A deep ensemble learning method for colorectal polyp classification with optimized network parameters

Abstract

Colorectal Cancer (CRC), a leading cause of cancer-related deaths, can be abated by timely polypectomy. Computer-aided classification of polyps helps endoscopists to resect timely without submitting the sample for histology. Deep learning-based algorithms are promoted for computer-aided colorectal polyp classification. However, the existing methods do not accommodate any information on hyperparametric settings essential for model optimisation. Furthermore, unlike the polyp types, i.e., hyperplastic and adenomatous, the third type, serrated adenoma, is difficult to classify due to its hybrid nature. Moreover, automated assessment of polyps is a challenging task due to the similarities in their patterns; therefore, the strength of individual weak learners is combined to form a weighted ensemble model for an accurate classification model by establishing the optimised hyperparameters. In contrast to existing studies on binary classification, multiclass classification require evaluation through advanced measures. This study compared six existing Convolutional Neural Networks in addition to transfer learning and opted for optimum performing architecture only for ensemble models. The performance evaluation on UCI and PICCOLO dataset of the proposed method in terms of accuracy (96.3%, 81.2%), precision (95.5%, 82.4%), recall (97.2%, 81.1%), F1-score (96.3%, 81.3%) and model reliability using Cohen’s Kappa Coefficient (0.94, 0.62) shows the superiority over existing models. The outcomes of experiments by other studies on the same dataset yielded 82.5% accuracy with 72.7% recall by SVM and 85.9% accuracy with 87.6% recall by other deep learning methods. The proposed method demonstrates that a weighted ensemble of optimised networks along with data augmentation significantly boosts the performance of deep learning-based CAD.

Oral cancer detection using transfer learning-based framework from histopathology images

“…Transfer learning is an approach that includes using previously trained networks to initialize CNN weights 23 – 26 It has been discovered that the effectiveness of transfer learning outperforms training the model from scratch on a small dataset. This is because transfer learning makes use of previously learned information.…”

On the Analyses of Medical Images Using Traditional Machine Learning Techniques and Convolutional Neural Networks

“…206 individuals were enlisted who had a significant or medium risk of developing colorectal cancer. After completing complete bowel preparation, the individuals were examined in the upright and supine postures using an adaptive incremental reconstruction technique with the Computed Tomography settings set to 120kV, standard error 45 to 50 Younas et al [ 304 ] In contrast to transfer learning, this research examines six previous pretrained Convolutional neural networks models and chose the best outperforming architecture exclusively for traditional ensemble structures Tanwar et al [ 305 ] The colonoscopy images are primarily enhanced and filtered using dynamic histogram equalization and directed image filtering techniques. Then, colorectal polyps in colonoscopy images are accurately detected and classified using Single Shot MultiBox Detector.…”

Colonoscopy polyp classification via enhanced scattering wavelet convolutional neural network

“…Sharma et al [22] proposed a multiple CNNs (ResNet, GoogleNet, Xception) classifier consultation strategy to create an effective and powerful classifier for polyp identification, achieving a performance measure greater than 95% in each of the algorithm parameters. Younas et al [28] proposed an ensemble CNN-based approach for colorectal polyp classification, achieving a 96.3% F1 score on a public dataset.…”