Review of Histopathological Image Segmentation via current Deep Learning Approaches

Dabass, Manju; Vig, Rekha; Vashisth, Sharda

doi:10.1109/ccaa.2018.8777616

Cited by 9 publications

(5 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A less complicated model can also give better classification results. This technique can be further used to increase the analytical as well as diagnostic accuracy of various applications like Cancer Classification ( [20,21]) and Tumor Segmentation ( [22]).…”

Section: Discussionmentioning

confidence: 99%

Effect Analysis of Contrast Enhancement Techniques on Cancer Classification in Colon Histopathology Images Using Machine Learning

Dabass

Vashisth

Vig

2020

Data Science and Analytics

Self Cite

View full text Add to dashboard Cite

The glandular morphology analysis done within the colon histopathological images is a crucial step required for grade determination of colon cancer. But the manual segmentation is quite laborious as well as time-consuming. Also, it suffers from the subjectivity present among the pathologists. Thus in order to minimize these limitations, the rising computational pathology has guided the researchers towards the advancement of various unsupervised and automated computerized methods for the gland segmentation task. However, this automated gland segmentation still remains a tricky task owing to numerous aspects like the need for high-level resolution for exact delineation of glandular boundaries, etc. Thus, in order to help researchers to develop automated gland segmentation techniques, various image enhancement techniques based on contrast adjustment are explained in this paper and their enhancement results are analyzed based on both objective qualitative assessment as well as subjective assessment given in the form of scores by the pathologists. And thus based on the qualitative analysis, a new combined technique i.e. Colormap Enhanced Image Sharpening is proposed in order to get an enhanced image in which all the critical elements are easily detectable. These techniques' effectiveness is also checked by applying them to do the image classification into two classes having benign and malignant cancer using the machine learning technique i.e. Bag of Features. The classification accuracy is examined for each enhancement techniques. These techniques' results will thus help pathologists in better colon histopathology image analysis.

show abstract

Section: Discussionmentioning

confidence: 99%

Effect Analysis of Contrast Enhancement Techniques on Cancer Classification in Colon Histopathology Images Using Machine Learning

Dabass

Vashisth

Vig

2020

Data Science and Analytics

Self Cite

View full text Add to dashboard Cite

show abstract

“…Transfer learning has also been applied to several medical imaging challenges such as [6], [26]. There have been many cases where deep learning was applied to histopathological slides, such as [2], [4], [9], [27] and a more comprehensive overview can be gained through [7]. To our knowledge, there has only been one attempt at semantic segmentation of epidermis using a deep learning approach [20], specifically using an adapted U-Net [22].…”

Section: Related Workmentioning

confidence: 99%

Semantic Segmentation of Histopathological Slides for the Classification of Cutaneous Lymphoma and Eczema

Scheurer

Ferrari

Bom

et al. 2020

Communications in Computer and Information Science

View full text Add to dashboard Cite

Mycosis fungoides (MF) is a rare, potentially life threatening skin disease, which in early stages clinically and histologically strongly resembles Eczema, a very common and benign skin condition. In order to increase the survival rate, one needs to provide the appropriate treatment early on. To this end, one crucial step for specialists is the evaluation of histopathological slides (glass slides), or Whole Slide Images (WSI), of the patients' skin tissue. We introduce a deep learning aided diagnostics tool that brings a twofold value to the decision process of pathologists. First, our algorithm accurately segments WSI into regions that are relevant for an accurate diagnosis, achieving a Mean-IoU of 69% and a Matthews Correlation score of 83% on a novel dataset. Additionally, we also show that our model is competitive with the state of the art on a reference dataset. Second, using the segmentation map and the original image, we are able to predict if a patient has MF or Eczema. We created two models that can be applied in different stages of the diagnostic pipeline, potentially eliminating life-threatening mistakes. The classification outcome is considerably more interpretable than using only the WSI as the input, since it is also based on the segmentation map. Our segmentation model, which we call EU-Net, extends a classical U-Net with an EfficientNet-B7 encoder which was pre-trained on the Imagenet dataset.

show abstract

“…The semantic segmentation of histopathological images is a prevalent computer vision problem, and various methods have been proposed to perform it. An approach that has shown great promise in recent years is the use of deep convolutional neural networks, 28 which can learn high-level feature information needed for pathology image segmentation. Deep learning algorithms of many forms have been used for several types of histopathological segmentation, including the segmentation of tumor regions.…”

Section: Introductionmentioning

confidence: 99%

“…Deep learning algorithms of many forms have been used for several types of histopathological segmentation, including the segmentation of tumor regions. 28 The analysis of labeled histopathological images presents various challenges including their large dimensions, the inter-and intra-observer variability in the labels created by pathologists, and the scarcity of accurately labeled data. 29 Deep learning for tumor segmentation Methods have been developed to combat these challenges and various deep learning methods exist that can perform tumor segmentation on histopathological images with a high level of accuracy.…”

mentioning

confidence: 99%

Deep learning-based tumor segmentation on digital images of histopathology slides for microdosimetry applications

Weishaupt

Torres

Camilleri-Broët

et al. 2021

Preprint

View full text Add to dashboard Cite

The goal of this study was (i) to use artificial intelligence to automate the traditionally labor-intensive process of manual segmentation of tumor regions in pathology slides performed by a pathologist and (ii) to validate the use of a deep learning architecture. Automation will reduce the human error involved in the manual process, increase efficiency, and result in more accurate and reproducible segmentation. This advancement will alleviate the bottleneck in the workflow in clinical and research applications due to a lack of pathologist time. Our application is patient-specific microdosimetry and radiobiological modeling, which builds on the contoured pathology slides. A deep neural network named UNet was used to segment tumor regions in pathology core biopsies of lung tissue with adenocarcinoma stained using hematoxylin and eosin. A pathologist manually contoured the tumor regions in 56 images with binary masks for training. To overcome memory limitations overlapping and non-overlapping patch extraction with various patch sizes and image downsampling were investigated individually. Data augmentation was used to reduce overfitting and artificially create more data for training. Using this deep learning approach, the UNet achieved accuracy of 0.91±0.06, specificity of 0.90±0.08, sensitivity of 0.92±0.07, and precision of 0.8±0.1. The F1/DICE score was 0.85±0.07, with a segmentation time of 3.24±0.03 seconds per image, thus achieving a 370±3 times increased efficiency over manual segmentation, which took 20 minutes per image on average. In some cases, the neural network correctly delineated the tumor's stroma from its epithelial component in tumor regions that were classified as tumor by the pathologist. The UNet architecture can segment images with a level of efficiency and accuracy that makes it suitable for tumor segmentation of histopathological images in fields such as radiotherapy dosimetry, specifically in the subfields of microdosimetry.

show abstract

Review of Histopathological Image Segmentation via current Deep Learning Approaches

Cited by 9 publications

References 15 publications

Effect Analysis of Contrast Enhancement Techniques on Cancer Classification in Colon Histopathology Images Using Machine Learning

Effect Analysis of Contrast Enhancement Techniques on Cancer Classification in Colon Histopathology Images Using Machine Learning

Semantic Segmentation of Histopathological Slides for the Classification of Cutaneous Lymphoma and Eczema

Deep learning-based tumor segmentation on digital images of histopathology slides for microdosimetry applications

Contact Info

Product

Resources

About