Publication
Bone age assessment using general use convolutional neural networks
| Summary: | Deep Learning methods have been applied to different medical imaging analysis tasks like, e.g., lesion classification and tissue segmentation. Bone age assessment is traditionally performed on an x-ray of the non-dominant hand applying the Greulich and Pyle or the Tanner Whitehouse methods. In this work, we first have tested several state-of-the-art Convolution Neural Networks models for assessing bone age that previously has shown great results in general computer vision tasks. Based on these results, we have developed/optimized a new model, which is presented here. For this purpose, we used transfer learning methods and trained the selected networks from scratch achieving a 7.89-month error rate when assessing bone age in females. |
|---|---|
| Main Authors: | Pinheiro, Gonçalo João Pereira |
| Other Authors: | Magalhães, Luís Gonzaga Mendes; Guevara, Miguel A. |
| Subject: | Medical Imaging X-rays Computer-Aided Diagnosis (CAD) Bone Age Bone Age Assessment Computer Vision Deep Learning Deep Neural Network Convolutional Neural Networks Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática Engenharia e Tecnologia::Engenharia Médica |
| Year: | 2019 |
| Country: | Portugal |
| Document type: | conference paper |
| Access type: | restricted access |
| Associated institution: | Universidade do Minho |
| Language: | English |
| Origin: | RepositóriUM - Universidade do Minho |
| Summary: | Deep Learning methods have been applied to different medical imaging analysis tasks like, e.g., lesion classification and tissue segmentation. Bone age assessment is traditionally performed on an x-ray of the non-dominant hand applying the Greulich and Pyle or the Tanner Whitehouse methods. In this work, we first have tested several state-of-the-art Convolution Neural Networks models for assessing bone age that previously has shown great results in general computer vision tasks. Based on these results, we have developed/optimized a new model, which is presented here. For this purpose, we used transfer learning methods and trained the selected networks from scratch achieving a 7.89-month error rate when assessing bone age in females. |
|---|
Funded activities
No funded projects found.