Related Experiment Video
Updated: Jun 27, 2025
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
Automatic Skeleton Segmentation in CT Images Based on U-Net
Eva Milara1, Adolfo Gómez-Grande2,3, Pilar Sarandeses2,3
1Biomedical Engineering and Telemedicine Centre, Center for Biomedical Technology, ETSI Telecomunicación, Universidad Politécnica de Madrid, 28040, Madrid, Spain.
Abstract:
Bone metastasis, emerging oncological therapies, and osteoporosis represent some of the distinct clinical contexts which can result in morphological alterations in bone structure. The visual assessment of these changes through anatomical images is considered suboptimal, emphasizing the importance of precise skeletal segmentation as a valuable aid for its evaluation. In the present study, a neural network model for automatic skeleton segmentation from bidimensional computerized tomography (CT) slices is proposed. A total of 77 CT images and their semimanual skeleton segmentation from two acquisition protocols (whole-body and femur-to-head) are used to form a training group and a testing group. Preprocessing of the images includes four main steps: stretcher removal, thresholding, image clipping, and normalization (with two different techniques: interpatient and intrapatient). Subsequently, five different sets are created and arranged in a randomized order for the training phase. A neural network model based on U-Net architecture is implemented with different values of the number of channels in each feature map and number of epochs. The model with the best performance obtains a Jaccard index (IoU) of 0.959 and a Dice index of 0.979. The resultant model demonstrates the potential of deep learning applied in medical images and proving its utility in bone segmentation.
Insights
This study introduces a deep learning model for automatic skeleton segmentation from CT scans. The U-Net based model accurately segments bone structures, aiding in the evaluation of conditions like osteoporosis and bone metastasis.
Area of Science:
- Medical Imaging
- Artificial Intelligence in Medicine
- Skeletal Biology
Background:
- Morphological bone alterations occur in various clinical contexts, including bone metastasis, oncological therapies, and osteoporosis.
- Visual assessment of bone changes from anatomical images is suboptimal.
- Precise skeletal segmentation is crucial for accurate bone evaluation.
Purpose of the Study:
- To propose a neural network model for automatic skeleton segmentation from 2D computerized tomography (CT) slices.
- To evaluate the model's performance in segmenting bone structures accurately.
Main Methods:
- Utilized 77 CT images with semimanual segmentation for training and testing.
- Implemented a U-Net based neural network architecture.
- Applied preprocessing steps including stretcher removal, thresholding, clipping, and normalization.
Main Results:
- Achieved a Jaccard index (IoU) of 0.959 and a Dice index of 0.979 with the best performing model.
- Demonstrated high accuracy in automatic bone segmentation.
Conclusions:
- Deep learning models, specifically U-Net architecture, show significant potential for medical image analysis.
- The developed model proves effective for precise bone segmentation, aiding clinical evaluation.
