Introduction: Image Processing of Brain Tumors From MRIs

Learn to detect tumor regions from MRI scans using this step-by-step process. Use MatLAB to process images of the brain and to locate the dimensions of a tumor that can be used for further analysis.

Step 1: Convert MRI Scan to Gray Scale

Convert the MRI scan to gray scale to get rid of any apparent color. The program will be able to detect which individual pixels are considered to be the tumor and which are not by labeling each pixel with a binary code.

Step 2: Sharpen Image Using High Pass Filter

Apply a high pass filter to the image so pixels with a higher brightness will be intensified over its neighbors. In other words, this form of filtering will amplify the tumor pixels but may also increase the noise of the overall image at the same time.

Step 3: Remove Noise Using Median Filter

The median filter will help remove the excess noise that was created with the high pass filter. The noise will be removed while still preserving the edges of the sharpened boundaries found from the previous step.

Step 4: Apply IMHIST and Find Threshold Value

Use built-in function "imhist" to plot a histogram of the intensity image and to determine the best threshold value. Select a threshold value depending on the slope of the curve and apply it to the filtered image. The goal is to have the output be the tumor alone so that it can be analyzed easily.

Step 5: Output Tumor Region

The final image will result in an MRI scan with the tumor region highlighted.