Search papers, labs, and topics across Lattice.
4
0
5
11
Curia-MAE achieves superior performance in 3D medical image segmentation with a frozen encoder, challenging the need for extensive fine-tuning even in data-scarce environments.
Jolia not only sets a new state of the art in 3D medical imaging but also reveals that localized concept alignment can dramatically enhance model performance in complex tasks.
Out-of-domain self-supervised pretraining on brain MRIs beats in-domain supervised learning when generalizing to real-world clinical data.
Billion-parameter Vision Transformers can now be effectively pre-trained on multi-modal CT and MRI data, achieving state-of-the-art performance on vision-focused radiology tasks.
