Self-supervised learning for classifying paranasal anomalies in the maxillary sinus
Standard
Self-supervised learning for classifying paranasal anomalies in the maxillary sinus. / Bhattacharya, Debayan; Behrendt, Finn; Becker, Benjamin Tobias; Maack, Lennart; Beyersdorff, Dirk; Petersen, Elina; Petersen, Marvin; Cheng, Bastian; Eggert, Dennis; Betz, Christian; Hoffmann, Anna Sophie; Schlaefer, Alexander.
In: INT J COMPUT ASS RAD, 2024.Research output: SCORING: Contribution to journal › SCORING: Journal article › Research › peer-review
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - JOUR
T1 - Self-supervised learning for classifying paranasal anomalies in the maxillary sinus
AU - Bhattacharya, Debayan
AU - Behrendt, Finn
AU - Becker, Benjamin Tobias
AU - Maack, Lennart
AU - Beyersdorff, Dirk
AU - Petersen, Elina
AU - Petersen, Marvin
AU - Cheng, Bastian
AU - Eggert, Dennis
AU - Betz, Christian
AU - Hoffmann, Anna Sophie
AU - Schlaefer, Alexander
N1 - © 2024. The Author(s).
PY - 2024
Y1 - 2024
N2 - PURPOSE: Paranasal anomalies, frequently identified in routine radiological screenings, exhibit diverse morphological characteristics. Due to the diversity of anomalies, supervised learning methods require large labelled dataset exhibiting diverse anomaly morphology. Self-supervised learning (SSL) can be used to learn representations from unlabelled data. However, there are no SSL methods designed for the downstream task of classifying paranasal anomalies in the maxillary sinus (MS).METHODS: Our approach uses a 3D convolutional autoencoder (CAE) trained in an unsupervised anomaly detection (UAD) framework. Initially, we train the 3D CAE to reduce reconstruction errors when reconstructing normal maxillary sinus (MS) image. Then, this CAE is applied to an unlabelled dataset to generate coarse anomaly locations by creating residual MS images. Following this, a 3D convolutional neural network (CNN) reconstructs these residual images, which forms our SSL task. Lastly, we fine-tune the encoder part of the 3D CNN on a labelled dataset of normal and anomalous MS images.RESULTS: The proposed SSL technique exhibits superior performance compared to existing generic self-supervised methods, especially in scenarios with limited annotated data. When trained on just 10% of the annotated dataset, our method achieves an area under the precision-recall curve (AUPRC) of 0.79 for the downstream classification task. This performance surpasses other methods, with BYOL attaining an AUPRC of 0.75, SimSiam at 0.74, SimCLR at 0.73 and masked autoencoding using SparK at 0.75.CONCLUSION: A self-supervised learning approach that inherently focuses on localizing paranasal anomalies proves to be advantageous, particularly when the subsequent task involves differentiating normal from anomalous maxillary sinuses. Access our code at https://github.com/mtec-tuhh/self-supervised-paranasal-anomaly .
AB - PURPOSE: Paranasal anomalies, frequently identified in routine radiological screenings, exhibit diverse morphological characteristics. Due to the diversity of anomalies, supervised learning methods require large labelled dataset exhibiting diverse anomaly morphology. Self-supervised learning (SSL) can be used to learn representations from unlabelled data. However, there are no SSL methods designed for the downstream task of classifying paranasal anomalies in the maxillary sinus (MS).METHODS: Our approach uses a 3D convolutional autoencoder (CAE) trained in an unsupervised anomaly detection (UAD) framework. Initially, we train the 3D CAE to reduce reconstruction errors when reconstructing normal maxillary sinus (MS) image. Then, this CAE is applied to an unlabelled dataset to generate coarse anomaly locations by creating residual MS images. Following this, a 3D convolutional neural network (CNN) reconstructs these residual images, which forms our SSL task. Lastly, we fine-tune the encoder part of the 3D CNN on a labelled dataset of normal and anomalous MS images.RESULTS: The proposed SSL technique exhibits superior performance compared to existing generic self-supervised methods, especially in scenarios with limited annotated data. When trained on just 10% of the annotated dataset, our method achieves an area under the precision-recall curve (AUPRC) of 0.79 for the downstream classification task. This performance surpasses other methods, with BYOL attaining an AUPRC of 0.75, SimSiam at 0.74, SimCLR at 0.73 and masked autoencoding using SparK at 0.75.CONCLUSION: A self-supervised learning approach that inherently focuses on localizing paranasal anomalies proves to be advantageous, particularly when the subsequent task involves differentiating normal from anomalous maxillary sinuses. Access our code at https://github.com/mtec-tuhh/self-supervised-paranasal-anomaly .
U2 - 10.1007/s11548-024-03172-5
DO - 10.1007/s11548-024-03172-5
M3 - SCORING: Journal article
C2 - 38850438
JO - INT J COMPUT ASS RAD
JF - INT J COMPUT ASS RAD
SN - 1861-6410
ER -