Method for classifying a sequence of input images representing a particle in a sample over time
AssigneesBiomerieux, Bioaster
InventorsPierre Mahé, Meriem El Azami, Elodie Degout-Charmette, Zohreh Sedaghat, Quentin Josso, Fabian Rol
Life of this patent
Estimating the term…
Abstract
A method for classifying a sequence of input images representing a target particle in a sample over time, includes the following steps performed by the data processing of a client, namely: (b) concatenation of the input images in the sequence as a three-dimensional stack; (c) direct classification of the three-dimensional stack using a convolutional neural network, CNN.
Claims12
A method for classifying a sequence of input images representing a target particle in a sample over time, the method being characterized in that it comprises implementing, by data processing means of a client, the following steps:
- (b) concatenating said input images of the sequence in the form of a three-dimensional stack;
- (c) directly classifying said three-dimensional stack by means of a convolutional neural network (CNN), with said CNN being composed of a succession of convolutional blocks composed of a ‘3D’ convolutional layer, applying four-dimensional filters to a four-dimensional input feature map so as to generate a four-dimensional output feature map, an activation layer and a 3D pooling layer, and then a flattening layer, and finally one or more fully connected layer(s).
A system for classifying a sequence of input images representing a target particle in a sample over time, comprising at least one client comprising data processing means, characterized in that said data processing means are configured to implement:
- concatenating said input images of the sequence in the form of a three-dimensional stack;
- directly classifying said three-dimensional stack by means of a convolutional neural network (CNN), with said CNN being composed of a succession of convolutional blocks composed of a ‘3D’ convolutional layer, applying four-dimensional filters to a four-dimensional input feature map so as to generate a four-dimensional output feature map, an activation layer and a 3D pooling layer, and then a flattening layer, and finally one or more fully connected layer(s).
Dependencies are inferred from the claim text.