Adapting fisher vectors for histopathology image classification

Song, Yang; Zou, Ju Jia; Chang, Hang; Cai, Weidong

doi:10.1109/isbi.2017.7950592

Cited by 68 publications

(58 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Surprisingly, we achieved 85.1% of accuracy using the TCNN without data augmentation, a performance comparable to the baseline which employs an AlexNet CNN with millions of trainable parameters. Approach Accuracy (%) CNN (Alexnet) [7] 84.6 TCNN (DA 1×) 85.1 Baseline [4] 85.1 TCNN Inc (DA 72×) 85.7 Deep Features (DeCaf) [12] 86.3 CNN+Fisher [13] 86.9 MI Approach [11] 87.2 Inception V3 FT (DA 72×) 87.4…”

Section: Resultsmentioning

confidence: 99%

Texture CNN for Histopathological Image Classification

Matos¹,

Britto

Oliveira

et al. 2019

2019 IEEE 32nd International Symposium on Computer-Based Medical Systems (CBMS)

View full text Add to dashboard Cite

Biopsies are the gold standard for breast cancer diagnosis. This task can be improved by the use of Computer Aided Diagnosis (CAD) systems, reducing the time of diagnosis and reducing the inter and intra-observer variability. The advances in computing have brought this type of system closer to reality. However, datasets of Histopathological Images (HI) from biopsies are quite small and unbalanced what makes difficult to use modern machine learning techniques such as deep learning. In this paper we propose a compact architecture based on texture filters that has fewer parameters than traditional deep models but is able to capture the difference between malignant and benign tissues with relative accuracy. The experimental results on the BreakHis dataset have show that the proposed texture CNN achieves almost 90% of accuracy for classifying benign and malignant tissues.

show abstract

Section: Resultsmentioning

confidence: 99%

Texture CNN for Histopathological Image Classification

Matos¹,

Britto

Oliveira

et al. 2019

2019 IEEE 32nd International Symposium on Computer-Based Medical Systems (CBMS)

View full text Add to dashboard Cite

show abstract

“…[105][106][107][108][109][110][111][112][113][114][115][116][117][118][119][120][121][122] The provided label for visual input data (eg, image) can correspond to either an entire image, [111][112][113][123][124][125][126][127][128] a window within the image, 108,[115][116][117] or at the pixel level. [105][106][107]109,110,[118][119][120][121] With the advent of deep learning methods that strongly benefit from pixel-level annotations, the latter is the most common type of problem being currently studied. Another popular scenario is when the training data come with a weaker form of annotation than what is expected as the output of the machine learning system.…”

Section: Machine Learning In Digital Pathology Image Analysismentioning

confidence: 99%

“…127 Additionally, unsupervised learning is commonly employed for extracting the important measurements, or features, from the data. 107,130 There exist other variants of data supervision, which has less commonly been used in digital histopathology, such as reinforcement learning, semi-supervised learning, and selfsupervised learning.…”

Section: Machine Learning In Digital Pathology Image Analysismentioning

confidence: 99%

Digital image analysis in breast pathology—from image processing techniques to artificial intelligence

Robertson

Azizpour

Smith

et al. 2018

Translational Research

245

134

View full text Add to dashboard Cite

“…With VGG-16, the input image goes through a series of convolutional layers before it finally produces a dense set of local feature descriptors of 512 dimensions at the last fully convolutional layer. In this study, 3D features are generated with the pre-trained VGG-16 model by encoding local features of last convolutional layer from all slices of the 3D image using the Fisher Vector (FV) [30,31]…”

Section: Deep Learningmentioning

confidence: 99%

“…Feature extraction from image patches often results in a large number of features, with many being redundant, leading to high computational cost and poor discriminative strength. Therefore feature representation techniques such as BOVW [7] and FV [30,31]…”

Section: Feature Representationmentioning

confidence: 99%

Estimation of Three-Dimensional Chromatin Morphology for Nuclear Classification and Characterisation

Rana

Sowmya

Meijering

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Classification and characterisation of cellular morphological states are vital for understanding cell differentiation, development, proliferation and diverse pathological conditions. As the onset of morphological changes transpires following genetic alterations in the chromatin configuration inside the nucleus, the nuclear texture as one of the low-level properties if detected and quantified accurately has the potential to provide insights on nuclear organisation and enable early diagnosis and prognosis. This study presents a three dimensional (3D) nuclear texture description method for cell nucleus classification and variation measurement in chromatin patterns on the transition to another phenotypic state. The proposed approach includes third plane information using hyperplanes into the design of the Sorted Random Projections (SRP) texture feature. The significance of including third plane information for low-resolution volumetric images is investigated by comparing the performance of 3D texture descriptor with its respective pseudo 3D form that ignores the interslice intensity correlations. Following classification, changes in chromatin pattern are estimated by computing the ratio of heterochromatin and euchromatin corresponding to their respective intensities and image gradient obtained by 3D SRP. The proposed method is evaluated on two publicly available 3D image datasets of human fibroblast and human prostate cancer cell lines in two phenotypic states obtained from the public Statistics Online Computational Resource. Experimental results show that 3D SRP and 3D Local Binary Pattern provide better results than other utilised handcrafted descriptors and deep learning features extracted using a pre-trained model. The results also show the advantage of utilising 3D feature descriptor for classification over its corresponding pseudo version. In addition, the proposed method validates that as the cell passes to another phenotypic state, there is a change in intensity and aggregation of heterochromatin.Author SummaryAutomated classification and measurement of cellular phenotypic traits can significantly impact clinical decision making. Early detection of diseases requires an accurate description of low-level cellular features to detect small-scale abnormalities in the few abnormal cells in the tissue microenvironment. The challenge is the development of a computational approach for 3D textural feature description that can capture the heterogeneous information in multiple dimensions and characterise the cells in their ultimate and intermediate phenotypic states effectively. Our work has proposed the method and metrics to measure chromatin condensation pattern and classify the phenotypic states. Experimental evaluation on the 3D image set of human fibroblast and human prostate cancer cell collections validates the proposed method for the classification of cell states. Results also signify the credibility of proposed metrics to characterise the cellular phenotypic states and contributes to studies related to early diagnosis, prognosis and drug resistance.

show abstract

Adapting fisher vectors for histopathology image classification

Cited by 68 publications

References 15 publications

Texture CNN for Histopathological Image Classification

Texture CNN for Histopathological Image Classification

Digital image analysis in breast pathology—from image processing techniques to artificial intelligence

Estimation of Three-Dimensional Chromatin Morphology for Nuclear Classification and Characterisation

Contact Info

Product

Resources

About