A Three Dimensional Modeling and Real Time Data Analysis Algorithm for Bronchoscopy Enhancement
CSEF · 2017 Computational Systems & Analysis
Overview
Objectives/Goals The objective of the project is to create a 3D computer model of the lung using simply the video feed from the tip of a bronchoscopy tool with minimal hardware modifications. Methods/Materials Laptop with MATLAB R2016b. A yoga mat was rolled to form a dark tunnel that provided me with a simulated bronchial tube. Then, I attached a webcam and flashlight to one end of a long stick using the masking tape to act as the bronchoscopy tool. I connected the webcam to my laptop's USB port and used the image processing toolbox to process snapshots from the real time video feed and curve fitting toolbox for creating the splines for the model. Results As the stick was pushed into the yoga mat tube, a 3D model is created by adding the ellipses the algorithm had calculated every second. The resulting model is hollow and also shows texture similar to the inside of the lungs. After each iteration, a new layer that is generated is added to the existing model in order to save time and memory rather than creating a whole new model from scratch. The tool is also able to pick up bends in the tube, indicating that lateral motion can be considered as the bronchoscope moves through the lung. Conclusions/Discussion It is possible to create a 3D model of the lungs without exposing the patient and doctor to harmful radiation. Moreover, one does not need to follow the traditional methods of 3D reconstruction. It is possible to use only a video feed without a depth sensor to create a streamlined process that produces a 3D model in real time. Through this project, I have accomplished 3 goals: Image Conversion to 3D Model, Real Time Image Processing, and the generated 3D model shows what is ahead. This algorithm can be used by doctors as a tool to prevent repetitive bronchoscopies as well as for the future of robotic endoscopic surgeries in which computer vision software will be paramount.
Summary statement
I created an algorithm to process and analyze the video feed from the bronchoscopy tool to create a real time 3D model of the lung's airways without the need for radiation.
Help received
None. I designed, coded, and implemented the algorithm and analyzed data myself.
Competition history
- CSEF 2017
Resources
Related projects
ISEF · 2015
It's a Matter of Life and Breath: A Novel Computer Diagnosis Tool for Early Detection of Lung Cancer SPNs towards Big Data Analysis
ISEF · 2019
Using Three-Dimensional Modeling to Analyze the Vascular System and Radiation-Induced Lung Damage
CSEF · 2018
A Machine Learning Based Approach to Decrease the Lung Cancer Malignancy Detection Threshold
ISEF · 2019
Utilizing Computer Vision and Machine Learning Systems to Develop a Live Time Navigational and Surgical Aid for Spinal Reconstructions
ISEF · 2019
Body Anomaly Detection through 3D Body Scanning, Image Processing, and Machine Learning
AJAS · 2017
Stereoscopic X-Ray Reconstruction Algorithms for Preventing Radiation Induced Cancer
CSEF · 2012
An Automated Process of Calculating Bone Mineral Density Using 3D Image Processing Algorithms
CSEF · 2014
Complete 3D Projection Using Head-Tracking and Stereoscopy
Closest projects by meaning, across every fair and year in the corpus.
Browse more like this
Source: California Science & Engineering Fair public projects