國立虎尾科技大學 |

One-Shot Learning Model for Cancer Diagnosis from Histopathological Images.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	One-Shot Learning Model for Cancer Diagnosis from Histopathological Images./
Author:	Yarlagadda, Dig Vijay Kumar.
Description:	1 online resource (35 pages)
Notes:	Source: Masters Abstracts International, Volume: 57-04.
Subject:	Computer science. -
Online resource:	click for full text (PQDT)
ISBN:	9780355615609

One-Shot Learning Model for Cancer Diagnosis from Histopathological Images.
Yarlagadda, Dig Vijay Kumar.

One-Shot Learning Model for Cancer Diagnosis from Histopathological Images. - 1 online resource (35 pages)

Source: Masters Abstracts International, Volume: 57-04.

Thesis (M.S.)--University of Missouri - Kansas City, 2018.

Includes bibliographical references

Cancer diagnosis from tissue biomarker scoring is a vital technique used in determining type and grade of cancer. This is a significant part of workload for pathologists, the process is tedious, time consuming, subjective, error prone and lacks inter-pathologist agreement. Thousands of patients are misdiagnosed each year, and several automated image analysis techniques using Deep Neural Networks (DNN) have been proposed for analyzing histopathology images for various cancer types and datasets. Typical challenges for a deep neural network to operate in this setting are limited datasets, gigapixel images and small percentage and high variability of nuclei indicative of malignant tumors. Previous approaches have focused on applying DNNs to different cancer imaging datasets, but their theoretical understanding of the problem is limited. In this work, we aim to gain fundamental insights into the nature of problem and propose a single model which can diagnose several types of cancers. Further, we employ recent advances in one-shot learning to enable our model to learn and expand to different types of cancer only from a few examples. We demonstrate good performance of our model on cervical cancer dataset.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2018

Mode of access: World Wide Web

ISBN: 9780355615609Subjects--Topical Terms:

573171
Computer science.
Index Terms--Genre/Form:

554714
Electronic books.

One-Shot Learning Model for Cancer Diagnosis from Histopathological Images.
LDR:02404ntm a2200349K 4500 001 912195
005 20180608102941.5
006 m o u
007 cr mn||||a|a||
008 190606s2018 xx obm 000 0 eng d
020 $a 9780355615609
035 $a (MiAaPQ)AAI10742962
035 $a (MiAaPQ)umkc:11213
035 $a AAI10742962
040 $a MiAaPQ $b eng $c MiAaPQ
100 1 $a Yarlagadda, Dig Vijay Kumar. $3 1184445
245 1 0 $a One-Shot Learning Model for Cancer Diagnosis from Histopathological Images.
264 0 $c 2018
300 $a 1 online resource (35 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: 57-04.
500 $a Adviser: Praveen Rao.
502 $a Thesis (M.S.)--University of Missouri - Kansas City, 2018.
504 $a Includes bibliographical references
520 $a Cancer diagnosis from tissue biomarker scoring is a vital technique used in determining type and grade of cancer. This is a significant part of workload for pathologists, the process is tedious, time consuming, subjective, error prone and lacks inter-pathologist agreement. Thousands of patients are misdiagnosed each year, and several automated image analysis techniques using Deep Neural Networks (DNN) have been proposed for analyzing histopathology images for various cancer types and datasets. Typical challenges for a deep neural network to operate in this setting are limited datasets, gigapixel images and small percentage and high variability of nuclei indicative of malignant tumors. Previous approaches have focused on applying DNNs to different cancer imaging datasets, but their theoretical understanding of the problem is limited. In this work, we aim to gain fundamental insights into the nature of problem and propose a single model which can diagnose several types of cancers. Further, we employ recent advances in one-shot learning to enable our model to learn and expand to different types of cancer only from a few examples. We demonstrate good performance of our model on cervical cancer dataset.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2018
538 $a Mode of access: World Wide Web
650 4 $a Computer science. $3 573171
650 4 $a Artificial intelligence. $3 559380
650 4 $a Oncology. $3 593951
650 4 $a Pathology. $3 668553
655 7 $a Electronic books. $2 local $3 554714
690 $a 0984
690 $a 0800
690 $a 0992
690 $a 0571
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a University of Missouri - Kansas City. $b Computer Science. $3 1182463
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10742962 $z click for full text (PQDT)