Cell analysis method and cell analyzer

1 . A cell analysis system to classify types of cells in a biological sample, comprising:

a cell measuring apparatus including (i) a flow cell through which cells in a measurement sample flows, (ii) a light source configured to irradiate the cells flowing through the flow cell for optical interrogation and (iii) a light detector configured to sense light from a respective one of optically interrogated cells, wherein the light sensed by the light detector carries an analog waveform signal indicative of a morphological feature of a respective one of cells optically sensed by the light detector, and wherein the cell measuring apparatus is further configured to sample the analog waveform signal at a predetermined sampling rate to convert the analog waveform signal into a matrix of cell feature values, which is a digital representation of the morphological feature of the respective one of the cells optically sensed by the light detector; and

a cell analyzer including a host processor and a parallel processing processor, the parallel processing processor including a plurality of arithmetic units each operable to perform a calculation on an assigned subset of a matrix operation, one of the host processor or the parallel processing processor being programmed with an artificial intelligence (AI) algorithm that has a neural network structure trained in advance with training data for classification of types of cells, the training data obtained from cells of already determined types,

the host processor being further programmed to perform a cell type identification process on respective target cells of unknown types, wherein the cell type identification process comprises:

receiving, from the cell measuring apparatus, the matrix of cell feature data representative of one target cell among the target cells of unknown types;

running the AI algorithm on the received matrix of cell feature values for calculation of a matrix operation to determine a type of said one target cell;

dividing the matrix operation into subsets of calculations and assigning the subsets of calculations to at least some of the arithmetic units of the parallel processing processor for parallel processing of the matrix operation; and

analyzing a result of the matrix operation performed by the parallel processing processor to determine the type of said one target cell.

2 . The cell analysis system according to claim 1 , further comprising a data transfer network operable to transfer the matrix of cell feature values from the cell measuring apparatus to the cell analyzer over the data transfer network.

3 . The cell analysis system according to claim 1 , wherein the at least one light detector is configured to sense non-fluorescent light or fluorescent light from the optically interrogated cells.

4 . The cell analysis system according to claim 3 , wherein the at least one light detector is configured to sense a forward scattered light, a side scattered light or a side fluorescence light from the optically interrogated cells in the biological sample.

5 . The cell analysis system according to claim 1 , wherein the cell measuring apparatus is operable to add identification information to the matrix of cell feature values, wherein the identification information includes any of: (1) an identification of the biological sample; (2) an identification of the optically interrogated cells; (3) an identification of a patient from which the biological sample is obtained; (4) an identification of a test performed on the cells in the biological sample; (5) an identification of the cell measuring apparatus; and (6) an identification of a test-related facility where the cell measuring apparatus is situated.

6 . The cell analysis system according to claim 5 , wherein the host processor is programmed to output the determined type of said one target cell with the identification information.

7 . The cell analysis system according to claim 1 , wherein the host processor is programmed to apply a digital filter to the received matrix of cell feature values to calculate cell features represented by the received matrix of cell feature values.

8 . The cell analysis system according to claim 1 , wherein said one target cell is a white blood cell, and the cell analyzer is programmed to determine a subtype of the white blood cell.

9 . A cell analysis method for classify types of cells in a biological sample, comprising:

irradiating the cells flowing through a flow cell for optical interrogation;

sensing by a light detector light from a respective one of optically interrogated cells, wherein the light sensed by the light detector carries an analog waveform signal indicative of a morphological feature of a respective one of cells optically sensed by the light detector;

sampling the analog waveform signal at a predetermined sampling rate to convert the analog waveform signal into a matrix of cell feature values, which is a digital representation of the morphological feature of the respective one of the cells optically sensed by the light detector; and

performing a cell type identification process on respective target cells of unknown types, wherein the cell type identification process comprises:

receiving the matrix of cell feature values representative of one target cell among the target cells of unknown types;

running an artificial intelligence (AI) algorithm on the received matrix of cell feature values for calculation of a matrix operation to determine a type of said one target cell, wherein the AI algorithm has a neural network structure trained in advance with training data for classification types of cells, and the training data is obtained from cells of already determined types;

dividing the matrix operation into subsets of calculations and assigning the subsets of calculations to arithmetic units of a parallel processing processor for parallel processing of the matrix operation; and

analyzing a result of the matrix operation performed by the parallel processing processor to determine the type of said one target cell.

10 . The cell analysis method according to claim 9 , wherein receiving the matrix of cell feature values representative of one target cell comprises receiving the matrix of cell feature values representative of said one target cell through a data transfer network.

11 . The cell analysis method according to claim 9 , wherein sensing light from optically interrogated cells comprises sensing non-fluorescent light or fluorescent light from the optically interrogated cells.

12 . The cell analysis method according to claim 9 , further comprising adding identification information to the matrix of cell feature values, wherein the identification information includes any of: (1) an identification of the biological sample; (2) an identification of the optically interrogated cells; (3) an identification of a patient from which the biological sample is obtained; (4) an identification of a test performed on the cells in the biological sample; (5) an identification of the cell measuring apparatus; and (6) an identification of a test-related facility where the cell measuring apparatus is situated.

13 . The cell analysis method according to claim 11 , wherein sensing light from optically interrogated cells comprises sensing a forward scattered light, a side scattered light or a side fluorescence light from optically interrogated cells.

14 . The cell analysis method according to claim 9 , further comprising applying a digital filter to the received matrix of cell feature values to calculate cell features represented by the set of matrix of cell feature values.

15 . The cell analysis method according to claim 12 , further comprising outputting the determined type of said one target cell with the identification information.

16 . The cell analysis method according to claim 9 , wherein said one target cell is a white blood cell, and determining the type of said one target cell comprising determining a subtype of the white blood cell.