Back to Search
Start Over
Agreement of two pre-trained deep-learning neural networks built with transfer learning with six pathologists on 6000 patches of prostate cancer from Gleason2019 Challenge
- Source :
- Romanian Journal of Morphology and Embryology
- Publication Year :
- 2020
- Publisher :
- Academy of Medical Sciences, Romanian Academy Publishing House, Bucharest, 2020.
-
Abstract
- Introduction While the visual inspection of histopathology images by expert pathologists remains the golden standard method for grading of prostate cancer the quest for developing automated algorithms for the job is set and deep-learning techniques have emerged on top of other approaches. Methods Two pre-trained deep-learning networks, obtained with transfer learning from two general purpose classification networks - AlexNet and GoogleNet, originally trained on a proprietary dataset of prostate cancer were used to classify 6000 cropped images from Gleason2019 Challenge. Results The average agreement between the two networks and the six pathologists was found to be substantial for AlexNet and moderate for GoogleNet. When tested against the majority vote of the six pathologists the agreement was perfect and moderate for AlexNet, and GoogleNet, respectively. Despite our expectations, the average inter-pathologist agreement was moderate, while between the two networks it was substantial. Resulted accuracy for AlexNet and GoogleNet when tested against the majority vote as ground truth was of 85.51% and 74.75%, respectively. This result was higher than the score obtained on the dataset that they were trained on, showing their generalization capabilities. Conclusions Both the agreement and the accuracy indicate a better performance of AlexNet over GoogleNet, making it suitable for clinical deployment thus could potentially contribute to faster, more accurate and with higher reproducibility prostate cancer diagnosis.
- Subjects :
- Male
Embryology
Computer science
030232 urology & nephrology
Machine learning
computer.software_genre
History, 21st Century
Pathology and Forensic Medicine
03 medical and health sciences
Prostate cancer
0302 clinical medicine
Gleason grading system
medicine
Humans
Ground truth
Original Paper
Gleason2019 Grand Challenge
Artificial neural network
business.industry
Deep learning
deep learning
Prostatic Neoplasms
Cell Biology
General Medicine
medicine.disease
prostate cancer
neural networks
Pathologists
General purpose
030220 oncology & carcinogenesis
Female
Artificial intelligence
Transfer of learning
business
computer
agreement
Developmental Biology
Subjects
Details
- Language :
- English
- ISSN :
- 20668279 and 12200522
- Volume :
- 61
- Issue :
- 2
- Database :
- OpenAIRE
- Journal :
- Romanian Journal of Morphology and Embryology
- Accession number :
- edsair.doi.dedup.....1babde92e1bfdd31d51cd67174883a78