Automated prostate cancer grading and diagnosis system using deep learning-based Yolo object detection algorithm

Developing an artificial intelligence-based prostate cancer detection and diagnosis system that can automatically determine important regions and accurately classify the determined regions on an input biopsy image. The Yolo general-purpose object detection algorithm was utilized to detect important...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Expert systems with applications Jg. 201; S. 117148
Hauptverfasser: Salman, Mehmet Emin, Çakirsoy Çakar, Gözde, Azimjonov, Jahongir, Kösem, Mustafa, Cedi̇moğlu, İsmail Hakkı
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York Elsevier Ltd 01.09.2022
Elsevier BV
Schlagworte:
Accuracy
Algorithms
Artificial intelligence
Biopsy
Computer vision
Data augmentation
Datasets
Deep learning
Diagnosis
Gleason grading
Image classification
Machine learning
Medical imaging
Object recognition
Prostate cancer
Prostate cancer detection
Prostate tissue classification
Test sets
Deep learning
Gleason grading
Prostate cancer detection
Prostate tissue classification
ISSN:0957-4174, 1873-6793
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Developing an artificial intelligence-based prostate cancer detection and diagnosis system that can automatically determine important regions and accurately classify the determined regions on an input biopsy image. The Yolo general-purpose object detection algorithm was utilized to detect important regions (for the localization task) and to grade the detected regions (for the classification task). The algorithm was re-trained with our prostate cancer dataset. The dataset was created by annotating 500 real prostate tissue biopsy images. The dataset was split into train/test parts as 450/50 real prostate tissue images, respectively, before the data augmentation process. Next, the training set consisting of 450 labeled biopsy images was pre-processed with the data augmentation method. This way, the number of biopsy images in the dataset was increased from 450 to 1776. Then, the algorithm was trained with the dataset and the automatic prostate cancer detection and diagnosis tool was developed. The developed tool was tested with two test sets. The first test set contains 50 images that are similar to the train set. Hence, 97% detection and classification accuracy has been achieved. The second test set contains 137 completely different real prostate tissue biopsy images, thus, 89% detection accuracy has been achieved. In this study, an automatic prostate cancer detection and diagnosis tool was developed. The test results show that high-accuracy (high-performance) prostate cancer diagnosis tools can be developed using AI (computer vision) methods such as object detection algorithms. These systems can decrease the inter-observer variability among pathologists, and help prevent the time delay in the diagnosis phase. •Automated prostate cancer detection and diagnosis system was developed.•The system was developed by applying the Gleason grading on prostate tissue images.•A new PCa dataset was created by a pathologist and reviewed by two other pathologists.•The fine-tuned Yolo object detection algorithm was re-trained via the dataset.•Grading outcomes of the Yolo algorithm was fused with ISUP to make the final decision.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0957-4174
1873-6793
DOI:10.1016/j.eswa.2022.117148