RESUMO

Objective.Multi-parametric magnetic resonance imaging (MP-MRI) has played an important role in prostate cancer diagnosis. Nevertheless, in the clinical routine, these sequences are principally analyzed from expert observations, which introduces an intrinsic variability in the diagnosis. Even worse, the isolated study of these MRI sequences trends to false positive detection due to other diseases that share similar radiological findings. Hence, the main objective of this study was to design, propose and validate a deep multimodal learning framework to support MRI-based prostate cancer diagnosis using cross-correlation modules that fuse MRI regions, coded from independent MRI parameter branches.Approach.This work introduces a multimodal scheme that integrates MP-MRI sequences and allows to characterize prostate lesions related to cancer disease. For doing so, potential 3D regions were extracted around expert annotations over different prostate zones. Then, a convolutional representation was obtained from each evaluated sequence, allowing a rich and hierarchical deep representation. Each convolutional branch representation was integrated following a special inception-like module. This module allows a redundant non-linear integration that preserves textural spatial lesion features and could obtain higher levels of representation.Main results.This strategy enhances micro-circulation, morphological, and cellular density features, which thereafter are integrated according to an inception late fusion strategy, leading to a better differentiation of prostate cancer lesions. The proposed strategy achieved a ROC-AUC of 0.82 over the PROSTATEx dataset by fusing regions ofKtransand apparent diffusion coefficient (ADC) maps coded from DWI-MRI.Significance.This study conducted an evaluation about how MP-MRI parameters can be fused, through a deep learning representation, exploiting spatial correlations among multiple lesion observations. The strategy, from a multimodal representation, learns branches representations to exploit radio-logical findings from ADC andKtrans. Besides, the proposed strategy is very compact (151 630 trainable parameters). Hence, the methodology is very fast in training (3 s for an epoch of 320 samples), being potentially applicable in clinical scenarios.

Assuntos

RESUMO

This paper presents a review of the state-of-the-art in histopathology image representation used in automatic image analysis tasks. Automatic analysis of histopathology images is important for building computer-assisted diagnosis tools, automatic image enhancing systems and virtual microscopy systems, among other applications. Histopathology images have a rich mix of visual patterns with particularities that make them difficult to analyze. The paper discusses these particularities, the acquisition process and the challenges found when doing automatic analysis. Second an overview of recent works and methods addressed to deal with visual content representation in different automatic image analysis tasks is presented. Third an overview of applications of image representation methods in several medical domains and tasks is presented. Finally, the paper concludes with current trends of automatic analysis of histopathology images like digital pathology.

Este artículo presenta una revisión del estado del arte en la representación de imágenes de histopatología utilizada en tareas de análisis automático. El análisis de imágenes hispatológicas es importante en la construcción de herramientas para el diagnóstico asistido por computador, sistemas de mejoramiento automático de imágenes y sistemas de microscopía virtual, entre otras aplicaciones. Estas imágenes tienen una gran mezcla de patrones visuales con características particulares que hacen de su análisis una tarea difícil. El artículo discute estas particularidades, el proceso de adquisición y los retos particulares al realizar un análisis automático. En la segunda sección se presenta una revisión de trabajos y métodos recientes enfocados a la representación del contenido visual en diferentes tareas de análisis automático. En tercer lugar, se presenta una visión general de las aplicaciones para los métodos de representación en diferentes dominios médicos. Finalmente el trabajo concluye con las actuales tendencias del análisis automático de imágenes de histopatología como la patología digital.

Este artigo é uma revisão do estado da arte na representação de imagens histopatológicas utilizadas nas tarefas de análise automáticos. O análise de imagens histopatológicas é importante na construção de ferramentas para o diagnóstico assistido por computador, sistemas de melhoramento automático de imagens e sistemas de microscopia virtual. Essas imagens tem uma grande mistura de padrões visuais com caraterísticas particulares, que fazem do análise uma tarefa difícil. O artigo discute essas particularidades, o processo de aquisição, e os desafios particulares no momento de realizar uma análise automático. Na segunda seção se apresenta uma revisão dos trabalhos e métodos recentes, com foco à representação do conteúdo visual em diferentes tarefas de análise automático. Na terceira, se apresenta uma visão geral das aplicações para os métodos de representação em diferentes domínios médicos. Finalmente, o artigo conclui com as atuais tendências do análise automático de imagens histopatológicas como a patologia digital.

Assuntos