Christiaan Perneel

A new approach to the validation of tissue microarrays

Although tissue microarrays (TMA) have been widely used for a number of years, it is still not clear how many core biopsies should be taken to determine a reliable value for percentage positivity or how much heterogeneity in marker expression influences this number. The first aim of this study was to validate the human visual semi‐quantitative scoring system for positive staining of tumour tissue with the exact values determined from computer‐generated images. The second aim was to determine the minimum number of core biopsies needed to estimate percentage positivity reliably when the immunohistochemical staining pattern is heterogeneous and scored in a non‐binary way. Tissue sections from ten colorectal cancer specimens were stained for carbonic anhydrase IX (CA IX). The staining patterns were digitized and 400 artificial computer‐generated images were generated to test the accuracy of the human scoring system. To determine the minimal number of core biopsies needed to account for tumour heterogeneity, 50 (artificial) core biopsies per section were taken from the tumoural region of the ten digitally recorded full tissue sections. Based on the semi‐quantitative scores from the 50 core biopsies per section, 2500 × n (n = 1–10 core biopsies) experimental core biopsies were then generated and scores recorded. After comparison with field‐by‐field analysis from the tumoural region of the whole tissue section, the number of core biopsies that need to be taken to minimize the influence of heterogeneity could be determined. In conclusion, visual scoring accurately estimated the percentage positivity and the percentage tumour present in a section, as judged by comparison with the artificial images. The exact number of core biopsies that has to be examined to determine tumour marker positivity using TMAs is affected by the degree of heterogeneity in the expression pattern of the protein, but for most purposes at least four is recommended. Copyright

Fusion of PolSAR and PolInSAR data for land cover classification.

Abstract The main research goal of this study is to investigate the complementarity and fusion of different frequencies (L- and P-band), polarimetric SAR (PolSAR) and polarimetric interferometric (PolInSAR) data for land cover classification. A large feature set was derived from each of these four modalities and a two-level fusion method was developed: Logistic regression (LR) as ‘feature-level fusion’ and the neural-network (NN) method for higher level fusion. For comparison, a support vector machine (SVM) was also applied. NN and SVM were applied on various combinations of the feature sets. The results show that for both NN and SVM, the overall accuracy for each of the fused sets is better than the accuracy for the separate feature sets. Moreover, that fused features from different SAR frequencies are complementary and adequate for land cover classification and that PolInSAR is complementary to PolSAR information and that both are essential for producing accurate land cover classification.

Optimization of fuzzy expert systems using genetic algorithms and neural networks

In this paper, fuzzy logic theory is used to build a specific decision-making system for heuristic search algorithms. Such algorithms are typically used for expert systems. To improve the performance of the overall system, a set of important parameters of the decision-making system is identified. Two optimization methods for the learning of the optimum parameters, namely genetic algorithms and gradient-descent techniques based on a neural network formulation of the problem, are used to obtain an improvement of the performance. The decision-making system and both optimization methods are tested on a target recognition system. >

A multi-institutional analysis comparing adjuvant and salvage radiation therapy for high-risk prostate cancer patients with undetectable PSA after prostatectomy.

BACKGROUND AND PURPOSE In men with adverse pathology at the time of radical prostatectomy (RP), the most appropriate timing to administer radiotherapy (RT) remains a subject for debate. To determine whether salvage radiotherapy (SRT) upon early prostate-specific antigen (PSA) relapse is equivalent to immediate adjuvant radiotherapy (ART) post RP. MATERIAL AND METHODS 130 patients receiving ART and 89 receiving SRT were identified. All had an undetectable PSA after RP. Homogeneous subgroups were built based on the status (±) of lymphatic invasion (LVI) and surgical margins (SM), to allow a comparison of ART and SRT. Biochemical disease-free survival (bDFS) was calculated from the date of surgery and from the end of RT. The multivariate analysis was performed using the Cox Proportional hazard model. RESULTS In the SM-/LVI- and SM+/LVI- groups, SRT was a significant predictor of a decreased bDFS from the date of surgery, while in the SM+/LVI+ group, there was a trend towards significance. From the end of RT, SRT was also a significant predictor of a decreased bDFS in three patient groups: SM-/LVI-, SM+/LVI- and SM+/LVI+. Gleason score >7 showed to be another factor on multivariate analysis associated with decreased bDFS in the SM-/LVI- group, from the date of surgery and end of RT. Preoperative PSA was a significant predictor in the SM-/LVI- group from the date of RP only. CONCLUSIONS Immediate ART post RP for patients with high risk features in the prostatectomy specimen significantly reduces bDFS after RP compared with early SRT upon PSA relapse.

Molecular and clinico-pathological markers in rectal cancer: a tissue micro-array study.

AimsThe aims of the study were to study the effect of pre-operative treatment on the expression of tumour-related proteins and to correlate the expression of these proteins with response and survival of patients with advanced rectal cancer.Materials and methodsTissue micro-arrays from pre- and post-treatment biopsies of 99 patients with rectal cancer treated with pre-operative (chemo)radiotherapy were stained for epidermal growth factor receptor (EGFR), carbonic anhydrase IX, Ki67, vascular endothelial growth factor, cyclo-oxygenase 2 (COX-2) and cleaved cytokeratin 18 (c-CK18). Also, fibro-inflammatory alterations after treatment were evaluated.ResultsPre-operative (chemo)radiotherapy caused fibro-inflammatory changes, a downregulation of proliferation (Ki67) and EGFR and an upregulation of apoptosis (cleaved CK18). Patients with a good regression during pre-operative treatment showed less proliferating and apoptotic cells in the resection specimen. Multivariate analysis showed that T downstaging, fibro-inflammatory changes in the resection specimen and COX-2 expression in the biopsy correlated with overall survival.ConclusionsPre-operative treatment has an effect on proliferation, apoptosis, inflammation and EGFR expression. The classical clinical parameters as well as fibro-inflammatory changes and COX-2 expression seem most valuable as predictors for survival.

Supervised feature-based classification of multi-channel SAR images

This paper describes a new method for a feature-based supervised classification of multi-channel SAR data. Classic feature selection and classification methods are inadequate due to the diverse statistical distributions of the input features. A method based on logistic regression (LR) and multinomial logistic regression (MNLR) for separating different classes is therefore proposed. Both methods, LR and MNLR, are less dependent on the statistical distribution of the input data. A new spatial regularization method is also introduced to increase consistency of the classification result. The classification method was applied to a project on humanitarian demining in which the relevant classes were defined by experts of a mine action center. A ground survey mission collected learning and validation samples for each class. Results of the proposed classification methods are shown and compared to a maximum likelihood classifier.

Comparative evaluation of hyperspectral anomaly detectors in different types of background

Anomaly detection (AD) in hyperspectral data has received a lot of attention for various applications. The aim of anomaly detection is to detect pixels in the hyperspectral data cube whose spectra differ significantly from the background spectra. Many anomaly detectors have been proposed in literature. They differ by the way the background is characterized and by the method used for determining the difference between the current pixel and the background. The most well-known anomaly detector is the RX detector that calculates the Mahalanobis distance between the pixel under test (PUT) and the background. Global RX characterizes the background of the complete scene by a single multi-variate normal distribution. In many cases this model is not appropriate for describing the background. For that reason a variety of other anomaly detection methods have been developed. This paper examines three classes of anomaly detectors: sub-space methods, local methods and segmentation-based methods. Representative examples of each class are chosen and applied on a set of hyperspectral data with different backgrounds. The results are evaluated and compared.

Edge and line detection in polarimetric SAR images

A scheme for detecting edges and lines in multichannel SAR images is proposed. The line detector is constructed from the edge detector. The latter is based on multivariate statistical hypothesis tests applied to log-intensity SAR images. The raw results are vectorized by a traditional bright line extraction process. The scheme is illustrated by extracting dark linear structures on various full-polarimetric SAR images.

Multilevel data fusion for the detection of targets using multispectral image sequences

An approach is presented to the long range automatic detec- tion of vehicles, using multisensor image sequences. The method is tested on a database of multispectral image sequences, acquired under diverse operational conditions. The approach consists of two parts. The first part uses a semisupervised approach, based on texture parameters, for detecting stationary targets. For each type of sensor one learning image is chosen. Texture parameters are calculated at each pixel of the learning images and are combined using logistic regression into a value that represents the conditional probability that the pixel belongs to a target given the texture parameters. The actual detection algorithm ap- plies the same combination to the texture features calculated on the remainder of the database (test images). When the results of this feature-level fusion are stored as an image, the local maxima correspond to likely target positions. These feature-level-fused images are calcu- lated for each sensor. In a sensor fusion step, the results obtained per sensor are then combined again. Region growing around the local maxima is then used to detect the targets. The second part of the algo- rithm searches for moving targets. To detect moving vehicles, any mo- tion of the sensor must be detected first. If sensor motion is detected, it is estimated using a Markov random field model. Available prior knowl- edge about the sensor motion is used to simplify the motion estimation. The estimate is used to warp past images onto the current image in a temporal fusion approach and moving targets are detected by threshold- ing the difference between the original and warped images. Decision level fusion combines the results from both parts of the algorithm.

Hyperspectral anomaly detection: comparative evaluation in scenes with diverse complexity

Anomaly detection (AD) in hyperspectral data has received a lot of attention for various applications. The aim of anomaly detection is to detect pixels in the hyperspectral data cube whose spectra differ significantly from the background spectra. Many anomaly detectors have been proposed in the literature. They differ in the way the background is characterized and in the method used for determining the difference between the current pixel and the background. The most well-known anomaly detector is the RX detector that calculates the Mahalanobis distance between the pixel under test (PUT) and the background. Global RX characterizes the background of the complete scene by a singlemultivariate normal probability density function. Inmany cases, this model is not appropriate for describing the background. For that reason a variety of other anomaly detection methods have been developed. This paper examines three classes of anomaly detectors: subspace methods, local methods, and segmentation-based methods. Representative examples of each class are chosen and applied on a set of hyperspectral data with diverse complexity. The results are evaluated and compared.