Sustainability of Artificial Intelligence and Deep Learning Algorithms for Medical Image Classification: Case of Cancer Pathology
| Autor: | Anouar Boudhir Abdelhakim, Ben Ahmed Mohamed, Dahdouh Yousra |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Pathology
medicine.medical_specialty Contextual image classification Computer science business.industry Deep learning Feature extraction Context (language use) Convolutional neural network Random forest ComputingMethodologies_PATTERNRECOGNITION Feature (machine learning) medicine Artificial intelligence F1 score business Algorithm |
| Zdroj: | Advances in Science, Technology & Innovation ISBN: 9783030534394 |
| DOI: | 10.1007/978-3-030-53440-0_3 |
| Popis: | In the context of Artificial Intelligence’s Sustainability (AI), deep learning has sparked tremendous global interest in recent years. Deep Learning has been widely adopted in image recognition, speech recognition, and natural language processing, but is only beginning to impact on health care. In pathology, artificial intelligence and especially deep learning algorithms has been applied to pathology image analysis tasks such as tumor region identification, prognosis prediction, and detection of cancer areas and classification. This chapter presents different deep learning methods such as convolutional networks (ConvNets) for automatic classification of medical images—cancer pathology. We have proposed different models and different architectures. The first proposed model consists of feature extraction and classification with simple CNN; the second one is composed of two parts: we have used CNN as feature extractor by removing the last classification layers and we have passed the features to Random Forest; and the last one is by using transfer learning–Fine-Tuning–pre-trained CNN “DenseNet201” as classifier. Finally, we have evaluated our models using three metrics: accuracy, precision, and F1 score. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|