Arkadiusz Gertych

Automatic bone age assessment for young children from newborn to 7-year-old using carpal bones.

A computer-aided-diagnosis (CAD) method has been previously developed based on features extracted from phalangeal regions of interest (ROI) in a digital hand atlas, which can assess bone age of children from ages 7 to 18 accurately. Therefore, in order to assess the bone age of children in younger ages, the inclusion of carpal bones is necessary. However, due to various factors including the uncertain number of bones appearing, non-uniformity of soft tissue, low contrast between the bony structure and soft tissue, automatic segmentation and identification of carpal bone boundaries is an extremely challenging task. Past research works on carpal bone segmentation were performed utilizing dynamic thresholding. However, due to the limitation of the segmentation algorithm, carpal bones have not been taken into consideration in the bone age assessment procedure. In this paper, we developed and implemented a knowledge-based method for fully automatic carpal bone segmentation and morphological feature analysis. Fuzzy classification was then used to assess the bone age based on the selected features. This method has been successfully applied on all cases in which carpal bones have not overlapped. CAD results of total about 205 cases from the digital hand atlas were evaluated against subject chronological age as well as readings of two radiologists. It was found that the carpal ROI provides reliable information in determining the bone age for young children from newborn to 7-year-old.

Integration of computer assisted bone age assessment with clinical PACS

Computer assisted bone age assessment (BAA) integrated with a clinical PACS is described. The image analysis is performed on a DICOM compliant workstation able to accept images from a PACS server or directly from an image modality (digital radiography or film scanner). Images can be processed in two modes. If the image is acquired from a normally developed subject, it can be added to the digital hand atlas. An image may also be subjected only to a diagnostic analysis for the BAA without archiving the features in the database. The image analysis is performed in three steps. A location of six region of interest is followed by their segmentation and feature extraction. The features analysis results in retrieving the closest image match from the standard database. Based on currently analyzed image data in the hand atlas, the standard deviation of the assessment bone age does not exceed 1 yr of age.

Computer-Assisted Bone Age Assessment: Graphical User Interface for Image Processing and Comparison

The current study is part of a project resulting in a computer-assisted analysis of a hand radiograph yielding an assessment of skeletal maturity. The image analysis is based on features selected from six regions of interest. At various stages of skeletal development different image processing problems have to be addressed. At the early stage, feature extraction is based on Lee filtering followed by the random Gibbs fields and mathematical morphology. Once the fusion starts, wavelet decomposition methods are implemented. The user interface displays the closest neighbors to each image under consideration. Results show the sensitivity of different regions to both stages of development and certain feature sensitivity within each region. At the early stage of development, the distal features are more reliable indicators, whereas at the stage of epiphyseal fusion, a larger dynamic range of middle features makes them more sensitive. In the current study, a graphical user interface has been designed and implemented for testing the image processing routines and comparing the results of quantitative image analysis with the visual interpretation of extracted regions of interest. The user interface may also serve as a teaching tool. At the later stage of the project it will be used as a classification tool.

Segmentation of regions of interest and post-segmentation edge location improvement in computer-aided bone age assessment

Segmentation of anatomical structures in radiological images is one of the important steps in the computerized approach to the bone age assessment. In this paper a method dealing with correct location of the borders in the epi-metaphyseal regions of interest is described. The well segmented bone structures are obtained utilizing the Gibbs random fields as the first segmentation step; however this method does not prove to be adequate in the correct outline of other tissues in the epi-metaphyseal area. In order to correct delineation of cartilage in this region, the second segmentation step utilizing the active contours serving as a post-segmentation edge location technique is applied. Controlling of tension and bending of the active contour requires a set of weights in the energy functional to be set. To adjust the weights and to initially test the methodology a model of region of interest containing three different anatomical structures corrupted with Gaussian noise has been designed. Combined methodology of Gibbs random fields and active contours with the final set of weights was applied to 200 regions of interest randomly selected from 1100 left hand radiographs. A meaningful improvement in terms of ultimate contour location and smoothing has been observed in regions with cartilage or bone convexity developed near the bottom region of the epiphysis.

Image database for digital hand atlas

Bone age assessment is a procedure frequently performed in pediatric patients to evaluate their growth disorder. A commonly used method is atlas matching by a visual comparison of a hand radiograph with a small reference set of old Greulich-Pyle atlas. We have developed a new digital hand atlas with a large set of clinically normal hand images of diverse ethnic groups. In this paper, we will present our system design and implementation of the digital atlas database to support the computer-aided atlas matching for bone age assessment. The system consists of a hand atlas image database, a computer-aided diagnostic (CAD) software module for image processing and atlas matching, and a Web user interface. Users can use a Web browser to push DICOM images, directly or indirectly from PACS, to the CAD server for a bone age assessment. Quantitative features on the examined image, which reflect the skeletal maturity, are then extracted and compared with patterns from the atlas image database to assess the bone age. The digital atlas method built on a large image database and current Internet technology provides an alternative to supplement or replace the traditional one for a quantitative, accurate and cost-effective assessment of bone age.

A Machine Learning Approach to Identify Prostate Cancer Areas in Complex Histological Images

Separating benign glands, and cancer areas from stroma is one of the vital steps towards automated grading of prostate cancer in digital images of H&E preparations. In this work we present a novel tool that utilizes a supervised classification of histograms of staining components in hematoxylin and eosin images to delineate areas of benign and cancer glands. Using high resolution images of whole slide prostatectomies we compared several image classification schemes which included intensity histograms, histograms of oriented gradients, and their concatenations to the manual annotations of tissues by a pathologist, and showed that joint intensity histograms of hematoxylin and eosin components performed with the highest accuracy.

Bone age assessment for young children from newborn to 7-year-old using carpal bones

A computer-aided-diagnosis (CAD) method has been previously developed based on features extracted from phalangeal regions of interest (ROI) in a digital hand atlas, which can assess bone age of children from ages 7 to 18 accurately. Therefore, in order to assess the bone age of children in younger ages, the inclusion of carpal bones is necessary. In this paper, we developed and implemented a knowledge-based method for fully automatic carpal bone segmentation and morphological feature analysis. Fuzzy classification was then used to assess the bone age based on the selected features. Last year, we presented carpal bone segmentation algorithm. This year, research works on procedures after carpal bone segmentation including carpal bone identification, feature analysis and fuzzy system for bone age assessment is presented. This method has been successfully applied on all cases in which carpal bones have not overlapped. CAD results of total about 205 cases from the digital hand atlas were evaluated against subject chronological age as well as readings of two radiologists. It was found that the carpal ROI provides reliable information in determining the bone age for young children from newborn to 7-year-old.

Carpal bone analysis in bone age assessment

A computer-aided-diagnosis (CAD) method has been previously developed in our Laboratory based on features extracted from regions of interest (ROI) in phalanges in a digital hand atlas. Due to various factors, including, the diversity of size, shape and orientation of carpal bones, non-uniformity of soft tissue, low contrast between the bony structure and soft tissue, the automatic identification and segmentation of bone boundaries is an extremely challenging task. Past research work on carpal bone segmentation has been done utilizing dynamic thresholding. However, due to the discrepancy of carpal bones developments and the limitations of segmentation algorithms, carpal bone ROI has not been taken into consideration in the bone age assessment procedure. In this paper, we present a method for fully automatic carpal bone segmentation and feature analysis in hand X-ray radiograph. The purpose of this paper is to automatically segment the carpal bones by anisotropic diffusion and Canny edge detection techniques. By adding their respective features extracted from carpal bones ROI to the phalangeal ROI feature space, the accuracy of bone age assessment can be improved especially when the image processing in the phalangeal ROI fails in younger children.

Active Contour Technique in Post-segmentation Edge Smoothing Applied to Hand Radiograph Regions of Interest

In the current study two various segmentation methods have been implemented sequentially in computer-aided approach to the assessment of skeletal maturity, where correct location of borders of anatomical structures is a crucial step. The first segmentation stage, based on the Gibbs random fields technique, correctly segments out a bony structure and roughly outlines the edges of cartilage while the second one using the active contour strategy, smoothes them and prepares for the feature extraction stage. A synthetic region of interest has been designed to test and adjust weights of the snake energy functional. These weights have afterwards been applied to a real image data. In comparison to our previous works we observe a significant improvement of boundary location mainly when cartilage is included in the epiphyses.

Spectral Classification of Dual Nuclear p16/Ki67 Positivity in Pap Smears

An automated detection and quantification of high-risk human papillomavirus immunostains in Pap smears can potentially improve the drawbacks of human observes, reduce the number of equivocal or misinterpreted cytologic specimens and increase the throughput of slide screening. Towards increasing the accuracy and efficacy of Pap smear screening, we tested a spectral imaging approach to quantify the dual p16/Ki67 immunoreactivity of epithelial cell nuclei in Pap smears. We demonstrated that the classification of spectral signatures extracted from nuclear pixels is helpful in detecting nuclei of cells that are positive for p16 and Ki67 and distinguishing them from other nuclei that are positive only for one or negative for both markers. Sensitivity of the proposed method was 84.4% whereas the specificity was 99.9%. Although our results are preliminary, they suggest that the implementation of spectral imaging and spectral classifications can potentially offer better nuclei screening performances than methods utilizing conventional RGB imaging.