One-shot domain adaptation in multiple sclerosis lesion segmentation using convolutional neural networks

Abstract

In recent years, several convolutional neural network (CNN) methods have been proposed for the automated white matter lesion segmentation of multiple sclerosis (MS) patient images, due to their superior performance compared with those of other state-of-the-art methods. However, the accuracies of CNN methods tend to decrease significantly when evaluated on different image domains compared with those used for training, which demonstrates the lack of adaptability of CNNs to unseen imaging data. In this study, we analyzed the effect of intensity domain adaptation on our recently proposed CNN-based MS lesion segmentation method. Given a source model trained on two public MS datasets, we investigated the transferability of the CNN model when applied to other MRI scanners and protocols, evaluating the minimum number of annotated images needed from the new domain and the minimum number of layers needed to re-train to obtain comparable accuracy. Our analysis comprised MS patient data from both a clinical center and the public ISBI2015 challenge database, which permitted us to compare the domain adaptation capability of our model to that of other state-of-the-art methods. In both datasets, our results showed the effectiveness of the proposed model in adapting previously acquired knowledge to new image domains, even when a reduced number of training samples was available in the target dataset. For the ISBI2015 challenge, our one-shot domain adaptation model trained using only a single case showed a performance similar to that of other CNN methods that were fully trained using the entire available training set, yielding a comparable human expert rater performance. We believe that our experiments will encourage the MS community to incorporate its use in different clinical settings with reduced amounts of annotated data. This approach could be meaningful not only in terms of the accuracy in delineating MS lesions but also in the related reductions in time and economic costs derived from manual lesion labeling.

Robustness of Simultaneous Lesion and Neuroanatomy Segmentation in Multiple Sclerosis Using Deep Neural Networks

“…Performance on an additional external validation set was also similar to the performance on our internal validation set, suggesting that domain adaptation may be less crucial when applying our method to data from different clinics. This contrasts to previously reported results, where performance on data from previously unseen centres was substantially degraded without domain adaptation [15].…”

Multiple sclerosis cortical and WM lesion segmentation at 3T MRI: a deep learning method based on FLAIR and MP2RAGE

“…Specifically, several convolutional neural network (CNN) architectures have been tailored for the segmentation of MS WMLs ( Kaur et al, 2020 ). Some of them employ 2D convolutional layers ( Aslani et al, 2019 , Roy et al, 2018 ), whereas others employ 3D convolutional layers to incorporate information from all three spatial directions simultaneously ( Hashemi et al, 2019 , La Rosa et al, 2019 , Valverde et al, 2017 , Valverde et al, 2019 ). The clear edge these methods have over classical approaches is the capability of automatically extracting the relevant features for the task.…”

“…They have often considered only 2D MRI sequences and segmentations were performed with a large minimum lesion volume threshold; for instance, Valverde et al (2017) set this value to 20 voxels for the clinical MS datasets. Moreover, apart from ( Valverde et al, 2019 ), all these deep learning methods are currently not publicly available. Finally, with the exception of our previous work ( La Rosa et al, 2019 ), they have not been evaluated on CLs.…”

Bridge Segmentation Performance Gap Via Evolving Shape Prior

“…It is hard for well-trained segmentors to generalize to unseen images with appearance shifts. Fully-supervised fine-tuning tackles the problem with many annotations [4], [15]. Weaklysupervised fine-tuning for a specific image reduces the need of extra annotations [6], [16], [17].…”

“…Gibson et al [4] proved that utilizing annotated images of as few as 8 subjects from the unseen site for calibration is possible to address the inter-site prostate segmentation in MR images. In [15], the model fine-tuned with a single annotated image achieved comparable results against full dataset trained competitors for lesion segmentation. However, the choice of annotated images may have a considerable impact on performances.…”