A Parametric Optimization Oriented, AFSA Based Random Forest Algorithm: Application to the Detection of Cervical Epithelial Cells

Abstract

Cervical cancer is one of the most common cancers among women in the world, over 570,000 patients are affected annually. Pathological examination for patients using Pap Smear becomes the mainstream of cervical cancer diagnoses. Accurate diagnoses and analyses largely rely on 3 factors: cell segmentation, feature extraction and selection as well as classification. Firstly, a 2-layer segmentation algorithm based on block Maximum Between-Class Variance (Otsu) and Gradient Vector Flow (GVF) Snake model is applied to obtain regions of interest (ROI). Then the features of chroma, shape and texture are extracted and selected for a better classification performance. The random forest algorithm based on Artificial Fish Swarms Algorithm (AFSA) is used to recognize and classify cervical epithelial cells. The proposed methods were tested on 200 cervical Pap Smear images. Experimental results show that cervical cells can be segmented with an effective segmentation result. The proposed feature selection method achieved an accuracy of 81.31% with the minimum feature number. The improved random forest algorithm with 2 and 7 classification under fivefold cross-validation reaches the highest classification accuracy (96.86%). Experimental results showed that the proposed method has obvious recognition advantages, and thus provides a practical classification frame for the diagnoses of cervical cancer. INDEX TERMS Artificial fish swarm algorithm, random forest, cell detection, image segmentation.

Student Performance Prediction in Mathematics Course Based on the Random Forest and Simulated Annealing

“…In recent years, many scholars have also incorporated intelligent algorithms into random forest algorithms to achieve good research results; for example, some researchers combined genetic algorithm and random forest algorithm for fault detection to reduce OOB outside the bag error [22]; a combination of particle algorithm and random forest algorithm is proposed for feature selection research and has good experimental results [23]; researchers incorporated AFSA into random forest algorithm for feature selection as well as parameter optimization [24], and the simulation experimental results prove that the algorithm OOB error is generally small and the generalization ability is relatively strong, which provides an effective guidance method for random forest algorithm feature selection and parameter estimation. ) + 1, and it is found in the experimental data that when the value of M is relatively small, the algorithm classification effect is the best when the hyperparameter O chooses log 2 ( �� M √ ) + 1, but it cannot obtain the optimal effect at any time.…”

Deploying Patch-Based Segmentation Pipeline for Fibroblast Cell Images at Varying Magnifications

“…Recent research on skin segmentation has largely focused on the outer areas of the skin, such as skin pores [5], chronic wounds [6], and lesions or cancer [7], [8], or the outer layer of the skin, such as epidermal tissue [9] rather than the cellular structure of the skin. Furthermore, most research on cellular segmentation has focused on cells that have less significant shape changes over time, such as blood cells [10], [11], cervical cells [12], [13], and nucleus [14], and has been conducted at single magnification level. These cells tend to have circular shapes and regular dimensions.…”