國立虎尾科技大學 |

Deep Learning Models to Characterize Smooth Muscle Fibers in Hematoxylin and Eosin Stained Histopathological Images of the Urinary Bladder

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Deep Learning Models to Characterize Smooth Muscle Fibers in Hematoxylin and Eosin Stained Histopathological Images of the Urinary Bladder/ Sridevi Kayyur Subramanya.
作者:	Kayyur Subramanya, Sridevi,
面頁冊數:	1 electronic resource (53 pages)
附註:	Source: Masters Abstracts International, Volume: 83-03.
Contained By:	Masters Abstracts International83-03.
標題:	Artificial intelligence. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28714267
ISBN:	9798538122974

Deep Learning Models to Characterize Smooth Muscle Fibers in Hematoxylin and Eosin Stained Histopathological Images of the Urinary Bladder
Kayyur Subramanya, Sridevi,

Deep Learning Models to Characterize Smooth Muscle Fibers in Hematoxylin and Eosin Stained Histopathological Images of the Urinary Bladder [eletronic resource] /Sridevi Kayyur Subramanya. - 1 electronic resource (53 pages)

Source: Masters Abstracts International, Volume: 83-03.

Muscularis propria (MP) and muscularis mucosa (MM), two types of smooth muscle fibers in the urinary bladder, are major benchmarks in staging bladder cancer to distinguish between muscle-invasive (MP invasion) and non-muscle-invasive (MM invasion) diseases. While patients with non-muscle-invasive tumor can be treated conservatively involving transurethral resection (TUR) only, more aggressive treatment options, such as removal of the entire bladder, known as radical cystectomy (RC) which may severely degrade the quality of patient's life, are often required in those with muscle-invasive tumor. Hence, given two types of image datasets, hematoxylin & eosin-stained histopathological images from RC and TUR specimens, we propose the first deep learning-based method for efficient characterization of MP. The proposed method is intended to aid the pathologists as a decision support system by facilitating accurate staging of bladder cancer. In this work, we aim to semantically segment the TUR images into MP and non-MP regions using two different approaches, patch-to-label and pixel-to-label. We evaluate four different state-of-the-art CNN-based models (VGG16, ResNet18, SqueezeNet, and MobileNetV2) and semantic segmentation-based models (U-Net, MA-Net, DeepLabv3+, and FPN) and compare their performance metrics at the pixel-level. The SqueezeNet model (mean Jaccard Index: 95.44%, mean dice coefficient: 97.66%) in patch-to-label approach and the MA-Net model (mean Jaccard Index: 96.64%, mean dice coefficient: 98.29%) in pixel-to-label approach are the best among tested models. Although pixel-to-label approach is marginally better than the patch-to-label approach based on evaluation metrics, the latter is computationally efficient using least trainable parameters.

English

ISBN: 9798538122974Subjects--Topical Terms:

559380
Artificial intelligence.
Subjects--Index Terms:

Bladder cancer

Deep Learning Models to Characterize Smooth Muscle Fibers in Hematoxylin and Eosin Stained Histopathological Images of the Urinary Bladder
LDR:03391nam a22004693i 4500 001 1172843
005 20260622113215.5
006 m o d
007 cr|nu||||||||
008 260803s2021 miu||||||m |||||||eng d
020 $a 9798538122974
035 $a (MiAaPQD)AAI28714267
035 $a AAI28714267
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Kayyur Subramanya, Sridevi, $e author. $3 1503447
245 1 0 $a Deep Learning Models to Characterize Smooth Muscle Fibers in Hematoxylin and Eosin Stained Histopathological Images of the Urinary Bladder $c Sridevi Kayyur Subramanya. $h [eletronic resource] /
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2021
300 $a 1 electronic resource (53 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Masters Abstracts International, Volume: 83-03.
500 $a Advisors: Cui, Feng Committee members: Miyamoto, Hiroshi; Li, Rui.
502 $b M.S. $c Rochester Institute of Technology $d 2021.
520 # $a Muscularis propria (MP) and muscularis mucosa (MM), two types of smooth muscle fibers in the urinary bladder, are major benchmarks in staging bladder cancer to distinguish between muscle-invasive (MP invasion) and non-muscle-invasive (MM invasion) diseases. While patients with non-muscle-invasive tumor can be treated conservatively involving transurethral resection (TUR) only, more aggressive treatment options, such as removal of the entire bladder, known as radical cystectomy (RC) which may severely degrade the quality of patient's life, are often required in those with muscle-invasive tumor. Hence, given two types of image datasets, hematoxylin & eosin-stained histopathological images from RC and TUR specimens, we propose the first deep learning-based method for efficient characterization of MP. The proposed method is intended to aid the pathologists as a decision support system by facilitating accurate staging of bladder cancer. In this work, we aim to semantically segment the TUR images into MP and non-MP regions using two different approaches, patch-to-label and pixel-to-label. We evaluate four different state-of-the-art CNN-based models (VGG16, ResNet18, SqueezeNet, and MobileNetV2) and semantic segmentation-based models (U-Net, MA-Net, DeepLabv3+, and FPN) and compare their performance metrics at the pixel-level. The SqueezeNet model (mean Jaccard Index: 95.44%, mean dice coefficient: 97.66%) in patch-to-label approach and the MA-Net model (mean Jaccard Index: 96.64%, mean dice coefficient: 98.29%) in pixel-to-label approach are the best among tested models. Although pixel-to-label approach is marginally better than the patch-to-label approach based on evaluation metrics, the latter is computationally efficient using least trainable parameters.
546 $a English
590 $a School code: 0465
650 # 4 $a Artificial intelligence. $3 559380
650 # 4 $a Histology. $3 644460
650 # 4 $a Oncology. $3 593951
650 # 4 $a Morphology. $3 1183868
650 # 4 $a Bioinformatics. $3 583857
653 # # $a Bladder cancer
653 # # $a Deep Learning
653 # # $a Hematoxylin and eosin stained images
653 # # $a Image analysis
653 # # $a Medical Images
653 # # $a Urinary Bladder
690 $a 0715
690 $a 0287
690 $a 0992
690 $a 0800
690 $a 0414
710 2 # $a Rochester Institute of Technology. $b Bioinformatics. $e degree granting institution. $3 1503448
720 1 $a Cui, Feng $e degree supervisor.
773 0 # $t Masters Abstracts International $g 83-03.
790 $a 0465
791 $a M.S.
792 $a 2021
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28714267