Medical image segmentation basics in Computer Vision - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is medical image segmentation?

Medical image segmentation is the process of dividing a medical image into meaningful parts, like separating organs or tumors from the background, to help doctors analyze and diagnose.

beginner

Why is medical image segmentation important?

It helps doctors see exact shapes and sizes of organs or abnormalities, making diagnosis and treatment planning more accurate and faster.

intermediate

Name two common types of medical image segmentation methods.

Two common methods are: 1) Thresholding, which separates parts based on pixel brightness, and 2) Deep learning models like U-Net, which learn to segment images automatically.

intermediate

What is a U-Net in medical image segmentation?

U-Net is a special deep learning model shaped like a U that helps computers learn to find and outline important parts in medical images with high accuracy.

intermediate

What metric is commonly used to measure segmentation accuracy?

The Dice coefficient is often used. It measures how much the predicted segmentation overlaps with the true area, with 1 meaning perfect overlap and 0 meaning no overlap.

What does medical image segmentation help doctors do?

A Increase image brightness

B Change image colors

C Separate important parts like organs or tumors in images

D Compress image size

Medical image segmentation helps separate meaningful parts in images to assist diagnosis.

Which model is popular for automatic medical image segmentation?

A U-Net

B Linear Regression

C Decision Tree

D K-Means Clustering

U-Net is a deep learning model designed for image segmentation tasks.

What does the Dice coefficient measure?

A Number of image pixels

B Image brightness

C Model training speed

D Overlap between predicted and true segmentation

Dice coefficient measures how well the predicted segmentation matches the true area.

Which of these is NOT a segmentation method?

A Histogram Equalization

B U-Net

C Thresholding

D Region Growing

Histogram Equalization is an image enhancement technique, not segmentation.

Why is segmentation useful in medical images?

A To add noise to images

B To identify and analyze specific structures

C To reduce file size

D To make images colorful

Segmentation helps identify and analyze specific parts like organs or tumors.

Explain in your own words what medical image segmentation is and why it matters.

Think about how doctors use images to find problems.

Describe the role of deep learning models like U-Net in medical image segmentation.

Consider how computers learn to find parts in images.