2023
DOI: 10.1016/j.media.2022.102703
|View full text |Cite
|
Sign up to set email alerts
|

A generalizable and robust deep learning algorithm for mitosis detection in multicenter breast histopathological images

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

0
14
0

Year Published

2023
2023
2024
2024

Publication Types

Select...
7
1

Relationship

1
7

Authors

Journals

citations
Cited by 22 publications
(14 citation statements)
references
References 36 publications
0
14
0
Order By: Relevance
“…It is also possible that our normalisation method was not sufficient to prevent a domain shift in the external test dataset. Other method such as Fourier-based data augmentation as described in Wang et al 24 could be adopted in a follow-up study to overcome this issue. We have chosen to train a U-Net model as this has been shown to be one of the most often-used models for automatic delineation in histopathology.…”
Section: Discussionmentioning
confidence: 99%
“…It is also possible that our normalisation method was not sufficient to prevent a domain shift in the external test dataset. Other method such as Fourier-based data augmentation as described in Wang et al 24 could be adopted in a follow-up study to overcome this issue. We have chosen to train a U-Net model as this has been shown to be one of the most often-used models for automatic delineation in histopathology.…”
Section: Discussionmentioning
confidence: 99%
“…As an emerging next-generation segmentation model emphasizing generalizability and prompt functionality, SAM demonstrates such potential but its performance on medical images remains unclear, especially for radiotherapy. 43 Current research on SAM in the field of medical imaging primarily follows two directions: Firstly, evaluating the performance of SAM across various types of medical images [44][45][46][47][48][49] ; and secondly, enhancing its performance through improvements to specific modules of SAM. [50][51][52][53][54][55] These studies, to some extent, demonstrated the generalizability of SAM.…”
Section: Introductionmentioning
confidence: 99%
“…However, most of the data used originated from public datasets, leading to a lack of performance assessment in realworld clinical settings. [44][45][46][47][48][49][50][51][52][53][54] Research reports on the application of SAM in auto-segmentation for clinical radiation therapy are still limited. Florian et al reported on the performance of SAM in the auto-segmentation of brain tumors in MRI images, but no further evaluation for the generalizability and human-AI interaction was reported, which is potentially the most important feature as the next generation auto-segmentation model for clinical radiation therapy.…”
Section: Introductionmentioning
confidence: 99%
“… 10 , 11 , 12 However, due to the diversity of mitotic nuclei shapes and textures, customizing these hand-crafted feature-based methods for mitosis detection tasks is difficult, and the outcomes are often unsatisfactory. 13 …”
Section: Introductionmentioning
confidence: 99%