東華大學圖書館 |

Cross-Domain Generalization of Deep Learning-Based Image Analysis Algorithms in Histopathology = = Domanenubergreifende Generalisierbarkeit von Deep Learning-Basierten Bildanalysealgorithmen in der Histopathologie /

Record Type:	Electronic resources : Monograph/item
Title/Author:	Cross-Domain Generalization of Deep Learning-Based Image Analysis Algorithms in Histopathology =/ Frauke Eva Wilm.
Reminder of title:	Domanenubergreifende Generalisierbarkeit von Deep Learning-Basierten Bildanalysealgorithmen in der Histopathologie /
Author:	Wilm, Frauke Eva,
Description:	1 electronic resource (195 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 86-10, Section: B.
Contained By:	Dissertations Abstracts International86-10B.
Subject:	Cancer. -
Online resource:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31978650
ISBN:	9798311949798

Cross-Domain Generalization of Deep Learning-Based Image Analysis Algorithms in Histopathology = = Domanenubergreifende Generalisierbarkeit von Deep Learning-Basierten Bildanalysealgorithmen in der Histopathologie /
Wilm, Frauke Eva,

Cross-Domain Generalization of Deep Learning-Based Image Analysis Algorithms in Histopathology =Domanenubergreifende Generalisierbarkeit von Deep Learning-Basierten Bildanalysealgorithmen in der Histopathologie /Frauke Eva Wilm. - 1 electronic resource (195 pages)

Source: Dissertations Abstracts International, Volume: 86-10, Section: B.

The visual assessment of microscopic samples by pathologists constitutes an essential component of cancer diagnostics. Traditional pathology workflows were based on the visual assessment of samples under the microscope. The development of designated slide scanners has facilitated the digitization of microscopy samples which not only allowed for digital archiving and remote expert consultancy but also facilitated the use of machine learning-based image analysis algorithms for computer-aided diagnosis. Meanwhile, a wide range of computer-aided systems has been developed in the field of histopathology, often matching the performance of trained pathologists. Previous work has shown that machine learning-based image analysis algorithms, and especially convolutional neural networks, can be very susceptible to changes in the visual appearance of images. In pathology, these domain shiftscan be caused when applying trained algorithms to different morphologies, or samples prepared at a different pathology lab. The preparation of histologic samples follows routine stages, including tissue fixation, dehydration, paraffin embedding, and microtome sectioning. Subsequently, a sample is usually stained with a specific dye and digitized with a designated scanning system. The visual manifestation of these sample preparation steps can be very unique for the respective pathology lab.This thesis investigates the impact of different domain shifts on the performance of convolutional neural networks in histopathology. For these experiments, three routine tasks in cancer diagnostics were considered: cross-scanner mitotic figure detection, cross-domain tumor segmentation, and pan-tumor T-lymphocyte detection on immunohistochemistry samples. For the task of cross-scanner mitotic figure detection, domain adversarial training was employed. Evaluations of the learned embeddings demonstrated the successful extraction of scanner-agnostic features. For the task of cross-domain tumor segmentation, representation learning and, in particular, self-supervised learning was explored as a pre-training strategy to align feature embeddings across domains and thereby enhance the domain agnosticity for the downstream task. The results provide insights into the applicability of self-supervised learning in the context of histopathology. To date, this technique has mostly been employed in the field of natural images. In a project addressing the detection of tumor-infiltrating lymphocytes in immunohistochemistry samples, fine-tuning was leveraged to bridge the domain gap between different tumor indications. Initial experiments exhibited degraded performance on out-of-distribution samples. By exploiting fine-tuning on a limited number of target domain samples, this degradation was effectively mitigated. The experiments allowed for recommendations on the development of robust algorithms for the detection of lymphocytes across different tumor morphologies.In the course of the thesis, several cross-domain datasets were curated, focusing on different sources of domain shift. This includes a fully annotated dataset of 350 whole slide images covering seven canine cutaneous tumor subtypes, which constitutes one of the most comprehensive open histopathology segmentation datasets to date. A high annotation quality of each published dataset was ensured through extensive multi-rater experiments on selected subsets of the data. By making these datasets publicly available, future work on the cross-domain generalization for histopathology was facilitated.

English

ISBN: 9798311949798Subjects--Topical Terms:

634186
Cancer.

Cross-Domain Generalization of Deep Learning-Based Image Analysis Algorithms in Histopathology = = Domanenubergreifende Generalisierbarkeit von Deep Learning-Basierten Bildanalysealgorithmen in der Histopathologie /
LDR:05121nmm a22003973i 4500 001 2432685
005 20250731101912.5
006 m o d
007 cr|nu||||||||
008 261011s2025 miu||||||m |||||||eng d
020 $a 9798311949798
035 $a (MiAaPQD)AAI31978650
035 $a (MiAaPQD)FAUErlangenFAU_35170_
035 $a AAI31978650
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Wilm, Frauke Eva, $e author. $3 3822003
245 1 0 $a Cross-Domain Generalization of Deep Learning-Based Image Analysis Algorithms in Histopathology = $b Domanenubergreifende Generalisierbarkeit von Deep Learning-Basierten Bildanalysealgorithmen in der Histopathologie / $c Frauke Eva Wilm.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2025
300 $a 1 electronic resource (195 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: Dissertations Abstracts International, Volume: 86-10, Section: B.
500 $a Advisors: Breininger, Katharina; Aubreville, Marc; Walter, Thomas.
502 $b Ph.D. $c Friedrich-Alexander-Universitaet Erlangen-Nuernberg (Germany) $d 2025.
520 $a The visual assessment of microscopic samples by pathologists constitutes an essential component of cancer diagnostics. Traditional pathology workflows were based on the visual assessment of samples under the microscope. The development of designated slide scanners has facilitated the digitization of microscopy samples which not only allowed for digital archiving and remote expert consultancy but also facilitated the use of machine learning-based image analysis algorithms for computer-aided diagnosis. Meanwhile, a wide range of computer-aided systems has been developed in the field of histopathology, often matching the performance of trained pathologists. Previous work has shown that machine learning-based image analysis algorithms, and especially convolutional neural networks, can be very susceptible to changes in the visual appearance of images. In pathology, these domain shiftscan be caused when applying trained algorithms to different morphologies, or samples prepared at a different pathology lab. The preparation of histologic samples follows routine stages, including tissue fixation, dehydration, paraffin embedding, and microtome sectioning. Subsequently, a sample is usually stained with a specific dye and digitized with a designated scanning system. The visual manifestation of these sample preparation steps can be very unique for the respective pathology lab.This thesis investigates the impact of different domain shifts on the performance of convolutional neural networks in histopathology. For these experiments, three routine tasks in cancer diagnostics were considered: cross-scanner mitotic figure detection, cross-domain tumor segmentation, and pan-tumor T-lymphocyte detection on immunohistochemistry samples. For the task of cross-scanner mitotic figure detection, domain adversarial training was employed. Evaluations of the learned embeddings demonstrated the successful extraction of scanner-agnostic features. For the task of cross-domain tumor segmentation, representation learning and, in particular, self-supervised learning was explored as a pre-training strategy to align feature embeddings across domains and thereby enhance the domain agnosticity for the downstream task. The results provide insights into the applicability of self-supervised learning in the context of histopathology. To date, this technique has mostly been employed in the field of natural images. In a project addressing the detection of tumor-infiltrating lymphocytes in immunohistochemistry samples, fine-tuning was leveraged to bridge the domain gap between different tumor indications. Initial experiments exhibited degraded performance on out-of-distribution samples. By exploiting fine-tuning on a limited number of target domain samples, this degradation was effectively mitigated. The experiments allowed for recommendations on the development of robust algorithms for the detection of lymphocytes across different tumor morphologies.In the course of the thesis, several cross-domain datasets were curated, focusing on different sources of domain shift. This includes a fully annotated dataset of 350 whole slide images covering seven canine cutaneous tumor subtypes, which constitutes one of the most comprehensive open histopathology segmentation datasets to date. A high annotation quality of each published dataset was ensured through extensive multi-rater experiments on selected subsets of the data. By making these datasets publicly available, future work on the cross-domain generalization for histopathology was facilitated.
546 $a English
590 $a School code: 0575
650 4 $a Cancer. $3 634186
650 4 $a Histopathology. $3 3821989
650 4 $a Open data. $3 3682807
650 4 $a Deep learning. $3 3554982
650 4 $a Neural networks. $3 677449
650 4 $a Lymphocytes. $3 895384
650 4 $a Digitization. $3 3681264
650 4 $a Pathology. $3 643180
650 4 $a Privacy. $3 528582
650 4 $a Immunohistochemistry. $3 673425
650 4 $a Tumors. $3 893964
650 4 $a Morphology. $3 591167
690 $a 0287
690 $a 0571
710 2 $a Friedrich-Alexander-Universitaet Erlangen-Nuernberg (Germany). $e degree granting institution. $3 3822004
720 1 $a Breininger, Katharina $e degree supervisor.
720 1 $a Aubreville, Marc $e degree supervisor.
720 1 $a Walter, Thomas $e degree supervisor.
773 0 $t Dissertations Abstracts International $g 86-10B.
790 $a 0575
791 $a Ph.D.
792 $a 2025
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31978650