László Czúni

Technical note on the validation of a semi-automated image analysis software application for estrogen and progesterone receptor detection in breast cancer

BackgroundThe immunohistochemical detection of estrogen (ER) and progesterone (PR) receptors in breast cancer is routinely used for prognostic and predictive testing. Whole slide digitalization supported by dedicated software tools allows quantization of the image objects (e.g. cell membrane, nuclei) and an unbiased analysis of immunostaining results. Validation studies of image analysis applications for the detection of ER and PR in breast cancer specimens provided strong concordance between the pathologists manual assessment of slides and scoring performed using different software applications.MethodsThe effectiveness of two connected semi-automated image analysis software (NuclearQuant v. 1.13 application for Pannoramic™ Viewer v. 1.14) for determination of ER and PR status in formalin-fixed paraffin embedded breast cancer specimens immunostained with the automated Leica Bond Max system was studied. First the detection algorithm was calibrated to the scores provided an independent assessors (pathologist), using selected areas from 38 small digital slides (created from 16 cases) containing a mean number of 195 cells. Each cell was manually marked and scored according to the Allred-system combining frequency and intensity scores. The performance of the calibrated algorithm was tested on 16 cases (14 invasive ductal carcinoma, 2 invasive lobular carcinoma) against the pathologists manual scoring of digital slides.ResultsThe detection was calibrated to 87 percent object detection agreement and almost perfect Total Score agreement (Cohens kappa 0.859, quadratic weighted kappa 0.986) from slight or moderate agreement at the start of the study, using the un-calibrated algorithm. The performance of the application was tested against the pathologists manual scoring of digital slides on 53 regions of interest of 16 ER and PR slides covering all positivity ranges, and the quadratic weighted kappa provided almost perfect agreement (κ = 0.981) among the two scoring schemes.ConclusionsNuclearQuant v. 1.13 application for Pannoramic™ Viewer v. 1.14 software application proved to be a reliable image analysis tool for pathologists testing ER and PR status in breast cancer.

Estimating the Optimal Quantization Parameter in H.264

The quantization parameter (QP) has a very important impact on the compression rate in H.264. In this paper we show that in order to achieve efficient rate-control coding a good estimate for the initial QP parameter is necessary. An extensive altering of this value, to keep the required titrate, results in significant fluctuation of the image quality and decreases the average quality of the whole coded sequence. We propose a simple and fast method to decide the starting value for QP for each part of the video sequence. Experimental results show stabilized image quality and significant gain in PSNR

Detection of unusual optical flow patterns by multilevel hidden Markov models

The analysis of motion information is one of the main tools for the understanding of complex behaviors in video. However, due to the quality of the optical flow of low-cost surveillance camera systems and the complexity of motion, new robust image-processing methods are required to generate reliable higher-level information. In our novel approach there is no need for tracking objects (vehicles, pedestrians) in order to recognize anomalous motion, but dense optical flow information is used to construct mixtures of Gaussians, which are analyzed temporally. We create a multilevel model, where low-level states of non-overlapping image regions are modeled by continuous hidden Markov models (HMMs). From low-level HMMs we compose high-level HMMs to analyze the occurrence of the low-level states. The processing of large numbers of data in traditional HMMs can result in a precision problem due to the multiplication of low probability values. Thus, besides introducing new motion models, we incorporate a scaling technique into the mathematical model of HMMs to avoid precision problems and to get an effective tool for the analysis of large numbers of motion vectors. We illustrate the use of our models with real-life traffic videos.

Object Oriented Motion-Segmentation for Video-Compression in the CNN-UM

Object-oriented motion segmentation is a basic step of the effective coding of image-series. Following the MPEG-4 standard we should define such objects. In this paper, a fully parallel and locally connected computation model is described for segmenting frames of image sequences based on spatial and motion information. The first type of the algorithm is called early segmentation. It is based on spatial information only and aims at providing an over-segmentation of the frame in real-time. Even if the obtained results do not minimize the number of regions, it is a good starting point for higher level post processing, when the decision on how to regroup regions in object can rely on both spatial and temporal information. In the second type of the algorithm stochastic optimization methods are used to form homogenous dense optical vector fields which act directly on motion vectors instead of 2D or 3D motion parameters. This makes the algorithm simple and less time consuming than many other relaxation methods. Then we apply morphological operators to handle disocclusion effects and to map the motion field to the spatial content. Computer simulations of the CNN architecture demonstrate the usefulness of our methods. All solutions in our approach suggest a fully parallel implementation in a newly developed CNN-UM VLSI chip architecture.

Adaptive Stabilization of Vibration on Archive Films

Image vibration is a typical type of degradation that is difficult to restore in an automatic film restoration system. It is usually caused by improper film transportation during the copying or the digitization process. We have developed a method for automatic image stabilization consisting of two main steps: estimating vibration then correction by drifting the whole frame. Earlier stabilization algorithms are unsuccessful in cases of multiple motions and human interaction is necessary to achieve satisfactory results. The proposed technique is automatic and avoids false results for most difficult situations as shown in examples. The paper describes the technique used for motion estimation of regions; the selection of image regions for adequate vibration estimation; and the method of stabilization.

Trainable blotch detection on high resolution archive films minimizing the human interaction

Film archives are continuously in need of automatic restoration tools to accelerate the correction of film artifacts and to decrease the costs. Blotches are a common type of film degradation and their correction needs a lot of manual interaction in traditional systems due to high false detection rates and the huge amount of data of high resolution images. Blotch detectors need reliable motion estimation to avoid the false detection of uncorrupted regions. In case of erroneous detection, usually an operator has to remove the false alarms manually, which significantly decreases the efficiency of the restoration process. To reduce manual intervention, we developed a two-step false alarm reduction technique including pixel- and object-based methods as post-processing. The proposed pixel-based algorithm compensates motion, decreasing false alarms at low computational cost, while the following object based method further reduces the residual false alarms by machine learning techniques. We introduced a new quality metric for detection methods by measuring the required amount of manual work after the automatic detection. In our novel evaluation technique, the ground truth is collected from digitized archive sequences where defective pixel positions are detected in an interactive process.

Depth-Based Indexing and Retrieval of Photographic Images

This paper proposes a new technique for image capture, indexing, and retrieval to implement a content-based image retrieval (CBIR) system more similar to the way people remember the real world [2]. The introduced technique uses range from focus technique to gather 3D information of a scene. The obtained depth-map is segmented and stored together with each individual image in database files. During retrieval the user can describe the query image not only in a conventional way but also with a layered representation where a few (typically 3) depth layers define the distance from the camera. This paper describes the beginning of our research with some preliminary results showing that depth information can be efficiently used in CBIR systems.

Content-based image retrieval using stochastic paintbrush transformation

We propose a new content based image retrieval method. The novelty of our approach lies in the applied image similarity measure: unlike traditional features, such as color, texture or shape, our measure is based on a painted representation of the original image. We use paintbrush stroke parameters as features. These strokes are produced by a stochastic paintbrush algorithm which simulates a painting process. Stroke parameters include color, orientation and location. Therefore, it provides information not only about the color content but also about the structural properties of an image. Experimental results on a database of more than 500 images show that the CBIR method using paintbrush features has a higher retrieval rate than methods using color features only.

HMM-based unusual motion detection without tracking

We propose novel pixel dense modeling of motion of urban traffic in noisy environments with the help of multidimensional Gaussian Mixture Models (GMMs) and Hidden Markov Models (HMMs). In our approach there is no need for object tracking in order to detect anomalous motion or to model and visualize the fluctuation of traffic. We propose a new scaling method introduced into the HMM to get a robust tool for the analysis of hundreds of motion vector samples at a time. We show the use of our model with a photorealistic video synthetized from real life recordings.

Trainable post-processing method to reduce false alarms in the detection of small blotches of archive films

We have developed a new semi-automatic neural network based method to detect blotches with low false alarm rate on archive films. Blotches can be modeled as temporal intensity discontinuities, hence false detection results originate from object motion (e.g. occlusion), non-rigid objects or erroneous motion estimation. In practice, usually, after the automatic detection step the false alarms are removed manually by an operator, significantly decreasing the efficiency of the restoration process. Our post-processing method classifies each detected blotch by its image features to minimize false results and the necessity of human intervention. The proposed method is tested on real archive sequences.