Predicting overall survival in early stage lung cancer with feature driven local cell graphs (FEDEG)

An apparatus for predicting overall survival in lung cancer, the apparatus comprising:

• a processor;
• a memory configured to store a digitized image of a region of tissue (ROT) demonstrating cancerous pathology, the digitized image having a plurality of pixels, a pixel having an intensity;
• an input/output (I/O) interface;
• a set of circuits;
• and
• an interface that connects the processor, the memory, the I/O interface, and the set of circuits, the set of circuits comprising:
• an image acquisition circuit configured to:
• access the digitized image of an ROT demonstrating cancerous pathology, the digitized image having the plurality of pixels, the pixel having the intensity;
• and
• detect a plurality of cells represented in the digitized image, where a member of the plurality of cells includes a cellular nucleus;
• a nuclear morphology circuit configured to:
• for each member of the plurality of cells:
• segment a cellular nucleus represented in the digitized image, respectively;
• and
• compute a set of nuclear morphology features based on the segmented cellular nucleus, where the set of nuclear morphology features includes at least six features;
• a feature driven local cell graph (FeDeG) circuit configured to:
• construct a FeDeG based on the segmented cellular nuclei and the set of nuclear morphology features using a mean-shift clustering approach;
• and
• compute a set of FeDeG features based on the FeDeG;
• a classification circuit configured to:
• generate a classification of the ROT as a long-term survivor or short-term survivor based on the set of FeDeG features and the FeDeG;
• a display circuit configured to display the classification.

A non-transitory computer-readable storage device storing computer-executable instructions that when executed control a computer to perform a method of predicting overall survival in cancerous pathology, the method comprising:

• accessing a digitized image of a region of tissue (ROT) demonstrating non-small cell lung cancer (NSCLC), the ROT including a plurality of cellular nuclei, the digitized image having a plurality of pixels, a pixel having an intensity;
• extracting a set of nuclear morphology features from a member of the plurality of cellular nuclei, the set of nuclear morphology features including at least six features;
• constructing a feature driven local cell graph (FeDeG) based on the plurality of cellular nuclei and the set of nuclear morphology features associated with each member of the plurality of cellular nuclei, respectively, using a mean-shift clustering approach;
• computing a set of FeDeG features based on the FeDeG, where the set of FeDeG features includes at least one intersection between different FeDeGs feature, at least one size of FeDeG feature, at least one disorder of nuclear morphology feature, and at least one architectural measures of FeDeGs feature;
• generating a classification of the ROT as a short-term survivor or long-term survivor based on a linear discriminant analysis of the set of FeDeG features;
• and
• displaying the classification.