Classification of Plant Somatic Embryos by Using Neural Network Classifiers

Abstract

The development of somatic embryos is characterized by a series of morphological changes. Quantitative kinetic studies have been hampered by the difficulties in enumerating and characterizing embryo populations. By employing neural networks, we have developed a pattern‐recognition system for characterizing the morphological features of carrot somatic embryos. This pattern‐recognition system employs a hierarchical decision tree to achieve optimal classification. It successfully classified carrot somatic embryos into normal and abnormal embryo classes. For normal embryo classes (globular, oblong, heart, and torpedo embryos), an accuracy of 90% or higher was achieved. The features identified by the neural network classifiers as most important for embryo classification are almost identical to those obtained by the branch‐and‐bound searching algorithm used previously. However, employing the neural networks shortens the system developing time greatly. Coupled with an image analysis system, this neural‐network‐based pattern‐recognition system shows great potential in embryo sorting and automation of synthetic seed production.

Protoplast Technology and Somatic Hybridisation in the Family Apiaceae

“…Shiotani et al used Multilayer perceptron (MLP) of artificial neural networks (ANNs) to classify the alive or dead cell status of Arabidopsis thaliana cultured protoplasts using digitalized imaged shape and color of cells [125]. In the family Apiaceae, MLP was used for the classification of somatic embryos and non-embryogenic structures of Apium graveolens that can be selected for transfer to the next culture phase [126,127]. Likewise, MLP was also applied for cell growth prediction and optimization in order to determine the final biomass level of Daucus carota by using initial inoculum and sugar concentration data [128].…”

Neural modeling of plant tissue cultures: a review

“…Classification and pattern recognition models built by ANNs are already being widely used in plant tissue cultures (Prasad and Dutta Gupta, 2008a), usually because the selection of somatic embryos in embryogenic cultures is tedious, costly and time-consuming. A system that recognizes the patterns and morphological features of carrot somatic embryos has been constructed using artificial intelligence technology (Ruan et al, 1997). A hierarchical decision tree consisting of three layers and four nodes was used to obtain an optimal classification.…”

“…The number of neurons in the hidden layer was varied and influenced the network's capacity to solve the problem. Ruan et al (1997) model categorized carrot somatic embryos with 90% accuracy or higher. Therefore, the pattern recognition system based on ANNs has Fig.…”

“…7. Hierarchical decision tree with four ANN nodes for somatic embryos classification (A) (according to Ruan et al, 1997) and with two ANN nodes (B) (according to Zhang et al, 1999) shown great potential for sorting embryos and artificial seed production automation. A system similar to the above classification has applied to the somatic embryos of the Douglas fir (Pseudotsuga menziesii) (Zhang et al, 1999).…”