MedicalSegmentationBase<T>

Base Classes · AiDotNet.ComputerVision.Segmentation.Common

Abstract base class for medical image segmentation models handling 2D slices and 3D volumes.

Inheritance: NeuralNetworkBase<T> → SegmentationModelBase<T> → MedicalSegmentationBase<T>

Implements: IMedicalSegmentation<T>

Known derived types: BiomedParse<T>, MedNeXt<T>, MedSAM2<T>, MedSAM<T>, MedSegDiffV2Segmentation<T>, MedSegDiffV2<T>, MedicalSAM2<T>, NnUNet<T>, SegMamba<T>, SwinUNETR<T>, TransUNet<T>, UMamba<T>, UniverSeg<T>

For Beginners

Medical segmentation helps doctors by automatically outlining organs, tumors, and other structures in medical images. These models handle special requirements like 3D volumetric processing (CT/MRI scans are stacks of slices), multi-modal imaging, and the very high accuracy needed for clinical use.

Models extending this base class: nnU-Net, TransUNet, Swin-UNETR, MedSAM, MedSAM 2, MedNeXt.

Properties

Property Summary
IReadOnlyList<string> SupportedModalities { get; }
bool Supports2D { get; }
bool Supports3D { get; }
bool SupportsFewShot { get; }

Methods

Method Summary
MedicalSegmentationResult<T> SegmentFewShot(Tensor<T> queryImage, Tensor<T> supportImages, Tensor<T> supportMasks)
MedicalSegmentationResult<T> SegmentSlice(Tensor<T> slice)
MedicalSegmentationResult<T> SegmentVolume(Tensor<T> volume)