Bio-inspired Swarm Techniques for Thermogram Breast Cancer Detection

Sayed, Gehad Ismail; Soliman, Mona; Hassanien, Aboul Ella

doi:10.1007/978-3-319-33793-7_21

Cited by 34 publications

(15 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From Table 6 many notices can be seen. As shown, the best results of statistical measurements are highlighted in Table 8 Parameter settings for PSO [50], ABC [51], CSO [56], GWO [52], CSA [53] and SCA [54] 6 are short for the DA algorithm using the ten chaotic maps that are listed in Table 3. As observed in Table 6, different versions of CDA outperformed the standard version of DA.…”

Section: The Performance Of Cda With Different Chaotic Mapsmentioning

confidence: 99%

“…The objective of this experiment is to compare the performance of CDA with Gauss chaotic map with seven other optimization algorithms, namely, PSO [50], ABC [51], GWO [52], CSA [53], SCA [54], SSA [55], and CSO [56]. The parameters settings for all meta-heuristic optimization algorithms are shown in Table 8.…”

Section: Cda Vs Other Meta-heuristic Optimization Algorithmsmentioning

confidence: 99%

See 1 more Smart Citation

Chaotic dragonfly algorithm: an improved metaheuristic algorithm for feature selection

2018

Self Cite

View full text Add to dashboard Cite

Selecting the most discriminative features is a challenging problem in many applications. Bio-inspired optimization algorithms have been widely applied to solve many optimization problems including the feature selection problem. In this paper, the most discriminating features were selected by a new Chaotic Dragonfly Algorithm (CDA) where chaotic maps embedded with searching iterations of the Dragonfly Algorithm (DA). Ten chaotic maps were employed to adjust the main parameters of dragonflies' movements through the optimization process to accelerate the convergence rate and improve the efficiency of DA. The proposed algorithm is employed for selecting features from the dataset that were extracted from the Drug bank database, which contained 6712 drugs. In this paper, 553 drugs that were bio-transformed into liver are used. This data have four toxic effects, namely, irritant, mutagenic, reproductive, and tumorigenic effect, where each drug is represented by 31 chemical descriptors. The proposed model is mainly comprised of three phases; data pre-processing, features selection, and the classification phase. In the data pre-processing phase, Synthetic Minority Over-sampling Technique (SMOTE) was used to solve the problem of the imbalanced dataset. At the features selection phase, the most discriminating features were selected using CDA. Finally, the selected features from CDA were used to feed Support Vector Machine (SVM) classifier at the classification phase. Experimental results proved the capability of CDA to find the optimal feature subset, which maximizing the classification performance and minimizing the number of selected features compared with DA and the other meta-heuristic optimization algorithms. Moreover, the experiments showed that Gauss chaotic map was the appropriate map to significantly boost the performance of DA. Additionally, the high obtained value of accuracy (81.82-96.08%), recall (80.84-96.11%), precision (81.45-96.08%) and F-Score (81.14-96.1%) for all toxic effects proved the robustness of the proposed model.

show abstract

Section: The Performance Of Cda With Different Chaotic Mapsmentioning

confidence: 99%

Section: Cda Vs Other Meta-heuristic Optimization Algorithmsmentioning

confidence: 99%

Chaotic dragonfly algorithm: an improved metaheuristic algorithm for feature selection

2018

Self Cite

View full text Add to dashboard Cite

show abstract

“…Hassanien, AboulElla, et al (2017) [8] used MFO based on rough sets for tomato disease detection. Sayed (2016) [9] also included MFO in swarm intelligence algorithms he used for biomedical segmentation problems. El Aziz (2017) [10] used MFO to find the optimum threshold value in order to remove the time consumption problem in multilevel thresholding problems.…”

Section: Related Workmentioning

confidence: 99%

Training Anfis System with Moth-Flame Optimization Algorithm

Canayaz

2019

ijisae

View full text Add to dashboard Cite

Adaptive Neuro Fuzzy Inference System (ANFIS) is an adaptive network that can use the computation and learning abilities of artificial neural network together with the inference feature of fuzzy logic. The ANFIS system, which is used in the solution of many problems such as classification and estimation of deep learning applications, meets the needs in many different areas such as modeling, control, and parameter estimation. In recent years, heuristic methods have been used for the training of this network, which requires initial and result parameters by its structure. Moth-Flame Optimization Algorithm (MFO) is one of the current heuristic methods modeled by the influence of the spiral movement of the moths towards the light source. In this study, the MFO algorithm was used for the first time for the optimization of initial and result parameters in the ANFIS system. In the determination of parameters, nonlinear system identification, time series estimation, classification problems were tried to be solved. When the results obtained for the ANFIS trained with the known heuristic methods such as Particle Swarm Optimization (PSO), Genetic Algorithm (GA) and Whale Optimization Algorithm (WOA) and the results of ANFIS trained by the MFO were examined, it was observed that the MFO had lower error values.

show abstract

“…Prakash et al [23] presented three segmentation techniques K-Means, Fuzzy C-Means and Gaussian Mixture Model -Expectation Maximization, in this paper also the color space transformation was important to achieve good results. Sayed et al [24] proposed a method using bioinspired swarm techniques forming clusters looking for the most optimal, the FA (Firefly Algorithm) algorithm was the best swarm version that obtained the highest accuracy, sensitivity, precision and specificity. Mejia et al [25] used morphological operators in order to enhance the ROI area, then thresholding and the Extended-minima transform are performed to search neighborhoods connected who delimited the ROI region.…”

Section: Introductionmentioning

confidence: 99%

Automatic Segmentation in Breast Thermographic Images Based on Local Pattern Variations

Sánchez-Ruiz¹,

Olmos-Pineda²,

Olvera-López³

2018

RCS

View full text Add to dashboard Cite

In order to detect any possible anomaly in the region of breast, the use of thermal images has experienced a considerable workload of research in the recent years, this due to the promising results of this technique. One of the principal tasks in this process is the segmentation of the Region of Interest (ROI), but this task is difficult, most of the proposed techniques perform a manual or semi-automatic process to extract it. In this paper, we propose an alternative technique to detect the ROI in thermal breast images. In the proposed technique, we focus on the higher temperature areas in the image. We apply local contrast enhancement, group higher temperature regions, spline cubic interpolation and statistical operations as a part of the method. The achieved results are competitive with the state of the art showing a new alternative to accomplish the automatic segmentation of thermal breast images.

show abstract

Bio-inspired Swarm Techniques for Thermogram Breast Cancer Detection

Cited by 34 publications

References 22 publications

Chaotic dragonfly algorithm: an improved metaheuristic algorithm for feature selection

Chaotic dragonfly algorithm: an improved metaheuristic algorithm for feature selection

Training Anfis System with Moth-Flame Optimization Algorithm

Automatic Segmentation in Breast Thermographic Images Based on Local Pattern Variations

Contact Info

Product

Resources

About