Lien, Yi-Hsuan
(2021)
Bone Age Assessment with less human intervention.
Master's Thesis, University of Pittsburgh.
(Unpublished)
Abstract
Biomedical imaging allows doctors to examine the condition of a patient’s organs or tissues without a surgical procedure. Various modalities of imaging techniques have been developed, such as X-radiation (X-ray), Magnetic Resonance Imaging (MRI), and Computed Tomography (CT). For example, the Bone Age Assessment (BAA) evaluates the maturity in infants, children, and adolescents using their hand radiographs. It plays an essential role in diagnosing a patient with growth disorders or endocrine disorders, such that needed treatments could be provided. Computer-aided diagnosis (CAD) systems have been introduced to extract features from regions of interest in this field automatically. Recently, several deep learning methods are proposed to perform automated bone age assessment by learning visual features. This study proposes a BAA model, including image preprocessing procedures and transfer learning with a limited number of annotated samples. The goal is to examine the efficiency of data augmentations by using a publicly available X-ray data set. The model achieves a comparable MAE of 5.8 months, RMSE of 7.3 months, and accuracy (within 1 year) of more than 90% on the data set. We also study whether generating samples by a Generative Adversarial Network could be a valuable technique for training the model and prevent it from overfitting when the samples are insufficient.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
3 September 2021 |
| Date Type: |
Publication |
| Defense Date: |
19 July 2021 |
| Approval Date: |
3 September 2021 |
| Submission Date: |
2 July 2021 |
| Access Restriction: |
No restriction; Release the ETD for access worldwide immediately. |
| Number of Pages: |
43 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
Swanson School of Engineering > Electrical and Computer Engineering |
| Degree: |
MS - Master of Science |
| Thesis Type: |
Master's Thesis |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
Bone Age assessment, Machine Learning, Generative Adversarial Network |
| Date Deposited: |
03 Sep 2021 16:01 |
| Last Modified: |
03 Sep 2021 16:01 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/41377 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[feed]](/images/feed-icon-32x32.png){kind=icon}