Arch Development Corporation

The University of Chicago

Shigehiko Katsuragawa

Kunio Doi

Qiang Li

Takayuki Ishida

Junji Morishita

Xin-Wei Xu

Hiroyuki Yoshida

Hidetaka Arimura

Yoshikazu Uchiyama

Xuan Chen

Yali Amit

An automated computerized scheme for detection and characterization of diffuse lung diseases on high-resolution computed tomography (HRCT) images including obtaining image data including pixels of an organ; segmenting the image data into organ image data and non-organ image data; extracting predetermined features from the organ image data to produce a set of image features; comparing the set of image features against a reference set of organ image features containing image data known to correspond to normal and abnormal conditions; and producing a comparison result.

System and method for computer-aided detection and characterization of diffuse lung disease

A method, system and computer readable medium for computerized processing of chest images including obtaining a digital first image of a chest (S); producing a second image which is a mirror image (S) of the first image; performing image warping on one of the first and second images to produce a warped image (S) which is registered to the other of the first and second images; and subtracting the warped image from the other image to generate a subtraction image (S). Another embodiment includes obtaining a digital first image of the chest of a subject; detecting ribcage edges on both sides of the lungs in the first chest image; determining average horizontal locations of the left and right ribcage edges at plural vertical locations; fitting the determined average horizontal locations to a straight line to derive a midline; rotating the chest image so that the midline is vertical; and shifting the rotated image to produce a lateral inclination corrected (S) second image with the midline centered in the lateral inclination corrected image.

System for computerized processing of chest radiographic images

A method, system and computer readable medium for a computer-automated method for identifying given image data, including obtaining template image data corresponding to said given image data; calculating correlation values between the given image data and said template image data; and identifying said image data based on the correlation values calculated in the calculating step.

Method, system and computer readable medium for identifying chest radiographs using image mapping and template matching techniques

A method to determine whether a candidate abnormality in a medical digital image is an actual abnormality, a system which implements the method, and a computer readable medium which stores program steps to implement the method, wherein the method includes obtaining a medical digital image including a candidate abnormality; obtaining plural first templates and plural second templates respectively corresponding to predetermined abnormalities and predetermined non-abnormalities; comparing the candidate abnormality with the obtained first and second templates to derive cross-correlation values between the candidate abnormality and each of the obtained first and second templates; determining the largest cross-correlation value derived in the comparing step and whether the largest cross-correlation value is produced by comparing the candidate abnormality with the first templates or with the second templates; and determining the candidate abnormality to be an actual abnormality when the largest cross-correlation value is produced by comparing the candidate abnormality with the first templates and determining the candidate abnormality to be a non-abnormality when the largest cross-correlation value is produced by comparing the candidate abnormality with the second templates. An actual abnormality is similarly classified as malignant or benign based on further cross-correlation values obtained by comparisons with additional templates corresponding to malignant and benign abnormalities.

Process, system and computer readable medium for pulmonary nodule detection using multiple-templates matching

A method, system, computer readable medium and apparatus for computerized detection of lung nodules in computer tomography images, by which mask images are created such that subtractions of the mask image from a targeted CT section image reveal or highlight small lung nodules in the target CT section. The mask image is created utilizing the targeted CT section image along with other CT section images generated from the same CT scan. Based on these other section CT images and the targeted CT section image, a mask image can be created that is very similar to the target CT section image, but without the presence of small lung nodules. When the mask image is subtracted from the targeted CT section image, the differences between the mask images and the CT section images reveal small lung nodules. The mask image may be created by linear interpolation or a morphological filtered image.

Subtraction technique for computerized detection of small lung nodules in computer tomography images

A method, system and computer readable medium of computerized processing of chest images including obtaining digital first and second images of a chest and detecting rib edges in at least one of the first and second images. The rib edges are detected by correlating points in the at least one of the first and second images to plural rib edge models using a Hough transform to identify approximate rib edges in one of the images, and delineating actual rib edges derived from the identified approximate rib edges using a snake model. The method system and computer readable medium further include deriving the shift values using the actual rib edges and warping one of the first and second images to produce a warped image which is registered to the other of the first and second images based at least in part on the shift values.

Method, system and computer readable medium for computerized processing of contralateral and temporal subtraction images using elastic matching

A method and apparatus for discrimination of nodules and false positives in digital chest radiographs, using a wavelet snake technique. The wavelet snake is a deformable contour designed to identify the boundary of a relatively round object. The shape of the snake is determined by a set of wavelet coefficients in a certain range of scales. Portions of the boundary of a nodule are first extracted using a multiscale edge representation. The multiscale edges are then fitted by a gradient descent procedure which deforms the shape of a wavelet snake by changing its wavelet coefficients. The degree of overlap between the fitted snake and the multiscale edges is calculated and used as a fit quality indicator for discrimination of nodules and false detections.

Method, apparatus, and storage medium for detection of nodules in

A method of computerized analysis of temporally sequential digital images, including (a) determining first shift values between pixels of a first digital image and corresponding pixels of a second digital image; (b) warping the second digital image based on the first shift values to obtain a first warped image in which spatial locations of pixels are varied in relation to the first shift values; (c) determining second shift values between pixels of the first digital image and pixels of the first warped image; and (d) warping the first warped image based on the second shift values to obtain a second warped image in which spatial locations of pixels of the first warped image are varied in relation to the second shift values. Additional iterations of image warping are possible to enhance image registration between the first digital image and the warped version of the second digital image, followed by image subtraction of the first digital image and the final warped image to produce a difference image from which diagnosis of temporal changes ensues. Temporal subtraction assists radiologists in the detection of interval changes on chest radiographs, and particularly to overcome severe misregistration errors in the temporally sequential images mainly due to differences in a subject's inclination and/or rotation. In the production of shift values used in image warping, initial shift values are obtained by cross-correlation techniques using a template and search regions of interest. Shift vectors and a histogram of shift vectors in each lung are obtained from initial shift values. The histograms of shift vectors are used in the selection of sets of shift values for smoothing, using two-dimensional fitting and subsequent use of fitted shift values in image warping.

Method, system and computer readable medium for iterative image warping

A computerized method for the detection and characterization of disease in an image derived from a chest radiograph, wherein an image in the chest radiograph is processed to determine the ribcage boundary, including lung top edges, right and left ribcage edges, and right and left hemidiaphragm edges. Texture measures including RMS variations of pixel values within regions of interest are converted to relative exposures and corrected for system noise existing in the system used to produce the image. Texture and/or geometric pattern indices are produced. A histogram(s) of the produced index (indices) is produced and values of the histograms) are applied as inputs to a trained artificial neural network, which classifies the image as normal or abnormal. In one embodiment, obviously normal and obviously abnormal images are determined based on the ratio of abnormal regions of interest to the total number of regions of interest in a rule-based method, so that only difficult cases to diagnose are applied to the artificial neural network.

Computer-aided method for automated image feature analysis and diagnosis

A method and computerized automated initial image matching technique for enhancing detection of interval changes between temporally subsequent radiographic images via image subtraction. The method includes the steps of digitizing images, normalizing density and contrast in the digital images, correcting for lateral inclination in the digital images, detecting edges of a same feature in each image, converting the images into low resolution matrices, blurring the low resolution images, segmenting portions of the blurred low resolution matrices based on the detected edges, matching the digital images based on a cross-correlation match between the segmented portions, performing non-linear warping to further match Regions of Interest (ROI), and performing image subtraction between the matched digital images. The low resolution matrices are greater than 64.times.64 in size and are produced by averaging. Blurring of the low resolution matrices is performed via a Gaussian filter that removes fine structures in each image such as small vessels, bronchia, etc. The method may be performed by a computer system according to instructions stored on a computer readable medium.

Method of detecting interval changes in chest radiographs utilizing

A computerized method for the detection and characterization of disease in an image derived from a chest radiograph, wherein an image in the chest radiograph is processed to determine the ribcage boundary, including lung top edges, right and left ribcage edges, and right and left hemidiaphragm edges. Texture measures including RMS variations of pixel values within regions of interest are converted to relative exposures and corrected for system noise existing in the system used to produce the image. Texture and/or geometric pattern indices are produced. A histogram(s) of the produced index (indices) is produced and values of the histogram(s) are applied as inputs to a trained artificial neural network, which classifies the image as normal or abnormal. In one embodiment, obviously normal and obviously abnormal images are determined based on the ratio of abnormal regions of interest to the total number of regions of interest in a rule-based method, so that only difficult cases to diagnose are applied to the artificial neural network.

An automated method and system for discriminating between normal lungs and abnormal lungs having interstitial disease and/or septal lines, wherein a large number of adjacent regions of interest (ROIs) are selected, corresponding to an area on a digital image of a patient's lungs. The ROIs each contain a number of square or rectangular pixel arrays and are selected to sequentially fill in the total selected area of the lungs to be analyzed. A background trend is removed from each individual ROI and the ROIs are then analyzed to determine those exhibiting sharp edges, i.e., high edge gradients. A percentage of these sharp-edged ROIs are removed from the original sample based on the edge gradient analysis, a majority of which correspond to rib-edge containing ROIs. After removal of the sharp-edged ROIs, texture measurements are taken on the remaining sample in order to compare such data with predetermined data for normal and abnormal lungs. Thus, a computerized scheme for quantitative analysis of interstitial lung diseases and/or septal lines appearing in digitized chest radiographs can be implemented in practical clinical situations.

Method and system for automated selection of regions of interest and

A computerized method and system based on quantitative analysis of geometric features of various infiltrate patterns in chest images for the detection and categorization of abnormalities related to the infiltrate patterns. Chest images are digitized and a lung texture analysis is performed on a number of small regions of interest (ROIs) in order to determine a classification of normal or abnormal of the particular patient's lungs. If the lungs are determined as being abnormal, large ROIs with a 128.times.128 matrix are selected in order to cover the detected areas of abnormality. Overall background trend correction is then performed in these large ROIs using a 2D-surface fitting technique for isolation of the fluctuating patterns of the underlying lung texture. Opacities of interstitial infiltrates are identified from two processed images which are obtained by employing thresholding with a morphological filter and a line enhancement filter. Finally, ROIs are classified into nodular, reticular and/or reticulo-nodular patterns by measurement of parameters corresponding to the type of the abnormality pattern detected.

Method and system for determining geometric pattern features of

A method and system for automated analysis of digital radiographic images in which regions-of-interest (ROI's) are first determined, and subsequently analyzed for abnormalities. To locate the ROI's, the outer ribcage and midline boundary locations of the chest image are determined from the digital image data. Vertical profiles are then selected and background trend is then removed from each vertical profile. A shift-variant sinusoidal function is fitted to each vertical profile and ROI's are selected based on the fitted functions. The non-uniform background trend is removed from the original image data of each selected ROI to obtain corrected data. The power spectrum of the lung texture is obtained from the 2D Fourier transform of the corrected data and is filtered by the human visual response. Finally, the root-mean-square (rms) variation, R, and the first moment of the power spectrum, M, are determined as quantitative texture measures for the magnitude and coarseness (or fineness), respectively, of the lung texture.

Growing community of inventors

Oita, Japan

Shigehiko Katsuragawa