Parallel multiple instance learning for extremely large histopathology image analysisReportar como inadecuado




Parallel multiple instance learning for extremely large histopathology image analysis - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

BMC Bioinformatics

, 18:360

Imaging, image analysis and data visualization

Abstract

BackgroundHistopathology images are critical for medical diagnosis, e.g., cancer and its treatment. A standard histopathology slice can be easily scanned at a high resolution of, say, 200,000×200,000 pixels. These high resolution images can make most existing imaging processing tools infeasible or less effective when operated on a single machine with limited memory, disk space and computing power.

ResultsIn this paper, we propose an algorithm tackling this new emerging -big data- problem utilizing parallel computing on High-Performance-Computing HPC clusters. Experimental results on a large-scale data set 1318 images at a scale of 10 billion pixels each demonstrate the efficiency and effectiveness of the proposed algorithm for low-latency real-time applications.

ConclusionsThe framework proposed an effective and efficient system for extremely large histopathology image analysis. It is based on the multiple instance learning formulation for weakly-supervised learning for image classification, segmentation and clustering. When a max-margin concept is adopted for different clusters, we obtain further improvement in clustering performance.

KeywordsHistopathology image analysis Microscopic image analysis Multiple instance learning Parallelization AbbreviationsCADComputer aided diagnosis

CBIRContent based image retrieval

ccMILContext-constrained multiple instance learning

GLCMGray level co-occurence matrix

GMGeneralized mean

HOGHistogram of oriented gradient

HPCHigh-performance-computing

IOInput-Output

KNNk-Nearest neighbor

LABLocally assembled binary

LSALatent semantic analysis

MCILMultiple clustered instance learning

MILMultiple instance learning

MIPMaximum intensity projection

MIL-BoostMultiple instance learning-boost

MPIMessage passing interface

OpenMPOpen multi-processing

PCPersonal computer

PMILParallel multiple instance learning

RAMRandom access memory

RAID6Redundant arrays of independent disk 6

RDMARemote direct memory access

ROCReceiver operating characteristic

SIFTScale invariant feature transform

SVMsSupport vector machines

Download fulltext PDF



Autor: Yan Xu - Yeshu Li - Zhengyang Shen - Ziwei Wu - Teng Gao - Yubo Fan - Maode Lai - Eric I-Chao Chang

Fuente: https://link.springer.com/







Documentos relacionados