Prostate cancer diagnosis using deep learning with 3D multiparametric MRI

  title={Prostate cancer diagnosis using deep learning with 3D multiparametric MRI},
  author={Saifeng Liu and Huaixiu Zheng and Yesu Feng and Wei Li},
  booktitle={Medical Imaging},
A novel deep learning architecture (XmasNet) based on convolutional neural networks was developed for the classification of prostate cancer lesions, using the 3D multiparametric MRI data provided by the PROSTATEx challenge. End-to-end training was performed for XmasNet, with data augmentation done through 3D rotation and slicing, in order to incorporate the 3D information of the lesion. XmasNet outperformed traditional machine learning models based on engineered features, for both train and… 

Deep Learning for Combined Lesion Detection and Classification in Prostate MRI

This study presents a model based on the 3D context enhanced deep convolutional neural network adapted to process prostate multi-parametric MRI and also predict the clinical significance of candidate lesions that reduces complexity and simplifies integration to clinical workflow.

Implementation of Convolutional Neural Network Architecture on 3D Multiparametric Magnetic Resonance Imaging for Prostate Cancer Diagnosis

A novel deep learning approach for automatic classification of prostate lesions in the corresponding magnetic resonance images by constructing a two-stage multimodal multi-stream convolutional neural network (CNN)-based architecture framework that outperformed most of the submitted methods and shared the highest value reported by the PROSTATEx Challenge organizer.

Prostate cancer malignancy detection and localization from mpMRI using auto-deep learning as one step closer to clinical utilization

Automatic diagnosis of malignant prostate cancer patients from mpMRI has been studied heavily in the past years. Model interpretation and domain drift have been the main road blocks for clinical

A Comprehensive Study of Data Augmentation Strategies for Prostate Cancer Detection in Diffusion-Weighted MRI Using Convolutional Neural Networks

This work has statically applied five most frequently used augmentation techniques to prostate diffusion-weighted magnetic resonance imaging training dataset of 217 patients separately and evaluated the effect of each method on the accuracy of prostate cancer detection.

A Transfer Learning Approach for Malignant Prostate Lesion Detection on Multiparametric MRI

The transfer learning–based deep learning approach is a promising method for prostate cancer detection on multiparametric magnetic resonance imaging images and features learned from ImageNet data set can be useful for medical images.

Semi-automatic classification of prostate cancer on multi-parametric MR imaging using a multi-channel 3D convolutional neural network

The diagnostic performance of the 3D CNN in detecting clinically significant prostate cancer is characterized by a good AUC and sensitivity and high specificity, and is comparable to that of an experienced radiologist.

An inception-based deep multiparametric net to classify clinical significance MRI regions of prostate cancer

An evaluation about how MP-MRI parameters can be fused, through a deep learning representation, exploiting spatial correlations among multiple lesion observations, leading to a better differentiation of prostate cancer lesions is conducted.

Deep Learning Regression for Prostate Cancer Detection and Grading in Bi-Parametric MRI

This work presents a neural network that simultaneously detects and grades cancer tissue in an end-to-end fashion and shows that the proposed model target outperforms standard multiclass classification and multi-label ordinal regression.

Enhancing classification performance of convolutional neural networks for prostate cancer detection on magnetic resonance images: a study with the semantic learning machine

The feasibility of using the Semantic Learning Machine (SLM) neuroevolution algorithm to replace the fully-connected architecture commonly used in the last layers of Convolutional Neural Networks (CNNs) is studied.



Decision forests for learning prostate cancer probability maps from multiparametric MRI

A machine learning framework for automatically detecting and localizing cancerous lesions within the prostate is developed and evaluated and shows that the combination of mpMRI and machine learning is a powerful tool for quantitatively diagnosing prostate cancer.

Dermatologist-level classification of skin cancer with deep neural networks

This work demonstrates an artificial intelligence capable of classifying skin cancer with a level of competence comparable to dermatologists, trained end-to-end from images directly, using only pixels and disease labels as inputs.

Computer-Aided Detection of Prostate Cancer in MRI

A fully automated computer-aided detection system which consists of two stages that detects initial candidates using multi-atlas-based prostate segmentation, voxel feature extraction, classification and local maxima detection and has potential in a first-reader setting.

Multiparametric MRI in prostate cancer management

The combination of conventional anatomical MRI and functional magnet resonance sequences—known as multiparametric MRI (mp-MRI)—is emerging as an accurate tool for identifying clinically relevant tumours owing to its ability to localize them.

Multiparametric MRI in Prostate Cancer

Prostate cancer (PCa) is the second most common cancer in men and an estimated 1.1 million men worldwide were diagnosed with prostate cancer in 2012 with almost 70% of the cases occurring in more

A radiomics model from joint FDG-PET and MRI texture features for the prediction of lung metastases in soft-tissue sarcomas of the extremities

In multivariable analysis, texture features extracted from fused scans significantly outperformed those from separate scans in terms of lung metastases prediction estimates, and the isotropic voxel size at which texture features were extracted had the most impact on predictive value.

ImageNet classification with deep convolutional neural networks

A large, deep convolutional neural network was trained to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes and employed a recently developed regularization method called "dropout" that proved to be very effective.

Very Deep Convolutional Networks for Large-Scale Image Recognition

This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers.