Transfer Learning-Based Multi-Scale Denoising Convolutional Neural Network for Prostate Cancer Detection

Abstract

Background: Prostate cancer is the 4th most common type of cancer. To reduce the workload of medical personnel in the medical diagnosis of prostate cancer and increase the diagnostic accuracy in noisy images, a deep learning model is desired for prostate cancer detection. Methods: A multi-scale denoising convolutional neural network (MSDCNN) model was designed for prostate cancer detection (PCD) that is capable of noise suppression in images. The model was further optimized by transfer learning, which contributes domain knowledge from the same domain (prostate cancer data) but heterogeneous datasets. Particularly, Gaussian noise was introduced in the source datasets before knowledge transfer to the target dataset. Results: Four benchmark datasets were chosen as representative prostate cancer datasets. Ablation study and performance comparison between the proposed work and existing works were performed. Our model improved the accuracy by more than 10% compared with the existing works. Ablation studies also showed average improvements in accuracy using denoising, multi-scale scheme, and transfer learning, by 2.80%, 3.30%, and 3.13%, respectively. Conclusions: The performance evaluation and comparison of the proposed model confirm the importance and benefits of image noise suppression and transfer of knowledge from heterogeneous datasets of the same domain.

CAMELYON 17 Challenge: A Comparison of Traditional Machine Learning (SVM) with the Deep Learning Method

“…This report compares and contrasts the efficacy of deep learning and machine learning approaches to the study of breast cancer [4]. Our work here was informed by algorithms developed for the CAMELYON 17 competition.…”

“…Project features are extracted into a lower-dimensional feature space, and the newly produced features are frequently composites of the original features. Principle component analysis (PCA) is one example of feature extraction techniques [4]. Both feature extraction and feature selection have the potential to improve learning performance, reduce computational complexity, develop more generalizable models, and reduce storage requirements.…”

Efficient Bioinspired Feature Selection and Machine Learning Based Framework Using Omics Data and Biological Knowledge Data Bases in Cancer Clinical Endpoint Prediction

“…In recent decades, many artificially assisted systems for cancer diagnosis have been thoroughly investigated, either through medical imaging analysis [19] or omic data analysis [5], [9], [20]. Although the incredible advancements brought about by AI in clinical applications cannot be understated, another need that must be met to improve genetic diagnosis, prognosis, and drug development is to provide models that assist biologists, clinicians, and the pharmaceutical industry in selecting molecular biomarkers with potential diagnosis, prognosis, and therapeutic targets.…”

Multiround Transfer Learning and Modified Generative Adversarial Network for Lung Cancer Detection

“…To examine the issue of model overftting and better fne-tuning the models, 5-fold cross-validation is adopted that has been justifed as a common setting of k-fold crossvalidation (with k � 5) [33,34]. Since 10 benchmark datasets are chosen, at most, the target model performs 9-round of MTL-MGAN from nine source datasets.…”