Semantic learning machine improves the CNN-Based detection of prostate cancer in non-contrast-enhanced MRI

Abstract

Considering that Prostate Cancer (PCa) is the most frequently diagnosed tumor in Western men, considerable attention has been devoted in computer-assisted PCa detection approaches. However, this task still represents an open research question. In the clinical practice, multiparametric Magnetic Resonance Imaging (MRI) is becoming the most used modality, aiming at defining biomarkers for PCa. In the latest years, deep learning techniques have boosted the performance in prostate MR image analysis and classification. This work explores the use of the Semantic Learning Machine (SLM) neuroevolution algorithm to replace the backpropagation algorithm commonly used in the last fully-connected layers of Convolutional Neural Networks (CNNs). We analyzed the non-contrast-enhanced multispectral MRI sequences included in the PROSTATEx dataset, namely: T2-weighted, Proton Density weighted, Diffusion Weighted Imaging. The experimental results show that the SLM significantly outperforms XmasNet, a state-of-the-art CNN. In particular, with respect to XmasNet, the SLM achieved higher classification accuracy (without neither pre-training the underlying CNN nor relying on the backprogation) as well as a speed-up of one order of magnitude. CCS CONCEPTS • Computing methodologies → Neural networks; Bio-inspired approaches; Search methodologies; Learning settings;

An Iterative Algorithm for Semisupervised Classification of Hotspots on Bone Scintigraphies of Patients with Prostate Cancer

“…An algorithm for the classification of automatically detected hotspots in bone scintigraphy images of patients with prostate cancer was proposed. Such an algorithm can be used in combination with computer-assisted PCa detection approaches such as the one described in [ 29 ], making it extremely useful in the medical community, as it provides the physicians with an aiding tool to quantify whole-body bone scans from patients with bone metastases.…”

Predicting High-Risk Prostate Cancer Using Machine Learning Methods

“…NPV does the same for negatives. F1 scores are correlated with a low rate of false positives and a low rate of false negatives [10].…”

“…Where training and test sets were from the same dataset, we evaluated the classifiers by using holdout and 10-fold cross-validation [25]. The metrics used were accuracy, AUC ROC [26], confusion matrices [10], sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and F1 score (F1). Sensitivity measures the ratio of correctly predicted positives to the total number of positives in the dataset, and the specificity does the same for negatives.…”

“…One screening method is magnetic resonance imaging (MRI), which has the potential to identify that someone with raised PSA levels actually has a low-risk cancer and hence does not need to have a biopsy [9][10][11]. Rundo et al have recently suggested automating the prostate segmentation component of MRI by using a technique based on the fuzzy c-means clustering algorithm, which would improve the time efficiency of MRI screening [11].…”

A Hybrid End-to-End Approach Integrating Conditional Random Fields into CNNs for Prostate Cancer Detection on MRI

“…Despite the growing interest in developing novel models for the task of PCa, little effort has been devoted to the addition of new types of layers. Recently, the Semantic Learning Machine (SLM) [26][27][28] neuroevolution algorithm was successfully employed to replace the backpropagation algorithm commonly used in the Fully-Connected (FC) layers of Convolutional Neural Networks (CNNs) [29,30]. When compared with backpropagation, SLM achieved higher classification accuracy in PCa detection as well as a training speed-up of one order of magnitude.…”