System, method, and article for detecting abnormal cells using multi-dimensional analysis

A device, comprising:

• memory;
• and
• processing circuitry coupled to the memory, wherein the processing circuitry, in operation and using the memory:
• defines a set of normal cell clusters in an n-dimensional space using a centroid line and radius, the radius originating on the line and a length of the radius being a function of a distance along the centroid line, the distance along the centroid line corresponding to a maturation level with a cell lineage, where n is greater than or equal to six;
• compares an indication of a distribution of biological cells of a test set of biological cells along the centroid line in the n-dimensional space to an indication of a distribution of biological cells of the defined set of normal cell clusters along the centroid line in the n-dimensional space, the comparing including:
• generating a histogram representing the indication of the distribution of biological cells of the defined set of normal cell clusters in a two-dimensional display having a first axis corresponding to a cell maturation level within a cell lineage and a second axis corresponding to a frequency of occurrence, wherein the histogram comprises a band corresponding to the centroid line and radius used to define the set of normal cell clusters;
• and
• comparing frequencies of occurrence of biological cells along the centroid line in the set of normal cell clusters to frequencies of occurrence of biological cells along the centroid line in the test set of biological cells using the histogram;
• characterizing the test set of biological cells based on the comparison of the indication of the distribution of biological cells of a test set of biological cells and the indication of the distribution of biological cells of the defined set of normal cell clusters in the histogram;
• and
• diagnosing cancer in the test set of biological cells based on the characterization of the test set of biological cells.

A non-transitory computer-readable medium having contents which configure a processing device to perform a method, the method comprising:

• (a) defining a set of normal cell clusters in an n-dimensional space using a centroid line and radius, the radius originating on the line and a length of the radius being a function of a distance along the centroid line, the distance along the centroid line corresponding to a maturation level with a cell lineage, where n is greater than or equal to six;
• (b) comparing an indication of a distribution of biological cells of a test set of biological cells along the centroid line in the n-dimensional space to an indication of a distribution of biological cells of the defined set of normal cell clusters along the centroid line in the n-dimensional space, the comparing including:
• (i) generating a histogram representing the indication of the distribution of biological cells of the defined set of normal cell clusters in a two-dimensional display having a first axis corresponding to a cell maturation level within a cell lineage and a second axis corresponding to a frequency of occurrence, wherein the histogram comprises a band corresponding to the centroid line and radius used to define the set of normal cell clusters;
• (ii) superimposing a representation of the indication of the distribution of biological cells of a test set of biological cells on the histogram;
• and
• (iii) comparing frequencies of occurrence of biological cells along the centroid line in the set of normal cell clusters to frequencies of occurrence of biological cells along the centroid line in the test set of biological cells using the superimposed representation on the histogram;
• (c) characterizing the test set of biological cells based on the comparison of the indication of the distribution of biological cells of a test set of biological cells and the indication of the distribution of biological cells of the defined set of normal cell clusters in the histogram;
• and
• (d) generating an indication of a diagnosis of cancer in the test set of biological cells based on the characterization of the test set of biological cells.