Cluster kernels for semisupervised classification of VHR urban images

In this paper, we present and apply a semisupervised support vector machine based on cluster kernels for the problem of very high resolution image classification. In the proposed setting, a base kernel working with labeled samples only is deformed by a likelihood kernel encoding similarities between unlabeled examples. The resulting kernel is used to train a standard support vector machine (SVM) classifier. Experiments carried out on very high resolution (VHR) multispectral and hyperspectral images using very few labeled examples show the relevancy of the method in the context of urban image classification. Its simplicity and the small number of parameters involved make it versatile and workable by unexperimented users.

Learning spatial filters for multispectral image segmentation

We present a novel filtering method for multispectral satellite image classification. The proposed method learns a set of spatial filters that maximize class separability of binary support vector machine (SVM) through a gradient descent approach. Regularization issues are discussed in detail and a Frobenius-norm regularization is proposed to efficiently exclude uninformative filters coefficients. Experiments carried out on multiclass one-against-all classification and target detection show the capabilities of the learned spatial filters.

Active Learning of Very-High Resolution Optical Imagery with SVM: Entropy vs Margin Sampling

An active learning method is proposed for the semi-automatic selection of training sets in remote sensing image classification. The method adds iteratively to the current training set the unlabeled pixels for which the prediction of an ensemble of classifiers based on bagged training sets show maximum entropy. This way, the algorithm selects the pixels that are the most uncertain and that will improve the model if added in the training set. The user is asked to label such pixels at each iteration. Experiments using support vector machines (SVM) on an 8 classes QuickBird image show the excellent performances of the methods, that equals accuracies of both a model trained with ten times more pixels and a model whose training set has been built using a state-of-the-art SVM specific active learning method

Classification of urban multi-angular image sequences by aligning their manifolds

When dealing with multi-angular image sequences, problems of reflectance changes due either to illumination and acquisition geometry, or to interactions with the atmosphere, naturally arise. These phenomena interplay with the scene and lead to a modification of the measured radiance: for example, according to the angle of acquisition, tall objects may be seen from top or from the side and different light scatterings may affect the surfaces. This results in shifts in the acquired radiance, that make the problem of multi-angular classification harder and might lead to catastrophic results, since surfaces with the same reflectance return significantly different signals. In this paper, rather than performing atmospheric or bi-directional reflection distribution function (BRDF) correction, a non-linear manifold learning approach is used to align data structures. This method maximizes the similarity between the different acquisitions by deforming their manifold, thus enhancing the transferability of classification models among the images of the sequence.

Automatic Mallampati Classification Using Active Appearance Models

Difficult tracheal intubation assessment is an important research topic in anesthesia as failed intubations are important causes of mortality in anesthetic practice. The modified Mallampati score is widely used, alone or in conjunction with other criteria, to predict the difficulty of intubation. This work presents an automatic method to assess the modified Mallampati score from an image of a patient with the mouth wide open. For this purpose we propose an active appearance models (AAM) based method and use linear support vector machines (SVM) to select a subset of relevant features obtained using the AAM. This feature selection step proves to be essential as it improves drastically the performance of classification, which is obtained using SVM with RBF kernel and majority voting. We test our method on images of 100 patients undergoing elective surgery and achieve 97.9% accuracy in the leave-one-out crossvalidation test and provide a key element to an automatic difficult intubation assessment system.

IEF pattern classification-derived criteria for the identification of epoetin-delta in urine.

Epoetin-delta (Dynepo Shire Pharmaceuticals, Basing stoke, UK) is a synthetic form of erythropoietin (EPO) whose resemblance with endogenous EPO makes it hard to identify using the classical identification criteria. Urine samples collected from six healthy volunteers treated with epoetin-delta injections and from a control population were immuno-purified and analyzed with the usual IEF method. On the basis of the EPO profiles integration, a linear multivariate model was computed for discriminant analysis. For each sample, a pattern classification algorithm returned a bands distribution and intensity score (bands intensity score) saying how representative this sample is of one of the two classes, positive or negative. Effort profiles were also integrated in the model. The method yielded a good sensitivity versus specificity relation and was used to determine the detection window of the molecule following multiple injections. The bands intensity score, which can be generalized to epoetin-alpha and epoetin-beta, is proposed as an alternative criterion and a supplementary evidence for the identification of EPO abuse.

Comparison of Energy Minimization Methods for 3-D Brain Tissue Classification

This paper presents 3-D brain tissue classificationschemes using three recent promising energy minimizationmethods for Markov random fields: graph cuts, loopybelief propagation and tree-reweighted message passing.The classification is performed using the well knownfinite Gaussian mixture Markov Random Field model.Results from the above methods are compared with widelyused iterative conditional modes algorithm. Theevaluation is performed on a dataset containing simulatedT1-weighted MR brain volumes with varying noise andintensity non-uniformities. The comparisons are performedin terms of energies as well as based on ground truthsegmentations, using various quantitative metrics.

Assessment of a new MRI classification of spinal stenosis based on cross-sectional morphology of the dural sac : FM2

Introduction: Quantitative measures of degree of lumbar spinal stenosis (LSS) such as antero-posterior diameter of the canal or dural sac cross sectional area vary widely and do not correlate with clinical symptoms or results of surgical decompression. In an effort to improve quantification of stenosis we have developed a grading system based on the morphology of the dural sac and its contents as seen on T2 axial images. The grading comprises seven categories ranging form normal to the most severe stenosis and takes into account the ratio of rootlet/CSF content. Material and methods: Fifty T2 axial MRI images taken at disc level from twenty seven symptomatic lumbar spinal stenosis patients who underwent decompressive surgery were classified into seven categories by five observers and reclassified 2 weeks later by the same investigators. Intra- and inter-observer reliability of the classification were assessed using Cohen's and Fleiss' kappa statistics, respectively. Results: Generally, the morphology grading system itself was well adopted by the observers. Its success in application is strongly influenced by the identification of the dural sac. The average intraobserver Cohen's kappa was 0.53 ± 0.2. The inter-observer Fleiss' kappa was 0.38 ± 0.02 in the first rating and 0.3 ± 0.03 in the second rating repeated after two weeks. Discussion: In this attempt, the teaching of the observers was limited to an introduction to the general idea of the morphology grading system and one example MRI image per category. The identification of the dimension of the dural sac may be a difficult issue in absence of complete T1 T2 MRI image series as it was the case here. The similarity of the CSF to possibly present fat on T2 images was the main reason of mismatch in the assignment of the cases to a category. The Fleiss correlation factors of the five observers are fair and the proposed morphology grading system is promising.

Automatic classification of MR scans in Alzheimer's disease.

To be diagnostically useful, structural MRI must reliably distinguish Alzheimer's disease (AD) from normal aging in individual scans. Recent advances in statistical learning theory have led to the application of support vector machines to MRI for detection of a variety of disease states. The aims of this study were to assess how successfully support vector machines assigned individual diagnoses and to determine whether data-sets combined from multiple scanners and different centres could be used to obtain effective classification of scans. We used linear support vector machines to classify the grey matter segment of T1-weighted MR scans from pathologically proven AD patients and cognitively normal elderly individuals obtained from two centres with different scanning equipment. Because the clinical diagnosis of mild AD is difficult we also tested the ability of support vector machines to differentiate control scans from patients without post-mortem confirmation. Finally we sought to use these methods to differentiate scans between patients suffering from AD from those with frontotemporal lobar degeneration. Up to 96% of pathologically verified AD patients were correctly classified using whole brain images. Data from different centres were successfully combined achieving comparable results from the separate analyses. Importantly, data from one centre could be used to train a support vector machine to accurately differentiate AD and normal ageing scans obtained from another centre with different subjects and different scanner equipment. Patients with mild, clinically probable AD and age/sex matched controls were correctly separated in 89% of cases which is compatible with published diagnosis rates in the best clinical centres. This method correctly assigned 89% of patients with post-mortem confirmed diagnosis of either AD or frontotemporal lobar degeneration to their respective group. Our study leads to three conclusions: Firstly, support vector machines successfully separate patients with AD from healthy aging subjects. Secondly, they perform well in the differential diagnosis of two different forms of dementia. Thirdly, the method is robust and can be generalized across different centres. This suggests an important role for computer based diagnostic image analysis for clinical practice.

Reliability of the image evaluation system PGMI

Purpose: Many countries used the PGMI (P=perfect, G=good, M=moderate, I=inadequate) classification system for assessing the quality of mammograms. Limits inherent to the subjectivity of this classification have been shown. Prior to introducing this system in Switzerland, we wanted to better understand the origin of this subjectivity in order to minimize it. Our study aimed at identifying the main determinants of the variability of the PGMI system and which criteria are the most subjected to subjectivity. Methods and Materials: A focus group composed of 2 experienced radiographers and 2 radiologists specified each PGMI criterion. Ten raters (6 radiographers and 4 radiologists) evaluated twice a panel of 40 randomly selected mammograms (20 analogic and 20 digital) according to these specified PGMI criteria. The PGMI classification was assessed and the intra- and inter-rater reliability was tested for each professional group (radiographer vs radiologist), image technology (analogic vs digital) and PGMI criterion. Results: Some 3,200 images were assessed. The intra-rater reliability appears to be weak, particularly in respect to inter-rater variability. Subjectivity appears to be largely independent of the professional group and image technology. Aspects of the PGMI classification criteria most subjected to variability were identified. Conclusion: Post-test discussions enabled to specify more precisely some criteria. This should reduce subjectivity when applying the PGMI classification system. A concomitant, important effort in training radiographers is also necessary.

Statistical assessment of dataset shift and model portability in multi-angle in-track image acquisitions

In this study we propose an evaluation of the angular effects altering the spectral response of the land-cover over multi-angle remote sensing image acquisitions. The shift in the statistical distribution of the pixels observed in an in-track sequence of WorldView-2 images is analyzed by means of a kernel-based measure of distance between probability distributions. Afterwards, the portability of supervised classifiers across the sequence is investigated by looking at the evolution of the classification accuracy with respect to the changing observation angle. In this context, the efficiency of various physically and statistically based preprocessing methods in obtaining angle-invariant data spaces is compared and possible synergies are discussed.

Recent advances in remote sensing image processing

Remote sensing image processing is nowadays a mature research area. The techniques developed in the field allow many real-life applications with great societal value. For instance, urban monitoring, fire detection or flood prediction can have a great impact on economical and environmental issues. To attain such objectives, the remote sensing community has turned into a multidisciplinary field of science that embraces physics, signal theory, computer science, electronics, and communications. From a machine learning and signal/image processing point of view, all the applications are tackled under specific formalisms, such as classification and clustering, regression and function approximation, image coding, restoration and enhancement, source unmixing, data fusion or feature selection and extraction. This paper serves as a survey of methods and applications, and reviews the last methodological advances in remote sensing image processing.

Validation of tissue modelization and classification techniques in T1-weighted MR brain images

We propose a deep study on tissue modelization andclassification Techniques on T1-weighted MR images. Threeapproaches have been taken into account to perform thisvalidation study. Two of them are based on FiniteGaussian Mixture (FGM) model. The first one consists onlyin pure gaussian distributions (FGM-EM). The second oneuses a different model for partial volume (PV) (FGM-GA).The third one is based on a Hidden Markov Random Field(HMRF) model. All methods have been tested on a DigitalBrain Phantom image considered as the ground truth. Noiseand intensity non-uniformities have been added tosimulate real image conditions. Also the effect of ananisotropic filter is considered. Results demonstratethat methods relying in both intensity and spatialinformation are in general more robust to noise andinhomogeneities. However, in some cases there is nosignificant differences between all presented methods.

Comparison and validation of tissue modelization and statistical classification methods in T1-weighted MR brain images.

This paper presents a validation study on statistical nonsupervised brain tissue classification techniques in magnetic resonance (MR) images. Several image models assuming different hypotheses regarding the intensity distribution model, the spatial model and the number of classes are assessed. The methods are tested on simulated data for which the classification ground truth is known. Different noise and intensity nonuniformities are added to simulate real imaging conditions. No enhancement of the image quality is considered either before or during the classification process. This way, the accuracy of the methods and their robustness against image artifacts are tested. Classification is also performed on real data where a quantitative validation compares the methods' results with an estimated ground truth from manual segmentations by experts. Validity of the various classification methods in the labeling of the image as well as in the tissue volume is estimated with different local and global measures. Results demonstrate that methods relying on both intensity and spatial information are more robust to noise and field inhomogeneities. We also demonstrate that partial volume is not perfectly modeled, even though methods that account for mixture classes outperform methods that only consider pure Gaussian classes. Finally, we show that simulated data results can also be extended to real data.