# Medical Image Analysis

MedGemma-4B-IT is a vision-language model specialized in the medical field, developed by Google and now converted to MLX format for efficient operation on Apple chips.

MedGemma-4B-it-Q8_0-GGUF is a GGUF format model converted from google/medgemma-4b-it, specifically designed for image-to-text tasks in the medical field.

This model is fine-tuned based on the DINOv2 architecture for disease classification of skin lesion images

PyTorch-based DeepLabV3Plus image segmentation model supporting efficient semantic segmentation tasks

This model is a GGUF format conversion of mlabonne/gemma-3-4b-it-abliterated, combined with the visual component of x-ray_alpha for a smoother multimodal experience.

AKI is a multimodal foundation model that achieves cross-modal mutual attention (MMA) by unlocking the causal attention mechanism in LLMs, addressing vision-language misalignment without additional parameters or training time.

A fine-tuned model based on Vision Transformer (ViT) architecture for classifying chest X-rays, trained on the CheXpert dataset.

An ensemble model for predicting breast cancer and breast density from screening mammograms, using 3 CNN networks with different resolutions for inference

EraX-VL-7B-V2.0-Preview is a powerful multimodal model designed for OCR and visual question answering, excelling in processing multiple languages including Vietnamese, with outstanding performance in recognizing medical forms, invoices, and other documents.

GenMedClip is a zero-shot image classification model based on the open_clip library, specializing in medical image analysis.

A PyTorch-implemented FPN image segmentation model that supports various encoder architectures, suitable for semantic segmentation tasks.

Linknet is a PyTorch-implemented image segmentation model suitable for semantic segmentation tasks.

An image classification model fine-tuned on the BrainMRI dataset based on Google's ViT base model

A multimodal biomedical model fine-tuned based on Llama-3-8B-Instruct, supporting text and image processing, suitable for biomedical research and clinical applications.

A lung cancer detection model fine-tuned based on Florence-2-base-ft, identifying lung cancer types through lung images

This model is used for bone age prediction, built upon the YassinHegazy/xray-model base model.

Virchow is a self-supervised vision Transformer pretrained on 1.5 million whole-slide histopathology images, serving as a slide-level feature extractor for computational pathology downstream tasks.

M3D-CLIP is a CLIP model specifically designed for 3D medical imaging, achieving visual and language alignment through contrastive loss.

This model can convert medical images (such as X-rays) into descriptive text to assist in medical diagnosis.

A deep learning model based on ViT architecture for identifying pneumonia symptoms in chest X-ray images

A facial image classification model using Vision Transformer (ViT) architecture for detecting dry, normal, and oily skin types

The AIMI Foundation Model Suite is a collection of foundation models for the radiology domain developed by the Stanford AIMI team, focusing on medical image analysis tasks.

A skin disease classification model fine-tuned based on the DINOv2 base model, achieving 95.57% accuracy on the ISIC 2018+Atlas Dermatology dataset.

Breast cancer segmentation model based on Segment Anything Model (SAM), used for tumor region identification in medical imaging

A retinal fundus image segmentation model based on the SegFormer architecture, specifically designed for precise segmentation of the optic disc and cup.

SkinSAM is a skin lesion segmentation model based on a 12-layer ViT-b architecture, fine-tuned on ISIC and PH2 datasets, focusing on precise segmentation of skin lesion images.

A dental X-ray image segmentation model fine-tuned based on the MIT-B0 architecture, specifically designed for precise segmentation of tooth regions in dental imaging

PubMedCLIP is a version of the CLIP model fine-tuned for the medical field, specifically designed to handle medical images and related text.

A Parkinson's disease prediction model based on the EfficientNet architecture, achieving approximately 83% accuracy by analyzing patient drawings

This model is a fine-tuned version of Google's ViT-base model on a chest X-ray classification dataset, designed for medical image analysis.

An image classification model fine-tuned based on google/vit-base-patch16-224, suitable for histology image analysis tasks.

ClipMD is a medical image-text matching model developed based on OpenAI's CLIP model, employing a sliding window text encoder specifically designed for medical image classification tasks.

A pneumonia detection model based on the ViT architecture, fine-tuned on a chest X-ray classification dataset with an accuracy rate of 97.68%

A fine-tuned model based on facebook/detr-resnet-50 for object detection tasks

An image classification model based on the Vision Transformer architecture, fine-tuned for BIRADS classification tasks

A brain tumor image classification model fine-tuned based on microsoft/resnet-50, achieving an accuracy of 91.71% on the evaluation set

A diabetic retinopathy classification model based on the Vision Transformer (ViT) architecture, achieving 72.87% accuracy on the evaluation set

A fine-tuned model based on the Swin Transformer architecture, specifically designed for skin cancer image classification tasks

A fine-tuned model based on Swin Transformer architecture, specifically designed for skin cancer image classification tasks

An image classification model fine-tuned on chest image datasets based on Google's ViT model, achieving 99% accuracy