Skip to content

Latest commit

 

History

History
34 lines (28 loc) · 966 Bytes

README.md

File metadata and controls

34 lines (28 loc) · 966 Bytes

Introduction

Using discrete wavelet transform for feature extraction of CT medical images. We aim to identify outliers that may be caused by medical discrepencies, poor calibration of the machine, or other outliers.

Data Set

CT scans of lungs
• dicom_dir contains 100 CT scans in the DICOM format
• tiff_images contains the same 100 CT scans in the TIFF format

Files

dwt_wavelet_name.m
• Simple dwt reconstruction of a wave. Plotted with labels

dwt_wavelet_and_scaling_features.m
• Adds scaling features to the previous file

dwt_image.m
• First attempt at using dwt functionality on an image

dwt_5_layers_disp.m
• Explored further extraction on one of the images from the dataset
• Display all of the layers that were extracted

scratchwork.m
• Where I work on the next step of the process.
• Currently using it to try and format a 2D array to store all of the sub-images for each DICOM file