Scalable histopathological image analysis via supervised hashing with multiple features

Jiang, Menglin; Zhang, Shaoting; Huang, Junzhou; Yang, Lin; Metaxas, Dimitris N.

doi:10.1016/j.media.2016.07.011

Cited by 41 publications

(31 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While new advanced approaches have improved image recognition (e.g., normal versus cancerous), the image interpretation of heterogeneous populations still suffers from lack of robust computerization approaches [66,11,26,37]. Current available automatic methods focus on classification of just one type of cancer versus the corresponding normal condition.…”

Section: Introductionmentioning

confidence: 99%

Deep Convolutional Neural Networks Enable Discrimination of Heterogeneous Digital Pathology Images

Khosravi

Kazemi

Imieliński

et al. 2017

Preprint

View full text Add to dashboard Cite

Pathological evaluation of tumor tissue is pivotal for diagnosis in cancer patients and automated image analysis approaches have great potential to increase precision of diagnosis and help reduce human error.In this study, we utilize various computational methods based on convolutional neural networks (CNN) and build a stand-alone pipeline to effectively classify different histopathology images across different types of cancer. In particular, we demonstrate the utility of our pipeline to discriminate between two subtypes of lung cancer, four biomarkers of bladder cancer, and five biomarkers of breast cancer. In addition, we apply our pipeline to discriminate among four immunohistochemistry (IHC) staining scores of bladder and breast cancers.Our classification pipeline utilizes a basic architecture of CNN, Google's Inceptions within three training strategies, and an ensemble of two state-of-the-art algorithms, Inception and ResNet. These strategies include training the last layer of Google's Inceptions, training the network from scratch, and fine-tunning the parameters for our data using two pre-trained version of Google's Inception architectures, Inception-V1 and Inception-V3.We demonstrate the power of deep learning approaches for identifying cancer subtypes, and the robustness of Google's Inceptions even in presence of extensive tumor heterogeneity. Our pipeline on average achieved accuracies of 100%, 92%, 95%, and 69% for discrimination of various cancer types, subtypes, biomarkers, and scores, respectively. Our pipeline and related documentation is freely available at https://github.com/ ih-lab/CNN_Smoothie.

show abstract

Section: Introductionmentioning

confidence: 99%

Deep Convolutional Neural Networks Enable Discrimination of Heterogeneous Digital Pathology Images

Khosravi

Kazemi

Imieliński

et al. 2017

Preprint

View full text Add to dashboard Cite

show abstract

“…t m,K indicates the time cost of retrieving K relevant images for the mth query image. The average/accumulated run time has been widely adopted for the evaluation, comparison and validation of large-scale medical image retrieval (Jiang et al, 2016a(Jiang et al, , 2015cZhang et al, 2015c,d). Still, run times need to be put in relationship to hardware resources available and are thus not always easy to interpret.…”

Section: Evaluation Measuresmentioning

confidence: 99%

Large-scale retrieval for medical image analytics: A comprehensive review

Zhang

Müller

et al. 2018

Medical Image Analysis

Self Cite

200

View full text Add to dashboard Cite

Over the past decades, medical image analytics was greatly facilitated by the explosion of digital imaging techniques, where huge amounts of medical images were produced with ever-increasing quality and diversity. However, conventional methods for analyzing medical images have achieved limited success, as they are not capable to tackle the huge amount of image data. In this paper, we review state-of-the-art approaches for large-scale medical image analysis, which are mainly based on recent advances in computer vision, machine learning and information retrieval. Specifically, we first present the general pipeline of large-scale retrieval, summarize the challenges/opportunities of medical image analytics on a large-scale. Then, we provide a comprehensive review of algorithms and techniques relevant to major processes in the pipeline, including feature representation, feature indexing, searching, etc. On the basis of existing work, we introduce the evaluation protocols and multiple applications of large-scale medical image retrieval, with a variety of exploratory and diagnostic scenarios. Finally, we discuss future directions of large-scale retrieval, which can further improve the performance of medical image analysis.

show abstract

“…We compared the output of proposed approachh with the existing approaches [23,40,41]. In the proposed CBMIR, we considered both 14 and 20 sub-images in matching process.…”

Section: Fig4 Precision (þ) For Proposed Andf Existing Approachee Smentioning

confidence: 99%

“…Seetharaman and Sathiamoorthy [22] introduced CBMIR for diverse medical image collection in the context of full range autoregressive model with Bayesian approach. In [23], retrieval of pathological images of breast cancer is reported and it performs supervised hashing for complementary features by combining several kernel functions. Later on, several CBMIR systems for histological images are reported [24][25][26][27][28][29][30].…”

Section: Introductionmentioning

confidence: 99%

Multi-Trend Structure Descriptor At Micro-Level For Histological Image Retrieval

Natarajan*¹

2019

IJRTE

View full text Add to dashboard Cite

Abstract: Since hospitals are generating and using image data extensively, medical im age databases and its size are rising rapidly. This led to difficulties in browsing and managing the huge databases. Therefore, the necessity for the development of efficient content-based medical image retrieval (CBMIR) system arises and is more challenging problem for researchers. In this paper, to alleviate the unbalanced distribution of image representation using multi-trend structure descriptor (MTSD), MTSD is computed at micro level i.e., image is divided int o number of sub-images and for each sub-image MTSD is exploited. In similarity measurement, we compared the MTSDs of corresponding sub-images in query and target images than the liner ordered collection of smallest similarity values between the sub-images are considered for retrieval. Experiments revels that computation of proposed feature at micro level retains the localized representation and considering the liner ordered collection of smallest similarity values between the sub-images provides consistency under illumination changes and noise and thus proposed CBMIR achieves better results. Pattern analysis, Image retrieval and classification. His research interests include Machine learning algorithms and Medical imageanalysis. He has published more than 25 papers in various International journals and more than 15 papers in national and international conferences. He has been on the reviewing and editorial board of many reputed journals.

show abstract

Scalable histopathological image analysis via supervised hashing with multiple features

Cited by 41 publications

References 33 publications

Deep Convolutional Neural Networks Enable Discrimination of Heterogeneous Digital Pathology Images

Deep Convolutional Neural Networks Enable Discrimination of Heterogeneous Digital Pathology Images

Large-scale retrieval for medical image analytics: A comprehensive review

Multi-Trend Structure Descriptor At Micro-Level For Histological Image Retrieval

Contact Info

Product

Resources

About