Image-Based Identification of Cell Cultures by Machine Learning
Journal Title: Journal of Advances in Mathematics and Computer Science - Year 2017, Vol 23, Issue 1
Abstract
Biomedical laboratories often use different cell types in the same assay or the same cell type in different assays. One cell type can become contaminated by another, or cells can be mis-identified, giving poor results. Addressing these issues by DNA analyses can be time-consuming, labor intensive or costly to implement. Here we uniquely employ Legendre moments (LM), Zernike moments (ZM), circularity and a genetic algorithm (GA) to advance a computer-based vision system, and we task it to identify four cell types used in virology: HeLa, Vero, BHK and PC3. By employing a k-nearest neighbor (kNN), multilayer perceptron (MLP), Convolutional Neural Networks (CNN) classifiers and a GA-selected 9-vector candidate comprising 4 ZMs, 4 LMs, and circularity, we provide adaptive system for deep machine learning. Our approach provides avenue to measure the performances of two of the conventional and popular classifiers (kNN and MLP) with a relatively recent classifier (CNN). We provide detailed mathematical treatments of the image signatures for accessibility and reproducibility in computer vision. Our methods are unique in biomedical applications. The performance of the kNN for k = 1, 2, and 3 using 10-fold cross-validation yielded accuracies of (83.59%, 82.03%, 81.25%) and (84.38%, 82.82%, 82.03%) for 8-class and 4-class training sets, respectively, drawn from the same data while those of the MLP and CNN were 86% and 87.25% respectively. These results establish the feasibility of reliable automated cell identification, with diverse applications in biological and biomedical research.
Authors and Affiliations
Oluleye Babatunde, Ashley Baltes, John Yin
Keywords
Related Articles
- EP ID EP321858
- DOI 10.9734/JAMCS/2017/34357
- Views 119
- Downloads 0
A Literature Study on Traditional Clustering Algorithms for Uncertain Data
Numerous traditional Clustering algorithms for uncertain data have been proposed in the literature such as k-medoid, global kernel k-means, k-mode, u-rule, uk-means algorithm, Uncertainty-Lineage database, Fuzzy c-means...
On the Notes of Quasi-Boundary Value Method for Solving Cauchy-Dirichlet Problem of the Helmholtz Equation
The Cauchy-Dirichlet problem of the Helmholtz equation yields unstable solution, which when solved with the Quasi-Boundary Value Method (Q-BVM) for a regularization parameter = 0. At this point of regularization parame...
A New Efficient Hybrid String Matching Algorithm to Solve the Exact String Matching Problem
The string matching algorithms are considered one of the most studied in the computer science field because the fundamental role they play in many different applications such as information retrieval, editors, security a...
Support Theorem for Random Evolution Equations in Hlderian Norm
In this paper, we purpose to prove the support theorem of the random evolution equation dX(t) = σ(X(t),Z(t))dWt + b(X(t), Y (t))dt (E) where X = {X(t), t ∈ [0; 1]} is the solution of (E) considered as which a random-vari...
Recurrent Neural Network Based Modeling of Gene Regulatory Network Using Bat Algorithm
Correct inference of genetic regulations inside a cell is one of the greatest challenges in post genomic era for the biologist and researchers. Several intelligent techniques and models were already proposed to identify...