Dynamic Learning Rate in Deep CNN Model for Metastasis Detection and Classification of Histopathology Images

Abstract

Diagnosis of different breast cancer stages using histopathology whole slide images (WSI) is the gold standard in determining the grade of tissue metastasis. Computer-aided diagnosis (CAD) assists medical experts as a second opinion tool in early detection to prevent further proliferation. The field of pathology has advanced so rapidly that it is possible to obtain high-quality images from glass slides. Patches from the region of interest in histopathology images are extracted and trained using artificial neural network models. The trained model primarily analyzes and predicts the histology images for the benign or malignant class to which it belongs. Classification of medical images focuses on the training of models with layers of abstraction to distinguish between these two classes with less false-positive rates. The learning rate is the crucial hyperparameter used during the training of deep convolutional neural networks (DCNN) to improve model accuracy. This work emphasizes the relevance of the dynamic learning rate than the fixed learning rate during the training of networks. The dynamic learning rate varies with preset conditions between the lower and upper boundaries and repeats at different iterations. The performance of the model thus improves and attains comparatively high accuracy with fewer iterations.

The Detection of COVID-19 in Chest X-Rays Using Ensemble CNN Techniques

“…Continuously using the same high learning rate throughout the training process makes it far more difficult for the weights to converge to their ideal values since their values are shifted by far greater amounts, just like only using only using coarse focus on a microscope. This can cause the model to converge at local minima instead [33]. Conversely, only using a low learning rate will substantially increase the computation time, and likewise may get stuck at local minima.…”

Computer Aided Classifier of Colorectal Cancer on Histopatological Whole Slide Images Analyzing Deep Learning Architecture Parameters

“…Table 5 summarizes the relevant methods that appear in the bibliography, which use deep learning models for multiclass classification. In Table 5, there are strategies for dynamic modification of the learning rate, such as cyclical learning rates (Smith et al [10]), polynomial learning rates (Purnendu et al [7]) or dynamic learning rates (Anil et al [9]). There are also static learning rate methods, such as that of Anil et al [8], and methodologies based on transfer learning (such as that of Alinsaif et al [14]).…”

“…Some studies compared the influence of not using a fixed LR in each epoch. Anil et al [9] proposed the use of a dynamic learning rate. Smith et al [10] proposed cyclic learning rates, a method that lets the learning rate vary cyclically between the appropriate thresholds.…”

Teachable Machine: Optimization of Herbal Plant Image Classification Based on Epoch Value, Batch Size and Learning Rate

“…For epoch values ranging from 10; 50; 100; 250; 750 and 1000 epochs that have been done by previous researchers [60], [61], [62], then for batch sizes of 16; 32; 64; 128; 256 and 512 available on Teachable Machine, these values are the minimum and maximum values that are already available without the need for modification. Then, for the learning rate value based on the average of previous research ranges from 0.00001; 0.0001; 0.001; 0.01; 0.1 and 1 [63], [64], [65].…”