The main intention of content based medical image retrieval (CBMIR) is to efficiently retrieve medical images that are visually similar to a query image. Medical images are usually retrieved on the basis of low level and high level features. This work deals with the concept of texture based spine MRI image retrieval in the wavelet compressed domain. We use two statistical methods such as Haralick features and texture??spectrum features for spine MRI image feature extraction and project the features to a set of signatures. The obtained statistical features are classifying, according to various types of spine MRI images using k-means clustering algorithm. Then the image retrieval is carried out by calculating the distance between the signatures in the database images and the query image. This method is applied around 500 spine MRI images and improvements of retrieval efficiency are found with standard precision and recall analysis.

This paper describes the last round of the medical image annotation task in ImageCLEF 2009. After four years, we defined the task as a survey of all the past experience. Seven groups participated to the challenge submitting nineteen runs. They were asked to train their algorithms on 12677 images, labelled according to four different settings, and to classify 1733 images in the four annotation frameworks. The aim is to understand how each strategy answers to the increasing number of classes and to the unbalancing. A plain classification scheme using support vector machines and local descriptors outperformed the other methods.

This work reports baseline results for the CLEF 2008 Medical Automatic Annotation Task (MAAT) by applying a classifier with a fixed parameter set to all tasks 2005 - 2008. A nearest-neighbor (NN) classifier is used, which uses a weighted combination of three distance and similarity measures operating on global image features: Scaled-down representations of the images are compared using models for the typical variability in the image data, mainly translation, local deformation, and radiation dose. In addition, a distance measure based on texture features is used. In 2008, the baseline classifier yields error scores of 170.34 and 182.77 for k = 1 and k = 5 when the full code is reported, which corresponds to error rates of 51.3% and 52.8% for 1-NN and 5-NN, respectively. Judging the relative increases of the number of classes and the error rates over the years, MAAT 2008 is estimated to be the most difficult in the four years.

The medical automatic annotation task issued by the cross language evaluation forum (CLEF) aims at a fair comparison of state-of-the art algorithms for medical content-based image retrieval (CBIR). The contribution of this work is twofold: at first, a logical decomposition of the CBIR task is presented, and key elements to support the relevant steps are identified: (i) implementation of algorithms for feature extraction, feature comparison, and classifier combination, (ii) visualization of extracted features and retrieval results, (iii) generic evaluation of retrieval algorithms, and (iv) optimization of the parameters for the retrieval algorithms and their combination. Data structures and tools to address these key elements are integrated into an existing framework for image retrieval in medical applications (IRMA). Secondly, baseline results for the CLEF annotation tasks 2005–2007 are provided applying the IRMA framework, where global features and corresponding distance measures are combined within a nearest neighbor approach. Using identical classifier parameters and combination weights for each year shows that the task difficulty decreases over the years. The declining rank of the baseline submission also indicates the overall advances in CBIR concepts. Furthermore, a rough comparison between participants who submitted in only one of the years becomes possible.

This paper provides baseline results for the medical automatic annotation task of CLEF 2007 by applying the image retrieval in medical applications (IRMA)-based algorithms previously used in 2005 and 2006, with identical parameterization. Three classifiers based on global image features are combined within a nearest neighbor (NN) approach: texture histograms and two distance measures, which are applied on down-scaled versions of the original images and model common variabilities in the image data. According to the evaluation scheme introduced in 2007, which uses the hierarchical structure of the coding scheme for the categorization, the baseline classifier yields scores of 51.29 and 52.54 when reporting full codes for 1-NN and 5-NN, respectively. This corresponds to error rates of 20.0% and 18.0% (rank 18 among 68 runs), respectively. Improvements via addressing the code hierarchy were not obtained. However, comparing the baseline results yields that the 2007 task was slightly easier than the previous ones.

The impact of image pattern recognition on accessing large databases of medical images has recently been explored, and content-based image retrieval (CBIR) in medical applications (IRMA) is researched. At the present, however, the impact of image retrieval on diagnosis is limited, and practical applications are scarce. One reason is the lack of suitable mechanisms for query refinement, in particular, the ability to (1) restore previous session states, (2) combine individual queries by Boolean operators, and (3) provide continuous-valued query refinement. This paper presents a powerful user interface for CBIR that provides all three mechanisms for extended query refinement. The various mechanisms of man–machine interaction during a retrieval session are grouped into four classes: (1) output modules, (2) parameter modules, (3) transaction modules, and (4) process modules, all of which are controlled by a detailed query logging. The query logging is linked to a relational database. Nested loops for interaction provide a maximum of flexibility within a minimum of complexity, as the entire data flow is still controlled within a single Web page. Our approach is implemented to support various modalities, orientations, and body regions using global features that model gray scale, texture, structure, and global shape characteristics. The resulting extended query refinement has a significant impact for medical CBIR applications.

Contemporary picture archiving and communication systems are limited in managing large and varied image collections, because content-based image retrieval (CBIR) methods are unavailable. In this paper, an XML-based data and resource exchange framework is defined using open standards and software to enable specialized CBIR systems to act as geographically distributed toolkits. The approach enables communication and collaboration between two or more geographically separated complementary systems with possibly different architectures and developed on different platforms, and specialized for different image modalities and characteristics. The resulting synergy provides the user with a rich functionality operating within a familiar Web browser interface, making the combined system portable and independent of location and underlying user operating systems. We describe the coupling of the Image Retrieval in Medical Applications (IRMA) system and the Spine Pathology and Image Retrieval System (SPIRS) as proof of this concept.

The relevant contents of medical images can often be described as a composition of objects with distinct relationships. For a given setting, a generalization by the statistical distributions of regional and relational attributes can be obtained. These yield a structural prototype graph which in turn can be used for the identification of objects in new images. As new image contents are represented by hierarchical attributed region adjacency graphs, the task of object identification corresponds to the problem of inexact graph matching. For this purpose, the graph edit distance as well as a Hopfield-net are used and evaluated for the example application of boneidentification in hand-radiographs. The structural prototypes improve recall by 9% and 17%, respectively, in comparison to a traditional approach without relational information.

Objectives: Content-based access (CBA) to medical image archives, i.e. data retrieval by means of image-based numerical features computed automatically, has capabilities to improve diagnostics, research and education. In this study, the applicability of CBA methods in dentomaxillofacial radiology is evaluated. Methods: Recent research has discovered numerical features that were successfully applied for an automatic categorization of radiographs. In our experiments, oral and maxillofacial radiographs were obtained from the day-to-day routine of a university hospital and labelled by an experienced dental radiologist regarding the technique and direction of imaging, as well as the displayed anatomy and biosystem. In total, 2000 radiographs of 71 classes with at least 10 samples per class were analysed. A combination of co-occurrence-based texture features and correlation-based similarity measures was used in leaving-one-out experiments for automatic classification. The impact of automatic detection and separation of multi-field images and automatic separability of biosystems were analysed. Results: Automatic categorization yielded error rates of 23.20%, 7.95% and 4.40% with respect to a correct match within the first, fifth and tenth best returns. These figures improved to 23.05%, 7.00%, 4.20%, and 20.05%, 5.65% and 3.25% if automatic decomposition was applied and the classifier was optimized to the dentomaxillofacial imagery, respectively. The dentulous and implant systems were difficult to distinguish. Experiments on non-dental radiographs (10 000 images of 57 classes) yielded 12.6%, 5.6% and 3.6%. Conclusion: Using the same numerical features as in medical radiology, oral and maxillofacial radiographs can be reliably indexed by global texture features for CBA and data mining.

The ImageCLEF 2006 medical automatic annotation task encompasses 11,000 images from 116 categories, compared to 57 categories for 10,000 images of the similar task in 2005. As a baseline for comparison, a run using the same classifiers with the identical parameterization as in 2005 is submitted. In addition, the parameterization of the classifier was optimized according to the 9,000/1,000 split of the 2006 training data. In particular, texture-based classifiers are combined in parallel with classifiers, which use spatial intensity information to model common variabilities among medical images. However, all individual classifiers are based on global features, i.e. one feature vector describes the entire image. The parameterization from 2005 yields an error rate of 21.7%, which ranks 13th among the 28 submissions. The optimized classifier yields 21.4% error rate (rank 12), which is insignificantly better.

A combination of several classifiers using global features for the content description of medical images is proposed. Beside two texture features, downscaled representations of the original images are used, which preserve spatial information and utilize distance measures which are robust regarding common variations in radiation dose, translation, and local deformation. No query refinement mechanisms are used. The single classifiers are used within a parallel combination scheme, with the optimization set being used to obtain the best weighing parameters. For the medical automatic annotation task, a categorization rate of 78.6% is obtained, which ranks 12th among 28 submissions. When applied in the medical retrieval task, this combination of classifiers yields a mean average precision (MAP) of 0.0172, which is rank 11 of 11 submitted runs for automatic, visual only systems.

With the increasing amount of medical images that are routinely acquired in today’s healthcare institutions, common methods of image management become inefficient. Even in modern picture archiving and communication systems that are based on the DICOM standard, image data is addressed by alphanumerical indexes such as patient name and examination date. In this software demonstration, the capabilities of content-based access to medical images is exemplified. Radiographs that are arbitrarily selected from the internet can be compared to the reference database by means of numerical features extracted from the pixel values. This allows for automatic identification of image content such as body region or the biomedical system images, as well as the determination of the modality used for image acquisition. Since the features of the system capture global, local, and structural image properties, content-based access impacts research, teaching and patient care.

Medizinische Bildinhalte sind häufig aus mehreren Bildregionen zusammengesetzt, die zueinander in bestimmten Verhältnissen stehen. Diese lassen sich über Wahrscheinlichkeitsverteilungen regionaler und relationaler Attribute zu einem Strukturprototypen verallgemeinern und mit Szenen unbekannten Inhalts vergleichen. Die semiautomatische Synthese eines Strukturprototyps sowie seine Anwendung zur Szenenanalyse wird am Beispiel von Handradiographien vorgestellt. Durch Verwendung der Strukturinformation wird hier eine Steigerung der Erkennungsrate um 17% erzielt.

A combination of several classifiers using global features for the content description of medical images is proposed. Beside well known texture histogram features, downscaled representations of the original images are used, which preserve spatial information and utilize distance measures which are robust with regard to common variations in radiation dose, translation, and local deformation. These features were evaluated for the annotation task and the retrieval task in ImageCLEF 2005 without using additional textual information or query refinement mechanisms. For the annotation task, a categorization rate of 86.7% was obtained, which ranks second among all submissions. When applied in the retrieval task, the image content descriptors yielded a mean average precision (MAP) of 0.0751, which is rank 14 of 28 submitted runs. As the image deformation model is not fit for interactive retrieval tasks, two mechanisms are evaluated with regard to the trade-off between loss of accuracy and speed increase: hierarchical filtering and prototype selection.

Dieser Beitrag präsentiert Lösungsvorschläge zur effektiven Unterstützung bei der Implementierung und Bereitstellung von Verfahren zum inhaltsbasierten Zugriff auf Bilddatenbanken. Hierzu ist eine strikte Trennung zwischen Merkmalsextraktion, Merkmalsspeicherung, Merkmalsvergleich und den Benutzerschnittstellen sinnvoll. Dadurch wird es möglich, existierende Komponenten in verschiedenen Kontexten wiederzuverwenden, was Softwarequalität, Entwicklungszeit und einheitlichen Richtlinien bei den Benutzerschnittstellen zugute kommt.

Segmentation of medical images is fundamental for many high-level applications. Unsupervised techniques such as region growing or merging allow automated processing of large data amounts. The regions are usually described by a mean feature vector, and the merging decisions are based on the Euclidean distance. This kind of similarity model is strictly local, since the feature vector of each region is calculated without evaluating the region's surrounding. Therefore, region merging often fails to extract visually comprehensible and anatomically relevant regions. In our approach, the local model is extended. Regional similarity is calculated for a pair of adjacent regions, e.g. considering the contrast along their common border. Global similarity components are obtained by analyzing the entire image partitioning before and after a hypothetical merge. Hierarchical similarities are derived from the iteration history. Local, regional, global, and hierarchical components are combined task-specifically guiding the iterative region merging process. Starting with an initial watershed segmentation, the process terminates when the entire image is represented as a single region. A complete segmentation takes only a few seconds. Our approach is evaluated contextually on plain skeletal radiographs of human hands acquired for bone maturity assessment. Region merging based on a local model fails to detect most bones, while a correct localization and delineation is obtained with the combined model. For quantitative evaluation of delineation precision, a gold standard is computed from ten reference segmentations of each radiograph obtained manually. The relative error of labeled pixels is 15.7 %, which is slightly more than the mean error of the ten manual references to the gold standard (12 %). The flexible and powerful similarity model can be adopted to many other segmentation tasks in medical imaging.

Categorization of medical images means selecting the appropriate class for a given image out of a set of pre-defined categories. This is an important step for data mining and content-based image retrieval (CBIR). So far, published approaches are capable to distinguish up to 10 categories. In this paper, we evaluate automatic categorization into more than 80 categories describing the imaging modality and direction as well as the body part and biological system examined. Based on 6231 reference images from hospital routine, 85.5% correctness is obtained combining global texture features with scaled images. With a frequency of 97.7%, the correct class is within the best ten matches, which is sufficient for medical CBIR applications.