Mitosis domain generalization in histopathology images -- The MIDOG challenge

Aubreville, Marc; Stathonikos, Nikolas; Bertram, Christof A.; Klopleisch, Robert; Hoeve, Natalie ter; Ciompi, Francesco; Wilm, Frauke; Marzahl, Christian; Donovan, Taryn A.; Maier, Andreas; Breen, Jack; Ravikumar, Nishant; Chung, Youjin; Park, Jinah; Nateghi, Ramin; Pourakpour, Fattaneh; Fick, Rutger; Hadj, Saïma Ben; Jahanifar, Mostafa; Rajpoot, Nasir M.; Dexl, Jakob; Wittenberg, Thomas; Kondo, Satoshi; Lafarge, Maxime W.; Koelzer, Viktor H.; Liang, Jingtang; Wang, Yübo; Liu, Xi; Liu, Jingxin; Razavi, Salar; Khademi, April; Yang, Sen; Wang, Xiyue; Veta, Mitko; Breininger, Katharina

doi:10.48550/arxiv.2204.03742

Cited by 3 publications

(2 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The most common datasets used for mitosis detection are as follows: ICPR12 [3], ICPR14 [19], TUPAC16 [20], MIDOG22 [21], and AMIDA13 [22]. In this study, ICPR14 dataset is selected as the dataset to carry out experiments.…”

Section: Dataset Initializationmentioning

confidence: 99%

A Mitosis Detection and Classification Methodology with YOLOv5 and Fuzzy Classifiers

Nemati,

Samet,

Hancer

et al. 2023

World Congress on Electrical Engineering and Computer Systems and Science

View full text Add to dashboard Cite

Histopathological images are examined by pathologists to diagnose cancer. A major step in classifying nuclei as cancerous or non-cancerous is to detect and classify mitosis. Detecting and classifying mitosis, however, can be challenging due to its complex form of proliferation and high similarity to non-mitosis. Typically, pathologists use manual methods to diagnose cancer. However, it is a very laborious, time-consuming and costly method. Computer-aided diagnosis helps pathologists in the early detection and recognition of cancer and increases diagnostic precision. Many methods have been proposed over the years, but researchers have not been able to develop a system that provides high accuracy and reliability for a wide range of applications. This issue motivates us to develop a new methodology for identifying and classifying mitosis in breast histopathological images. First, mitotic-shaped cells are detected with YOLOv5. Both mitotic and non-mitotic cells can be detected by YOLOv5. In results of YOLOv5 diagnosis accuracy and reliability are reduced. After the detection process of mitotic-shaped cells with YOLOv5, fuzzy-based classifiers such as Fuzzy-based K Nearest Neighbor, Fuzzy Min-Max, and Fuzzy Random Forest are applied to distinguish mitotic cells from non-mitotic cells. The performance verification of the proposed methodology is conducted on the MITOS ICPR14 dataset in terms of Precision, Recall and F1-Score.

show abstract

Section: Dataset Initializationmentioning

confidence: 99%

A Mitosis Detection and Classification Methodology with YOLOv5 and Fuzzy Classifiers

Nemati,

Samet,

Hancer

et al. 2023

World Congress on Electrical Engineering and Computer Systems and Science

View full text Add to dashboard Cite

show abstract

“…The Mitosis Domain Generalization Challenge 2021 (MI-DOG2021) (2) was held to evaluate the performance of several methods for mitosis detection by using breast cancer cases digitized by different scanners. MIDOG2022 (3) is held as extension of MIDOG2021.…”

Section: Introductionmentioning

confidence: 99%