Adaptive color deconvolution for histological WSI normalization

Zheng, Yushan; Jiang, Zhiguo; Zhang, Haopeng; Xie, Fengying; Shi, Jun; Xue, Chenghai

doi:10.1016/j.cmpb.2019.01.008

Cited by 70 publications

(59 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We contained such batch effects in our input data through hand-picked ROIs and normalization. However, more sophisticated H&E normalization standards needed to be developed to allow a comprehensive application of deep learning for the large spectrum of disease conditions [42,43]. In particular, integration of cell-type-specific brain somatic gene information into disease classification will advance inter-and intra-observer liability of histopathology diagnosis as well as better understanding of underlying pathomechanisms [44].…”

Section: Different Limitations and Possible Solutions Moving Into Thementioning

confidence: 99%

Same same but different: a web-based deep learning application for the histopathologic distinction of cortical malformations

Kubach

Mühlebner

Söylemezoğlu

et al. 2019

Preprint

View full text Add to dashboard Cite

We trained a convolutional neural network (CNN) to classify H.E. stained microscopic images of focal cortical dysplasia type IIb (FCD IIb) and cortical tuber of tuberous sclerosis complex (TSC). Both entities are distinct subtypes of human malformations of cortical development that share histopathological features consisting of neuronal dyslamination with dysmorphic neurons and balloon cells. The microscopic review of routine stainings of such surgical specimens remains challenging. A digital processing pipeline was developed for a series of 56 FCD IIb and TSC cases to obtain 4000 regions of interest and 200.000 sub-samples with different zoom and rotation angles to train a CNN.Our best performing network achieved 91% accuracy and 0.88 AUCROC (area under the receiver operating characteristic curve) on a hold-out test-set. Guided gradient-weighted class activation maps visualized morphological features used by the CNN to distinguish both entities. We then developed a web application, which combined the visualization of whole slide images (WSI) with the possibility for classification between FCD IIb and TSC on demand by our pretrained and build-in CNN classifier.This approach might help to introduce deep learning applications for the histopathologic diagnosis of rare and difficult-to-classify brain lesions.

show abstract

Section: Different Limitations and Possible Solutions Moving Into Thementioning

confidence: 99%

Same same but different: a web-based deep learning application for the histopathologic distinction of cortical malformations

Kubach

Mühlebner

Söylemezoğlu

et al. 2019

Preprint

View full text Add to dashboard Cite

show abstract

“…Several deep learning approaches for lung cancer histopathological classification have gained success, in a supervision or weakly supervision manner, via single or multiple convolutional neural network (CNN) models [ 16 – 21 ] (Table 1 ). Computational tools have been developed for viewing, annotating, and data mining of whole slide images (WSIs) [ 22 – 26 ] (Table 1 ). Notably, QuPath [ 22 ], DeepFocus [ 23 ], ConvPath [ 24 ], HistQC [ 25 ], and ACD Model [ 26 ] are referenced in Table 1 as general WSI analysing tools, not specific for lung cancer.…”

Section: Introductionmentioning

confidence: 99%

“…Computational tools have been developed for viewing, annotating, and data mining of whole slide images (WSIs) [ 22 – 26 ] (Table 1 ). Notably, QuPath [ 22 ], DeepFocus [ 23 ], ConvPath [ 24 ], HistQC [ 25 ], and ACD Model [ 26 ] are referenced in Table 1 as general WSI analysing tools, not specific for lung cancer. Additionally, the relationships between molecular genotypes and morphological phenotypes have been explored in several pioneering studies [ 16 , 17 ] (Table 1 ).…”

Section: Introductionmentioning

confidence: 99%

Deep learning-based six-type classifier for lung cancer and mimics from histopathological whole slide images: a retrospective study

Yang

Chen

Cheng

et al. 2021

BMC Med

View full text Add to dashboard Cite

Background Targeted therapy and immunotherapy put forward higher demands for accurate lung cancer classification, as well as benign versus malignant disease discrimination. Digital whole slide images (WSIs) witnessed the transition from traditional histopathology to computational approaches, arousing a hype of deep learning methods for histopathological analysis. We aimed at exploring the potential of deep learning models in the identification of lung cancer subtypes and cancer mimics from WSIs. Methods We initially obtained 741 WSIs from the First Affiliated Hospital of Sun Yat-sen University (SYSUFH) for the deep learning model development, optimization, and verification. Additional 318 WSIs from SYSUFH, 212 from Shenzhen People’s Hospital, and 422 from The Cancer Genome Atlas were further collected for multi-centre verification. EfficientNet-B5- and ResNet-50-based deep learning methods were developed and compared using the metrics of recall, precision, F1-score, and areas under the curve (AUCs). A threshold-based tumour-first aggregation approach was proposed and implemented for the label inferencing of WSIs with complex tissue components. Four pathologists of different levels from SYSUFH reviewed all the testing slides blindly, and the diagnosing results were used for quantitative comparisons with the best performing deep learning model. Results We developed the first deep learning-based six-type classifier for histopathological WSI classification of lung adenocarcinoma, lung squamous cell carcinoma, small cell lung carcinoma, pulmonary tuberculosis, organizing pneumonia, and normal lung. The EfficientNet-B5-based model outperformed ResNet-50 and was selected as the backbone in the classifier. Tested on 1067 slides from four cohorts of different medical centres, AUCs of 0.970, 0.918, 0.963, and 0.978 were achieved, respectively. The classifier achieved high consistence to the ground truth and attending pathologists with high intraclass correlation coefficients over 0.873. Conclusions Multi-cohort testing demonstrated our six-type classifier achieved consistent and comparable performance to experienced pathologists and gained advantages over other existing computational methods. The visualization of prediction heatmap improved the model interpretability intuitively. The classifier with the threshold-based tumour-first label inferencing method exhibited excellent accuracy and feasibility in classifying lung cancers and confused nonneoplastic tissues, indicating that deep learning can resolve complex multi-class tissue classification that conforms to real-world histopathological scenarios.

show abstract

“…However, most of the existing methods have been optimised for a specific problem domain using a narrow dataset and fail to generalize to new domains, such as images from different labs or different tissues due to the high variability present in the histological images. A lot of work has been done in order to address the generalization challenge using techniques such as staining normalization [13]- [15], extensive data augmentation [16], [17], or utilization of datasets containing high variability such as multi-tissue datasets [8]. Although these approaches have achieved prominent results, they have still left room for further research, and some variability sources are yet to be studied, such as different tissue preparation techniques.…”

Section: Introductionmentioning

confidence: 99%

Generalized Fixation Invariant Nuclei Detection Through Domain Adaptation Based Deep Learning

Valkonen

Högnäs

Bova

et al. 2021

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

Nucleus detection is a fundamental task in histological image analysis and an important tool for many follow up analyses. It is known that sample preparation and scanning procedure of histological slides introduce a great amount of variability to the histological images and poses challenges for automated nucleus detection. Here, we studied the effect of histopathological sample fixation on the accuracy of a deep learning based nuclei detection model trained with hematoxylin and eosin stained images. We experimented with training data that includes three methods of fixation; PAXgene, formalin and frozen, and studied the detection accuracy results of various convolutional neural networks. Our results indicate that the variability introduced during sample preparation affects the generalization of a model and should be considered when building accurate and robust nuclei detection algorithms. Our dataset includes over 67 000 annotated nuclei locations from 16 patients and three different sample fixation types. The dataset provides excellent basis for building an accurate and robust nuclei detection model, and combined with unsupervised domain adaptation, the workflow allows generalization to images from unseen domains, including different tissues and images from different labs.

show abstract

Adaptive color deconvolution for histological WSI normalization

Cited by 70 publications

References 28 publications

Same same but different: a web-based deep learning application for the histopathologic distinction of cortical malformations

Same same but different: a web-based deep learning application for the histopathologic distinction of cortical malformations

Deep learning-based six-type classifier for lung cancer and mimics from histopathological whole slide images: a retrospective study

Generalized Fixation Invariant Nuclei Detection Through Domain Adaptation Based Deep Learning

Contact Info

Product

Resources

About