Marco Loog

Comparative study of retinal vessel segmentation methods on a new publicly available database

In this work we compare the performance of a number of vessel segmentation algorithms on a newly constructed retinal vessel image database. Retinal vessel segmentation is important for the detection of numerous eye diseases and plays an important role in automatic retinal disease screening systems. A large number of methods for retinal vessel segmentation have been published, yet an evaluation of these methods on a common database of screening images has not been performed. To compare the performance of retinal vessel segmentation methods we have constructed a large database of retinal images. The database contains forty images in which the vessel trees have been manually segmented. For twenty of those forty images a second independent manual segmentation is available. This allows for a comparison between the performance of automatic methods and the performance of a human observer. The database is available to the research community. Interested researchers are encouraged to upload their segmentation results to our website (http://www.isi.uu.nl/Research/Databases). The performance of five different algorithms has been compared. Four of these methods have been implemented as described in the literature. The fifth pixel classification based method was developed specifically for the segmentation of retinal vessels and is the only supervised method in this test. We define the segmentation accuracy with respect to our gold standard as the performance measure. Results show that the pixel classification method performs best, but the second observer still performs significantly better.

Multiclass linear dimension reduction by weighted pairwise Fisher criteria

We derive a class of computationally inexpensive linear dimension reduction criteria by introducing a weighted variant of the well-known K-class Fisher criterion associated with linear discriminant analysis (LDA). It can be seen that LDA weights contributions of individual class pairs according to the Euclidean distance of the respective class means. We generalize upon LDA by introducing a different weighting function.

Linear dimensionality reduction via a heteroscedastic extension of LDA: the Chernoff criterion

We propose an eigenvector-based heteroscedastic linear dimension reduction (LDR) technique for multiclass data. The technique is based on a heteroscedastic two-class technique which utilizes the so-called Chernoff criterion, and successfully extends the well-known linear discriminant analysis (LDA). The latter, which is based on the Fisher criterion, is incapable of dealing with heteroscedastic data in a proper way. For the two-class case, the between-class scatter is generalized so to capture differences in (co)variances. It is shown that the classical notion of between-class scatter can be associated with Euclidean distances between class means. From this viewpoint, the between-class scatter is generalized by employing the Chernoff distance measure, leading to our proposed heteroscedastic measure. Finally, using the results from the two-class case, a multiclass extension of the Chernoff criterion is proposed. This criterion combines separation information present in the class mean as well as the class covariance matrices. Extensive experiments and a comparison with similar dimension reduction techniques are presented.

Segmentation of anatomical structures in chest radiographs using supervised methods: a comparative study on a public database.

The task of segmenting the lung fields, the heart, and the clavicles in standard posterior-anterior chest radiographs is considered. Three supervised segmentation methods are compared: active shape models, active appearance models and a multi-resolution pixel classification method that employs a multi-scale filter bank of Gaussian derivatives and a k-nearest-neighbors classifier. The methods have been tested on a publicly available database of 247 chest radiographs, in which all objects have been manually segmented by two human observers. A parameter optimization for active shape models is presented, and it is shown that this optimization improves performance significantly. It is demonstrated that the standard active appearance model scheme performs poorly, but large improvements can be obtained by including areas outside the objects into the model. For lung field segmentation, all methods perform well, with pixel classification giving the best results: a paired t-test showed no significant performance difference between pixel classification and an independent human observer. For heart segmentation, all methods perform comparably, but significantly worse than a human observer. Clavicle segmentation is a hard problem for all methods; best results are obtained with active shape models, but human performance is substantially better. In addition, several hybrid systems are investigated. For heart segmentation, where the separate systems perform comparably, significantly better performance can be obtained by combining the results with majority voting. As an application, the cardio-thoracic ratio is computed automatically from the segmentation results. Bland and Altman plots indicate that all methods perform well when compared to the gold standard, with confidence intervals from pixel classification and active appearance modeling very close to those of a human observer. All results, including the manual segmentations, have been made publicly available to facilitate future comparative studies.

Quantitative Comparison of Spot Detection Methods in Fluorescence Microscopy

Quantitative analysis of biological image data generally involves the detection of many subresolution spots. Especially in live cell imaging, for which fluorescence microscopy is often used, the signal-to-noise ratio (SNR) can be extremely low, making automated spot detection a very challenging task. In the past, many methods have been proposed to perform this task, but a thorough quantitative evaluation and comparison of these methods is lacking in the literature. In this paper, we evaluate the performance of the most frequently used detection methods for this purpose. These include seven unsupervised and two supervised methods. We perform experiments on synthetic images of three different types, for which the ground truth was available, as well as on real image data sets acquired for two different biological studies, for which we obtained expert manual annotations to compare with. The results from both types of experiments suggest that for very low SNRs ( ¿ 2), the supervised (machine learning) methods perform best overall. Of the unsupervised methods, the detectors based on the so-called h -dome transform from mathematical morphology or the multiscale variance-stabilizing transform perform comparably, and have the advantage that they do not require a cumbersome learning stage. At high SNRs ( > 5), the difference in performance of all considered detectors becomes negligible.

A computer-aided diagnosis system for detection of lung nodules in chest radiographs with an evaluation on a public database

A computer algorithm for nodule detection in chest radiographs is presented. The algorithm consists of four main steps: (i) image preprocessing; (ii) nodule candidate detection; (iii) feature extraction; (iv) candidate classification. Two optional extensions to this scheme are tested: candidate selection and candidate segmentation. The output of step (ii) is a list of circles, which can be transformed into more detailed contours by the extra candidate segmentation step. In addition, the candidate selection step (which is a classification step using a small number of features) can be used to reduce the list of nodule candidates before step (iii). The algorithm uses multi-scale techniques in several stages of the scheme: Candidates are found by looking for local intensity maxima in Gaussian scale space; nodule boundaries are detected by tracing edge points found at large scales down to pixel scale; some of the features used for classification are taken from a multi-scale Gaussian filterbank. Experiments with this scheme (with and without the segmentation and selection steps) are carried out on a previously characterized, publicly available database, that contains a large number of very subtle nodules. For this database, counting as detections only those nodules that were indicated with a confidence level of 50% or more, radiologists previously detected 70% of the nodules. For our algorithm, it turns out that the selection step does have an added value for the system, while segmentation does not lead to a clear improvement. With the scheme with the best performance, accepting on average two false positives per image results in the identification of 51% of all nodules. For four false positives, this increases to 67%. This is close to the previously reported 70% detection rate of the radiologists.

Identification of progressors in osteoarthritis by combining biochemical and MRI-based markers

IntroductionAt present, no disease-modifying osteoarthritis drugs (DMOADS) are approved by the FDA (US Food and Drug Administration); possibly partly due to inadequate trial design since efficacy demonstration requires disease progression in the placebo group. We investigated whether combinations of biochemical and magnetic resonance imaging (MRI)-based markers provided effective diagnostic and prognostic tools for identifying subjects with high risk of progression. Specifically, we investigated aggregate cartilage longevity markers combining markers of breakdown, quantity, and quality.MethodsThe study included healthy individuals and subjects with radiographic osteoarthritis. In total, 159 subjects (48% female, age 56.0 ± 15.9 years, body mass index 26.1 ± 4.2 kg/m2) were recruited. At baseline and after 21 months, biochemical (urinary collagen type II C-telopeptide fragment, CTX-II) and MRI-based markers were quantified. MRI markers included cartilage volume, thickness, area, roughness, homogeneity, and curvature in the medial tibio-femoral compartment. Joint space width was measured from radiographs and at 21 months to assess progression of joint damage.ResultsCartilage roughness had the highest diagnostic accuracy quantified as the area under the receiver-operator characteristics curve (AUC) of 0.80 (95% confidence interval: 0.69 to 0.91) among the individual markers (higher than all others, P < 0.05) to distinguish subjects with radiographic osteoarthritis from healthy controls. Diagnostically, cartilage longevity scored AUC 0.84 (0.77 to 0.92, higher than roughness: P = 0.03). For prediction of longitudinal radiographic progression based on baseline marker values, the individual prognostic marker with highest AUC was homogeneity at 0.71 (0.56 to 0.81). Prognostically, cartilage longevity scored AUC 0.77 (0.62 to 0.90, borderline higher than homogeneity: P = 0.12). When comparing patients in the highest quartile for the longevity score to lowest quartile, the odds ratio of progression was 20.0 (95% confidence interval: 6.4 to 62.1).ConclusionsCombination of biochemical and MRI-based biomarkers improved diagnosis and prognosis of knee osteoarthritis and may be useful to select high-risk patients for inclusion in DMOAD clinical trials.

Segmentation of the posterior ribs in chest radiographs using iterated contextual pixel classification

The task of segmenting the posterior ribs within the lung fields of standard posteroanterior chest radiographs is considered. To this end, an iterative, pixel-based, supervised, statistical classification method is used, which is called iterated contextual pixel classification (ICPC). Starting from an initial rib segmentation obtained from pixel classification, ICPC updates it by reclassifying every pixel, based on the original features and, additionally, class label information of pixels in the neighborhood of the pixel to be reclassified. The method is evaluated on 30 radiographs taken from the JSRT (Japanese Society of Radiological Technology) database. All posterior ribs within the lung fields in these images have been traced manually by two observers. The first observers segmentations are set as the gold standard; ICPC is trained using these segmentations. In a sixfold cross-validation experiment, ICPC achieves a classification accuracy of 0.86 /spl plusmn/ 0.06, as compared to 0.94 /spl plusmn/ 0.02 for the second human observer.

A texton-based approach for the classification of lung parenchyma in CT images

In this paper, a texton-based classification system based on raw pixel representation along with a support vector machine with radial basis function kernel is proposed for the classification of emphysema in computed tomography images of the lung. The proposed approach is tested on 168 annotated regions of interest consisting of normal tissue, centrilobular emphysema, and paraseptal emphysema. The results show the superiority of the proposed approach to common techniques in the literature including moments of the histogram of filter responses based on Gaussian derivatives. The performance of the proposed system, with an accuracy of 96.43%, also slightly improves over a recently proposed approach based on local binary patterns.

Multiple instance learning with bag dissimilarities

Multiple instance learning (MIL) is concerned with learning from sets (bags) of objects (instances), where the individual instance labels are ambiguous. In this setting, supervised learning cannot be applied directly. Often, specialized MIL methods learn by making additional assumptions about the relationship of the bag labels and instance labels. Such assumptions may fit a particular dataset, but do not generalize to the whole range of MIL problems. Other MIL methods shift the focus of assumptions from the labels to the overall (dis)similarity of bags, and therefore learn from bags directly. We propose to represent each bag by a vector of its dissimilarities to other bags in the training set, and treat these dissimilarities as a feature representation. We show several alternatives to define a dissimilarity between bags and discuss which definitions are more suitable for particular MIL problems. The experimental results show that the proposed approach is computationally inexpensive, yet very competitive with state-of-the-art algorithms on a wide range of MIL datasets. HighlightsA general bag dissimilarities framework for multiple instance learning is explored.Point set distances and distribution distances are considered.Metric dissimilarities are not necessarily more informative.Results are competitive with, or outperform state-of-the-art algorithms.Practical suggestions for end-users are provided.