Effective Metadata Discovery for Dynamic Filtering of Queries to a Radiology Image Search Engine

We sought to demonstrate the effectiveness of techniques to index radiology images using metadata discovered in their free-text figure captions. The ARRS GoldMiner™ image library incorporated 94,256 figures from 11,712 articles published in peer-reviewed online radiology journals. Algorithms were developed to discover metadata—age, sex, and imaging modality—from the figures’ free-text captions. Age was recorded in years, and was classified as infant (less than 2 years), child (2 to 17 years), or adult (18+ years). Each figure was assigned to one of eight imaging modalities. A random sample of 1,000 images was examined to measure accuracy of the metadata. The patient’s age was identified in 58,994 cases (63%), and the patient’s sex was identified in 58,427 cases (62%). An imaging modality was assigned to 80,402 (85%) of the figures. Based on the 1,000 sampled cases, recall values for age, sex, and imaging modality were 97.2%, 99.7%, and 86.4%, respectively. Precision values for age, sex, and imaging modality were 100%, 100%, and 97.2%, respectively. Automated techniques can accurately discover age, sex, and imaging modality metadata from captions of figures published in radiology journals. The metadata can be used to dynamically filter queries for an image search engine.     

Integration of Imaging Signs into RadLex

Imaging signs form an important part of the language of radiology, but are not represented in established lexicons. We sought to incorporate imaging signs into RSNA''s RadLex® ontology of radiology terms. Names of imaging signs and their definitions were culled from books, journal articles, dictionaries, and biomedical web sites. Imaging signs were added into RadLex as subclasses of the term “imaging sign,” which was defined in RadLex as a subclass of “imaging observation.” A total of 743 unique imaging signs were added to RadLex with their 392 synonyms to yield a total of 1,135 new terms. All included definitions and related RadLex terms, including imaging modality, anatomy, and disorder, when appropriate. The information will allow RadLex users to identify imaging signs by modality (e.g., ultrasound signs) and to find all signs related to specific pathophysiology. The addition of imaging signs to RadLex augments its use to index the radiology literature, create and interpret clinical radiology reports, and retrieve relevant cases and images.     

Using Keynote to Present Radiology Images

Numerous articles have offered instructions for working with advanced radiology images in Microsoft PowerPoint (Redmond, WA); however, no articles have detailed instructions to do the same on alternative presentation software. Apple Macintosh (Cupertino, CA) computers are gaining popularity with many radiologists, due in part to the availability of a powerful, free, open-source Digital Imaging and Communications in Medicine (DICOM) viewing and manipulating software OsiriX (). Apple’s own presentation software, Keynote, is particularly effective in dealing with medical images and cine clips. This article demonstrates how to use Apple’s Keynote software to present radiology images and scrollable image stacks, without third-party add-on software. The article also illustrates how to compress media files and protect patient information in Keynote presentations. Lastly, it addresses the steps to converting between PowerPoint and Keynote file formats. Apple’s Keynote software enables quick and efficient addition of multiple static images or scrollable image stacks, compression of media files, and removal of patient information. These functions can be accomplished by inexperienced users with no software modifications.     

Decision tool for the early diagnosis of trauma patient hypovolemia

We present a classifier for use as a decision assist tool to identify a hypovolemic state in trauma patients during helicopter transport to a hospital, when reliable acquisition of vital-sign data may be difficult. The decision tool uses basic vital-sign variables as input into linear classifiers, which are then combined into an ensemble classifier. The classifier identifies hypovolemic patients with an area under a receiver operating characteristic curve (AUC) of 0.76 (standard deviation 0.05, for 100 randomly-reselected patient subsets). The ensemble classifier is robust; classification performance degrades only slowly as variables are dropped, and the ensemble structure does not require identification of a set of variables for use as best-feature inputs into the classifier. The ensemble classifier consistently outperforms best-features-based linear classifiers (the classification AUC is greater, and the standard deviation is smaller, p < 0.05). The simple computational requirements of ensemble classifiers will permit them to function in small fieldable devices for continuous monitoring of trauma patients.     

Automatic figure classification in bioscience literature

Millions of figures appear in biomedical articles, and it is important to develop an intelligent figure search engine to return relevant figures based on user entries. In this study we report a figure classifier that automatically classifies biomedical figures into five predefined figure types: Gel-image, Image-of-thing, Graph, Model, and Mix. The classifier explored rich image features and integrated them with text features. We performed feature selection and explored different classification models, including a rule-based figure classifier, a supervised machine-learning classifier, and a multi-model classifier, the latter of which integrated the first two classifiers. Our results show that feature selection improved figure classification and the novel image features we explored were the best among image features that we have examined. Our results also show that integrating text and image features achieved better performance than using either of them individually. The best system is a multi-model classifier which combines the rule-based hierarchical classifier and a support vector machine (SVM) based classifier, achieving a 76.7% F1-score for five-type classification. We demonstrated our system at http://figureclassification.askhermes.org/.     

An Ensemble Method for Classifying Regional Disease Patterns of Diffuse Interstitial Lung Disease Using HRCT Images from Different Vendors

We propose the use of ensemble classifiers to overcome inter-scanner variations in the differentiation of regional disease patterns in high-resolution computed tomography (HRCT) images of diffuse interstitial lung disease patients obtained from different scanners. A total of 600 rectangular 20 × 20-pixel regions of interest (ROIs) on HRCT images obtained from two different scanners (GE and Siemens) and the whole lung area of 92 HRCT images were classified as one of six regional pulmonary disease patterns by two expert radiologists. Textual and shape features were extracted from each ROI and the whole lung parenchyma. For automatic classification, individual and ensemble classifiers were trained and tested with the ROI dataset. We designed the following three experimental sets: an intra-scanner study in which the training and test sets were from the same scanner, an integrated scanner study in which the data from the two scanners were merged, and an inter-scanner study in which the training and test sets were acquired from different scanners. In the ROI-based classification, the ensemble classifiers showed better (p < 0.001) accuracy (89.73%, SD = 0.43) than the individual classifiers (88.38%, SD = 0.31) in the integrated scanner test. The ensemble classifiers also showed partial improvements in the intra- and inter-scanner tests. In the whole lung classification experiment, the quantification accuracies of the ensemble classifiers with integrated training (49.57%) were higher (p < 0.001) than the individual classifiers (48.19%). Furthermore, the ensemble classifiers also showed better performance in both the intra- and inter-scanner experiments. We concluded that the ensemble classifiers provide better performance when using integrated scanner images.     

Automatic medical image annotation and keyword-based image retrieval using relevance feedback

This paper presents novel multiple keywords annotation for medical images, keyword-based medical image retrieval, and relevance feedback method for image retrieval for enhancing image retrieval performance. For semantic keyword annotation, this study proposes a novel medical image classification method combining local wavelet-based center symmetric-local binary patterns with random forests. For keyword-based image retrieval, our retrieval system use the confidence score that is assigned to each annotated keyword by combining probabilities of random forests with predefined body relation graph. To overcome the limitation of keyword-based image retrieval, we combine our image retrieval system with relevance feedback mechanism based on visual feature and pattern classifier. Compared with other annotation and relevance feedback algorithms, the proposed method shows both improved annotation performance and accurate retrieval results.     

Comparative Performance Analysis of State-of-the-Art Classification Algorithms Applied to Lung Tissue Categorization

In this paper, we compare five common classifier families in their ability to categorize six lung tissue patterns in high-resolution computed tomography (HRCT) images of patients affected with interstitial lung diseases (ILD) and with healthy tissue. The evaluated classifiers are naive Bayes, k-nearest neighbor, J48 decision trees, multilayer perceptron, and support vector machines (SVM). The dataset used contains 843 regions of interest (ROI) of healthy and five pathologic lung tissue patterns identified by two radiologists at the University Hospitals of Geneva. Correlation of the feature space composed of 39 texture attributes is studied. A grid search for optimal parameters is carried out for each classifier family. Two complementary metrics are used to characterize the performances of classification. These are based on McNemar’s statistical tests and global accuracy. SVM reached best values for each metric and allowed a mean correct prediction rate of 88.3% with high class-specific precision on testing sets of 423 ROIs.     

Ensemble selection for feature-based classification of diabetic maculopathy images

As diabetic maculopathy (DM) is a prevalent cause of blindness in the world, it is increasingly important to use automated techniques for the early detection of the disease. In this paper, we propose a decision system to classify DM fundus images into normal, clinically significant macular edema (CMSE), and non-clinically significant macular edema (non-CMSE) classes. The objective of the proposed decision system is three fold namely, to automatically extract textural features (both region specific and global), to effectively choose subset of discriminatory features, and to classify DM fundus images to their corresponding class of disease severity. The system uses a gamut of textural features and an ensemble classifier derived from four popular classifiers such as the hidden naïve Bayes, naïve Bayes, sequential minimal optimization (SMO), and the tree-based J48 classifiers. We achieved an average classification accuracy of 96.7% using five-fold cross validation.     

Dynamic “Inline” Images: Context-Sensitive Retrieval and Integration of Images into Web Documents

Integrating relevant images into web-based information resources adds value for research and education. This work sought to evaluate the feasibility of using “Web 2.0” technologies to dynamically retrieve and integrate pertinent images into a radiology web site. An online radiology reference of 1,178 textual web documents was selected as the set of target documents. The ARRS GoldMiner™ image search engine, which incorporated 176,386 images from 228 peer-reviewed journals, retrieved images on demand and integrated them into the documents. At least one image was retrieved in real-time for display as an “inline” image gallery for 87% of the web documents. Each thumbnail image was linked to the full-size image at its original web site. Review of 20 randomly selected Collaborative Hypertext of Radiology documents found that 69 of 72 displayed images (96%) were relevant to the target document. Users could click on the “More” link to search the image collection more comprehensively and, from there, link to the full text of the article. A gallery of relevant radiology images can be inserted easily into web pages on any web server. Indexing by concepts and keywords allows context-aware image retrieval, and searching by document title and subject metadata yields excellent results. These techniques allow web developers to incorporate easily a context-sensitive image gallery into their documents.     

IntelliHealth: A medical decision support application using a novel weighted multi-layer classifier ensemble framework

Accuracy plays a vital role in the medical field as it concerns with the life of an individual. Extensive research has been conducted on disease classification and prediction using machine learning techniques. However, there is no agreement on which classifier produces the best results. A specific classifier may be better than others for a specific dataset, but another classifier could perform better for some other dataset. Ensemble of classifiers has been proved to be an effective way to improve classification accuracy. In this research we present an ensemble framework with multi-layer classification using enhanced bagging and optimized weighting. The proposed model called “HM-BagMoov” overcomes the limitations of conventional performance bottlenecks by utilizing an ensemble of seven heterogeneous classifiers. The framework is evaluated on five different heart disease datasets, four breast cancer datasets, two diabetes datasets, two liver disease datasets and one hepatitis dataset obtained from public repositories. The analysis of the results show that ensemble framework achieved the highest accuracy, sensitivity and F-Measure when compared with individual classifiers for all the diseases. In addition to this, the ensemble framework also achieved the highest accuracy when compared with the state of the art techniques. An application named “IntelliHealth” is also developed based on proposed model that may be used by hospitals/doctors for diagnostic advice.     

Detection of prostate cancer by integration of line-scan diffusion, T2-mapping and T2-weighted magnetic resonance imaging; a multichannel statistical classifier

A multichannel statistical classifier for detecting prostate cancer was developed and validated by combining information from three different magnetic resonance (MR) methodologies: T2-weighted, T2-mapping, and line scan diffusion imaging (LSDI). From these MR sequences, four different sets of image intensities were obtained: T2-weighted (T2W) from T2-weighted imaging, Apparent Diffusion Coefficient (ADC) from LSDI, and proton density (PD) and T2 (T2 Map) from T2-mapping imaging. Manually segmented tumor labels from a radiologist, which were validated by biopsy results, served as tumor "ground truth." Textural features were extracted from the images using co-occurrence matrix (CM) and discrete cosine transform (DCT). Anatomical location of voxels was described by a cylindrical coordinate system. A statistical jack-knife approach was used to evaluate our classifiers. Single-channel maximum likelihood (ML) classifiers were based on 1 of the 4 basic image intensities. Our multichannel classifiers: support vector machine (SVM) and Fisher linear discriminant (FLD), utilized five different sets of derived features. Each classifier generated a summary statistical map that indicated tumor likelihood in the peripheral zone (PZ) of the prostate gland. To assess classifier accuracy, the average areas under the receiver operator characteristic (ROC) curves over all subjects were compared. Our best FLD classifier achieved an average ROC area of 0.839(+/-0.064), and our best SVM classifier achieved an average ROC area of 0.761(+/-0.043). The T2W ML classifier, our best single-channel classifier, only achieved an average ROC area of 0.599(+/-0.146). Compared to the best single-channel ML classifier, our best multichannel FLD and SVM classifiers have statistically superior ROC performance (P=0.0003 and 0.0017, respectively) from pairwise two-sided t-test. By integrating the information from multiple images and capturing the textural and anatomical features in tumor areas, summary statistical maps can potentially aid in image-guided prostate biopsy and assist in guiding and controlling delivery of localized therapy under image guidance.     

A Presentation System for Just-in-time Learning in Radiology

There is growing interest in bringing medical educational materials to the point of care. We sought to develop a system for just-in-time learning in radiology. A database of 34 learning modules was derived from previously published journal articles. Learning objectives were specified for each module, and multiple-choice test items were created. A web-based system—called TEMPO—was developed to allow radiologists to select and view the learning modules. Web services were used to exchange clinical context information between TEMPO and the simulated radiology work station. Preliminary evaluation was conducted using the System Usability Scale (SUS) questionnaire. TEMPO identified learning modules that were relevant to the age, sex, imaging modality, and body part or organ system of the patient being viewed by the radiologist on the simulated clinical work station. Users expressed a high degree of satisfaction with the system’s design and user interface. TEMPO enables just-in-time learning in radiology, and can be extended to create a fully functional learning management system for point-of-care learning in radiology.     

An Interactive System for Computer-Aided Diagnosis of Breast Masses

Although mammography is the only clinically accepted imaging modality for screening the general population to detect breast cancer, interpreting mammograms is difficult with lower sensitivity and specificity. To provide radiologists “a visual aid” in interpreting mammograms, we developed and tested an interactive system for computer-aided detection and diagnosis (CAD) of mass-like cancers. Using this system, an observer can view CAD-cued mass regions depicted on one image and then query any suspicious regions (either cued or not cued by CAD). CAD scheme automatically segments the suspicious region or accepts manually defined region and computes a set of image features. Using content-based image retrieval (CBIR) algorithm, CAD searches for a set of reference images depicting “abnormalities” similar to the queried region. Based on image retrieval results and a decision algorithm, a classification score is assigned to the queried region. In this study, a reference database with 1,800 malignant mass regions and 1,800 benign and CAD-generated false-positive regions was used. A modified CBIR algorithm with a new function of stretching the attributes in the multi-dimensional space and decision scheme was optimized using a genetic algorithm. Using a leave-one-out testing method to classify suspicious mass regions, we compared the classification performance using two CBIR algorithms with either equally weighted or optimally stretched attributes. Using the modified CBIR algorithm, the area under receiver operating characteristic curve was significantly increased from 0.865 ± 0.006 to 0.897 ± 0.005 (p < 0.001). This study demonstrated the feasibility of developing an interactive CAD system with a large reference database and achieving improved performance.     

A Filmless Radiology Teaching Conference System for Pertinent Displaying and Image Searching

To solve the problems of image displays in filmless radiology conferences for the purpose of teaching, we made an experimental design of a conference system with dual 50-in. plasma monitors for displaying larger images and a shared folder containing shortcuts to images for quick display during conferences on the desktop of each client computer in a picture archival and communication system. The image quality of the monitors was evaluated using the TG18-QC test pattern. The display time of images was measured in 20 cases when the shared folder was used and when it was not. Monitor screen size and image quality, operability, display time of images, and overall impression given by the system were evaluated subjectively by five radiologists. Although the image quality of the monitor was not as high as that of the high-resolution monitors used for diagnostic radiology, its performance was good enough for teaching. The average display time using the shared folder (2.6 ± 0.39 s) was significantly shorter than without it (16.9 ± 5.04 s; p = 2.85 × 10⁻⁶). Despite the need for certain improvements in monitor size and in the operability of the system, the radiologists considered the system suitable for radiology teaching conferences. We believe that this system is useful for institutions that intend to introduce a filmless system for filmless radiology teaching conferences.     

DCE-MRI及DWI影像特征对乳腺癌病理组织学分级及Ki-67表达的预测研究

联合动态增强磁共振成像(DCE-MRI)以及弥散加权成像(DWI)的影像特征,通过建立模型,分别对乳腺癌的组织学分级以及Ki-67的表达进行预测。对144例未经过任何手术或化疗的乳腺浸润性导管癌患者的数据进行回顾性分析,这些患者均采用3T 扫描仪进行术前乳腺 MRI 检查,从中获取DCE-MRI以及DWI影像,并从 DWI 中计算得到表观扩散系数 (ADC)。对不同参数磁共振影像进行肿瘤分割,并分别从整个肿瘤区域中提取纹理特征、统计特征、形态特征等。采用无监督判别特征选择方法(UDFS)和Fisher Score算法进行特征选择,将分类模型分别应用于DCE-MRI及DWI图像数据,将得到的不同分类器进行多分类器模型融合,最终得到多参数图像的联合预测结果。为了评估所建立模型的分类性能, 通过留一法交叉验证 (LOOCV) 的方法计算ROC曲线下的面积(AUC)。对于分级任务,DCE-MRI的第二增强序列达到0.780的最优AUC,(特异度为0.647,灵敏度为0.934);对于Ki-67预测任务,DWI序列达到0.756的最优AUC(特异度为0.806,灵敏度为0.695)。经过融合,分级的预测结果提高到AUC为0.808(特异度为0.706,灵敏度为0.895),Ki-67的预测结果提高到AUC为0.783(特异度为0.778,灵敏度为0.722)。结果表明,相比采用单一参数的磁共振图像数据,DCE-MRI和DWI的影像特征联合可以提高分类器的性能。     

Quality Degradation in Lossy Wavelet Image Compression

The objective of this study was to develop a method for measuring quality degradation in lossy wavelet image compression. Quality degradation is due to denoising and edge blurring effects that cause smoothness in the compressed image. The peak Moran z histogram ratio between the reconstructed and original images is used as an index for degradation after image compression. The Moran test is applied to images randomly selected from each medical modality, computerized tomography, magnetic resonance imaging, and computed radiography and compressed using the wavelet compression at various levels. The relationship between the quality degradation and compression ratio for each image modality agrees with previous reports that showed a preference for mildly compressed images. Preliminary results show that the peak Moran z histogram ratio can be used to quantify the quality degradation in lossy image compression. The potential for this method is applications for determining the optimal compression ratio (the maximized compression without seriously degrading image quality) of an image for teleradiology.     

Development of an Automatic Classification System for Differentiation of Obstructive Lung Disease using HRCT

The motivation is to introduce new shape features and optimize the classifier to improve performance of differentiating obstructive lung diseases, based on high-resolution computerized tomography (HRCT) images. Two hundred sixty-five HRCT images from 82 subjects were selected. On each image, two experienced radiologists selected regions of interest (ROIs) representing area of severe centrilobular emphysema, mild centrilobular emphysema, bronchiolitis obliterans, or normal lung. Besides 13 textural features, additional 11 shape features were employed to evaluate the contribution of shape features. To optimize the system, various ROI size (16 × 16, 32 × 32, and 64 × 64 pixels) and other classifier parameters were tested. For automated classification, the Bayesian classifier and support vector machine (SVM) were implemented. To assess cross-validation of the system, a five-folding method was used. In the comparison of methods employing only the textural features, adding shape features yielded the significant improvement of overall sensitivity (7.3%, 6.1%, and 4.1% in the Bayesian and 9.1%, 7.5%, and 6.4% in the SVM, in the ROI size 16 × 16, 32 × 32, 64 × 64 pixels, respectively; t test, P < 0.01). After feature selection, most of cluster shape features were survived ,and the feature selected set shows better performance of the overall sensitivity (93.5 ± 1.0% in the SVM in the ROI size 64 × 64 pixels; t test, P < 0.01). Adding shape features to conventional texture features is much useful to improve classification performance of obstructive lung diseases in both Bayesian and SVM classifiers. In addition, the shape features contribute more to overall sensitivity in smaller ROI.     

Practical guidelines for radiographers to improve computed radiography image quality

Computed Radiography (CR) has become a major digital imaging modality in a modern radiological department. CR system changes workflow from the conventional way of using film/screen by employing photostimulable phosphor plate technology. This results in the changing perspectives of technical, artefacts and quality control issues in radiology departments. Guidelines for better image quality in digital medical enterprise include professional guidelines for users and the quality control programme specifically designed to serve the best quality of clinical images. Radiographers who understand technological shift of the CR from conventional method can employ optimization of CR images. Proper anatomic collimation and exposure techniques for each radiographic projection are crucial steps in producing quality digital images. Matching image processing with specific anatomy is also important factor that radiographers should realise. Successful shift from conventional to fully digitised radiology department requires skilful radiographers who utilise the technology and a successful quality control program from teamwork in the department.