Figure 3 from Merlin: A Vision Language Foundation Model for 3D Computed Tomography | Semantic Scholar (2024)

[PDF] Semantic Reader

I3D models considerably improve upon the state-of-the-art in action classification, reaching 80.2% on HMDB-51 and 97.9% on UCF-101 after pre-training on Kinetics, and a new Two-Stream Inflated 3D Conv net that is based on 2D ConvNet inflation is introduced.

The first large-scale application of the phenome-wide association study (PheWAS) paradigm within electronic medical records (EMRs) is reported, an unbiased approach to replication and discovery that interrogates relationships between targeted genotypes and multiple phenotypes in EMR-based cohorts.

PMC-15M is a novel dataset that is two orders of magnitude larger than existing biomedical multimodal datasets such as MIMIC-CXR, and spans a diverse range of biomedical image types, and pretrained BiomedCLIP, a multimodal foundation model with domain-specific adaptations tailored to biomedical vision-language processing.

This research highlights the need to understand more fully the role that language plays in the development of identity and how language and identity politics play a role in the creation of identity.

A two-stage paradigm consisting of noninvasive cine-based CVD screening followed by cine and late gadolinium enhancement-based diagnosis is proposed, which holds the potential to substantially advance the efficiency and scalability of CMR interpretation, thereby improving CVD screening and diagnosis.

This work conducts extensive experiments around pre-trained models, architecture choice, data, and training methods, and develops Idefics2, an efficient foundational VLM of 8 billion parameters that achieves state-of-the-art performance within its size category across various multimodal benchmarks.

EchoCLIP is a vision–language foundation model for echocardiography that learns the relationship between cardiac ultrasound images and the interpretations of expert cardiologists across a wide range of patients and indications for imaging.

An open-source and interpretable AI algorithm accurately detects contrast phases in abdominal CT scans, with high accuracy and F1 scores in internal and external validation, confirming its generalization capability.

This study introduces CT-RATE, the first 3D medical imaging dataset that pairs images with textual reports, and developed CT-CLIP, a CT-focused contrastive language-image pre-training framework that outperforms state-of-the-art, fully supervised methods in multi-abnormality detection across all key metrics, thus eliminating the need for manual annotation.

...

...