Zobrazeno 1 - 10
of 32
pro vyhledávání: '"Wang, Alan Q."'
Distribution shifts between sites can seriously degrade model performance since models are prone to exploiting unstable correlations. Thus, many methods try to find features that are stable across sites and discard unstable features. However, unstabl
Externí odkaz:
http://arxiv.org/abs/2409.05996
Autor:
Nguyen, Minh, Karaman, Batuhan K., Kim, Heejong, Wang, Alan Q., Liu, Fengbei, Sabuncu, Mert R.
Deep learning models can extract predictive and actionable information from complex inputs. The richer the inputs, the better these models usually perform. However, models that leverage rich inputs (e.g., multi-modality) can be difficult to deploy wi
Externí odkaz:
http://arxiv.org/abs/2405.20448
We present a keypoint-based foundation model for general purpose brain MRI registration, based on the recently-proposed KeyMorph framework. Our model, called BrainMorph, serves as a tool that supports multi-modal, pairwise, and scalable groupwise reg
Externí odkaz:
http://arxiv.org/abs/2405.14019
Healthcare data often come from multiple sites in which the correlations between confounding variables can vary widely. If deep learning models exploit these unstable correlations, they might fail catastrophically in unseen sites. Although many metho
Externí odkaz:
http://arxiv.org/abs/2310.15766
Autor:
Wang, Alan Q., Karaman, Batuhan K., Kim, Heejong, Rosenthal, Jacob, Saluja, Rachit, Young, Sean I., Sabuncu, Mert R.
Interpretability for machine learning models in medical imaging (MLMI) is an important direction of research. However, there is a general sense of murkiness in what interpretability means. Why does the need for interpretability in MLMI arise? What go
Externí odkaz:
http://arxiv.org/abs/2310.01685
Machine learning models will often fail when deployed in an environment with a data distribution that is different than the training distribution. When multiple environments are available during training, many methods exist that learn representations
Externí odkaz:
http://arxiv.org/abs/2309.13377
We present KeyMorph, a deep learning-based image registration framework that relies on automatically detecting corresponding keypoints. State-of-the-art deep learning methods for registration often are not robust to large misalignments, are not inter
Externí odkaz:
http://arxiv.org/abs/2304.09941
Autor:
Wang, Alan Q., Sabuncu, Mert R.
In this paper, we empirically analyze a simple, non-learnable, and nonparametric Nadaraya-Watson (NW) prediction head that can be used with any neural network architecture. In the NW head, the prediction is a weighted average of labels from a support
Externí odkaz:
http://arxiv.org/abs/2212.03411
Deep learning based techniques achieve state-of-the-art results in a wide range of image reconstruction tasks like compressed sensing. These methods almost always have hyperparameters, such as the weight coefficients that balance the different terms
Externí odkaz:
http://arxiv.org/abs/2202.11009
The convolutional neural network (CNN) is one of the most commonly used architectures for computer vision tasks. The key building block of a CNN is the convolutional kernel that aggregates information from the pixel neighborhood and shares weights ac
Externí odkaz:
http://arxiv.org/abs/2202.02701