Browse Prior Art Database

Discrimination of cell cycle using Deep Learning

IP.com Disclosure Number: IPCOM000250468D
Publication Date: 2017-Jul-24
Document File: 1 page(s) / 46K

Publishing Venue

The IP.com Prior Art Database

This text was extracted from a Microsoft Word document.
This is the abbreviated version, containing approximately 55% of the total text.

TITLE:   Discrimination of cell cycle using Deep Learning

*Enter a brief abstract for this article

The present invention is utilized for automating the observation of the cell division period which is one of the properties of cells for elucidating the causes of diseases and responsiveness of medication.

Automatic discrimination of the following periodic classification by using color features from images expressing the difference of cycles generated by color to observe the division cycle of cells which divide into a large number over a long time in color.

(Green: S, G2, M phase, red: G1 phase, yellow: G1 to S phase, black: M to G1 phase)

*Enter the body of your Invention Disclosure Publication below.  You may include drawings and highlighting as appropriate.

Based on the preliminary learning result, periodic classification using color image information is performed by the following method. As a result, the present invention realizes periodic classification with no need for time series prediction and high accuracy.

1. Obtains the phase difference images of all the cells aligned in the time series of the inspection target period and all the images of the fluorescent images. It is also possible to sequentially select the previous time frame as an object to be inspected in chronological order from the last time frame on the premise that the cell exists continuously after splitting.

2. Acquire position information of all cells from the image at the first time (T0). The position...