield useful results that assist physicians in diagnosing breast tumors.
Breast tumors; Textural feature; Morphological feature; Artificial immune system algorithm; Support vector machine
In the UK, physicists and radiographers perform routine quality control (QC) of digital mammography equipment at daily, weekly and monthly intervals. The tests performed and tolerances are specified by standard protocols. The manual nature of many of the tests introduces variability due to the positioning of regions of interest (ROIs) and can be time consuming. The tools on workstations provided by manufacturers limit the range of analysis that radiographers can perform and do not allow for a standard set of tools and analysis because they are specific to a given manufacturer. Automated software provides a means of reducing the variability in the analysis and also provides the possibility of additional, more complex analysis than is currently performed on the daily, weekly and monthly checks by radiographers. To this end, a set of tools has been developed to analyse the routine images taken by radiographers. As well as automatically reproducing the usual measurements by radiographers more complex analysis is provided. A QC image collection system has been developed which automatically routes QC data from a clinical site to a centralised server for analysis. A Web-based interface has been created that allows the users to view the performance of the mammographic equipment. The pilot system obtained over 3000 QC images from seven X-ray units at a single screening centre over 2 years. The results show that these tools and methods of analysis can highlight changes in a detector over time that may otherwise go unnoticed with the conventional analysis.
QC; Mammography; NHSBSP; ImageJ; QA
Patient-specific 3D models obtained by the segmentation of volumetric diagnostic images play an increasingly important role in surgical planning. Surgeons use the virtual models reconstructed through segmentation to plan challenging surgeries. Many solutions exist for the different anatomical districts and surgical interventions. The possibility to bring the 3D virtual reconstructions with native radiological images in the operating room is essential for fostering the use of intraoperative planning. To the best of our knowledge, current DICOM viewers are not able to simultaneously connect to the picture archiving and communication system (PACS) and import 3D models generated by external platforms to allow a straight integration in the operating room. A total of 26 DICOM viewers were evaluated: 22 open source and four commercial. Two DICOM viewers can connect to PACS and import segmentations achieved by other applications: Synapse 3D® by Fujifilm and OsiriX by University of Geneva. We developed a software network that converts diffuse visual tool kit (VTK) format 3D model segmentations, obtained by any software platform, to a DICOM format that can be displayed using OsiriX or Synapse 3D. Both OsiriX and Synapse 3D were suitable for our purposes and had comparable performance. Although Synapse 3D loads native images and segmentations faster, the main benefits of OsiriX are its user-friendly loading of elaborated images and it being both free of charge and open source.
Digital Imaging and Communications in Medicine (DICOM); PACS; 3D imaging (imaging three-dimensional); 3D segmentation