MadhumitaSushil/Biomedical_Clinical_LLMs

Biomedical and clinical LLMs for language and multimodal inference

Encoder-only language models

Publicly Available:

• BioBERT
• clinicalBERT
• SciBERT
• PubmedBERT
• BlueBERT
• BioLM
• Clinical-Longformer
• Clinical-BigBird
• GatorTron
• DrBERT: A Robust Pre-trained Model in French for Biomedical and Clinical domains
• BioSimCSE
• BioLinkBERT

Proprietary:

• CancerBERT
• UCSF-BERT
• NYUTron

Encoder-decoder language models

Publicly Available:

• Clinical-T5

Proprietary:

Decoder-only language models

Publicly Available:

• clinicalXLNet
• BiomedLM
• BioGPT
• PMC-Llama
• MedAlpaca
• Clinical Camel
• ChatDoctor
• MediTron
• Med42
• BioMistral: A Collection of Open-Source Pretrained Large Language Models for Medical Domains
• Me LLaMA: Foundation Large Language Models for Medical Applications
• Palmyra-Med
• Taiyi
• JSL-MedMNX-7B-SFT
• JSL-MedMNX-7B
• Internist.ai 7B
• Saama AI OpenBioLLM-8B
• Saama AI OpenBioLLM-70B

Models trained on MIMIC but unavailable:

• DRG-LLaMA
• Clinical Llama-LORA

Proprietary:

• GatorTronGPT
• MedPalm
• MedPalm2
• Polaris by HippocraticAI

Multimodal models

Publicly Available:

• BiomedCLIP
• PMC-VQA | MedVInt
• LLaVA-Med
• MedFlamingo
• XRayGPT
• BiomedGPT
• KAD
• CONCH Computational pathology model

Proprietary:

• ELIXR
• Flamingo-CXR
• MedPalm-M

Structured data models

Publicly-available:

• EHR-SHOT | CLMBR

Proprietary:

• EHRBert
• MedBERT
• RareBERT
• CEHR-Bert
• Claim-PT
• BEHRT: Transformer for Electronic Health Records

Application-specific models

• Stanford-Penn MIDRC Deidentifier

Models using synthetic data

• Asclepius
• GatorTron-S