Segmentation and Detection

Segmentation and Detection

Tumor Detection and Segmentation in DICOM Tumor Detection:

The process starts by detecting the presence of a tumor inside the DICOM slices.

Detection can be done using image processing or AI-based algorithms .

Once a suspicious region(tumor) is found, a bounding box is drawn around it.

The bounding box acts as a rough boundary that helps isolate the tumor from the rest of the tissues.

1. 2D Segmentation:

After detection, the next step is to perform segmentation on each 2D slice of the DICOM dataset.

In this step, we carefully outline the tumor within the bounding box to separate it from surrounding tissue.

This can be done manually or automatically using thresholding, edge detection, or other AI based algorithms.

The goal is to create accurate segments in the 2D slices that highlight the tumor’s exact shape in each slice.

2. Segmentation in  3D:

At this stage, the complex thing would be is to build a complex polygonal mesh/object     that represents the tumor shape in 3D or use a “shrink-wrap” method that wraps a smooth outer layer around the segmented region to form a realistic 3D model.