2017
DOI: 10.1016/j.compmedimag.2017.06.001
|View full text |Cite
|
Sign up to set email alerts
|

Deep convolutional neural networks for automatic classification of gastric carcinoma using whole slide images in digital histopathology

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
159
0
4

Year Published

2018
2018
2023
2023

Publication Types

Select...
7
2

Relationship

0
9

Authors

Journals

citations
Cited by 274 publications
(163 citation statements)
references
References 22 publications
0
159
0
4
Order By: Relevance
“…It seems that the deeper and more complex structures of CNN have great potential to deal with unbalanced data problems. In this work, we developed complete detection and classification algorithms for M. tuberculosis identification using microscopy images. In further study, we will spread the system to other pathological image classifications such as gastric carcinoma (Sharma, Zerbe, Klempert, Hellwich, & Hufnagl, ) and hepatocellular carcinoma nuclei grading (Li, Jiang, & Pang, ).…”
Section: Resultsmentioning
confidence: 99%
“…It seems that the deeper and more complex structures of CNN have great potential to deal with unbalanced data problems. In this work, we developed complete detection and classification algorithms for M. tuberculosis identification using microscopy images. In further study, we will spread the system to other pathological image classifications such as gastric carcinoma (Sharma, Zerbe, Klempert, Hellwich, & Hufnagl, ) and hepatocellular carcinoma nuclei grading (Li, Jiang, & Pang, ).…”
Section: Resultsmentioning
confidence: 99%
“…This is directly followed by max pooling with window size of (2, 2), and identically sized strides. The output from this part is of size (16,16,64), where the values are: reduced width, reduced height and the number of filters. These operations decrease the spatial dimensions by a factor of 8, which in turn significantly reduces memory usage and can be considered an adaptation of network topology to a relatively simple texture structure of the input 51 .…”
Section: Network Architecturementioning
confidence: 99%
“…Since 2012, when the groundbreaking AlexNet was created by Alex Krizevsky 6 , the state of the art has rapidly shifted from machine learning algorithms using manual feature engineering (henceforth referred to as 'traditional' machine learning approaches) to new deep learning ones 7 . Medical imaging in general [8][9][10][11][12] , and histopathological image classification in particular [13][14][15][16][17][18][19][20] , became important applications of these methods. Multiple machine learning methods go beyond the tasks of tissue type classification and whole-slide segmentation, confirming there is more information about the patients encrypted in histopathological images than immediately visible by eye 5 .…”
Section: Introductionmentioning
confidence: 99%
“…AlexNet has been used widely recently as a feature extractor in areas such as mineral processing [21], medical image analysis [22,23], audio processing [24], text analysis [25], botany [26], and remote sensing [27].…”
Section: Alexnetmentioning
confidence: 99%