Abstract An automated method and system for digital imaging processing of radiologic images, wherein digital image data is acquired and subjected to multiple phases of digital imaging processing. During the Pre-Processing stage, simultaneous box-rim filtering and k-nearest neighbor processing and subsequent global thresholding are performed on the image data to enhance object-to-background contrast, merge subclusters and determine gray scale thresholds for further processing. Next, during the Preliminary Selection phase, body part segmentation, morphological erosion processing, connected component analysis and image block segmentation occurs to subtract unwanted image data preliminarily identify potentials areas including abnormalities. During the Pattern Classification phase, feature patterns are developed for each area of interest, a supervised, back propagation neural network is trained, a feed forward neural network is developed and employed to detect true and several false positive categories, and two types of pruned neural networks are utilized in connection with a heuristic decision tree to finally determine whether the regions of interest are abnormalities or false positives.
An automated method for digital image processing of a radiographic image comprising the steps of: providing a digital radiographic image of a body part having microcalcifications; independently, but simultaneously, box-rim filtering and K-nearest neighbor processing the digital radiographic image; global thresholding the image resulting from the prior step; body part segmenting the globally thresholded image; morphological erosion processing of the body part segmented image; connected component analysis of the processed image to determine microcalcification candidates; and segmenting the provided image into image blocks centered at the microcalcification candidates.
Claim 2
Dependent Claim
88 words
An automated method for digital image processing of a radiographic image comprising the steps of: providing a digital radiographic image of a body part having microcalcifications; independently, but simultaneously, box-rim filtering and K-nearest neighbor processing the digital radiographic image; global thresholding the image resulting from the prior step; body part segmenting the globally thresholded image; morphological erosion processing of the body part segmented image; connected component analysis of the processed image to determine microcalcification candidates; and segmenting the provided image into image blocks centered at the microcalcification candidates.
Claim 3
Dependent Claim
23 words
The method of claim 1 including the step of processing each image block with feature map generation using sets of radially sampled pixels.
Claim 4
Dependent Claim
40 words
The method of claim 2 including the step of using a neural network with five output nodes specifying five distinct classes: (1) microcalcifications, (2) bright solitary dots, (3) medium solitary dots, (4) bright fat tissue, and (5) medium fat tissue.
Claim 5
Dependent Claim
20 words
The method of claim 3 including the step of using a heuristic decision rule to facilitate classification of true microcalcifications.
Claim 6
Dependent Claim
21 words
The method of claim 4 including the step of using an ANN decision tree in conjunction with said heuristic decision rule.
Claim 7
Dependent Claim
17 words
The method of claim 5 wherein said steps are implemented by two separate pruned neural network architectures.
Claim 8
Dependent Claim
21 words
The method of claim 6 wherein said two neural network architectures are a semi-trainable ANN implementation and a fully-trainable ANN implementation.