Abstract:
Cervical spine fracture localization in medical images is a challenging task that requires a large
amount of labeled data for accurate diagnosis. However, obtaining labeled data is time-consuming and
difficult, which limits the application of supervised learning methods. In this thesis, we propose a semisupervised
learning approach to improve the accuracy of cervical spine fracture localization by combining
a small amount of labeled data with a larger amount of unlabeled data.
Our approach leverages semi-supervised learning techniques to learn patterns and features in the
larger set of unlabeled CT scans, which improves the model's ability to generalize to new and unseen
cases. Additionally, our approach is more robust to noisy or inaccurate labeled data, as the model can
learn to ignore or weight the labeled data based on its confidence in the label.
To increase the amount of labeled data available for training, we also explore data augmentation
techniques, such as rotation, flipping, cropping. We demonstrate the effectiveness of our approach
through experiments on a dataset of CT scans for cervical spine fracture localization.
Our results show that our semi-supervised learning approach improves the accuracy of cervical
spine fracture localization compared to traditional supervised learning methods, even when trained on a
limited amount of labeled data. Overall, our approach has the potential to improve the diagnosis of CSFs
in medical images, which can ultimately lead to better patient outcomes.