Retraction Notice: Identification of Endometrial Cancer in Myometrium Using ANFIS and CNN Algorithms for Accuracy and Sensitivity Improvement
Pasala Nagamohan Reddy, R. Puviarasi, Murugesan · 2022 14th International Conference on Mathematics, Actuarial Science, Computer Science and Statistics (MACS) · 2022
The main aim of this work is to measure the accuracy and sensitivity using the Adaptive Neuro Fuzzy Inference System in comparison with the Convolution Neural Network algorithm for identification of Endometrial Cancer. Materials and Methods: Endometrial cancer images from mini-MIAS database are used in this research work for experiment. For endometrial cancer detection a total of 20 samples are used out of which 10 samples are used for group 1 and 10 for group 2 used with 80 % of G power and maximum accepted error is fixed as 0.5 that is calculated from Clincalc. Results: The predictive model using ANFIS algorithm shows the Accuracy of 94.46 ± 0.39215 and Sensitivity of 91.04 ± 1.07207 followed by CNN algorithm provides an accuracy of 88.63 ± 1.02746 and sensitivity of 86.15 ± 1.34268 with the significance value of 0.002 for accuracy and 0.000 (2-tailed test) for sensitivity obtained from SPSS tool. Conclusion: Based on the outcome of the proposed work ANFIS classifier shows higher performance than CNN classifier to identify the cancerous cells in the beginning stage of the disease.