Scribd
Upload
36 views2 pages

RA Selection Task

This project involves bone segmentation in 3D CT images, specifically focusing on the knee region. Tasks include segmenting the femur and tibia, expanding the segmented mask, randomizing the contour, and detecting key anatomical landmarks on the tibia. Submissions must be made via GitHub with a complete codebase, results folder, README file, and adherence to software development practices.

Uploaded by

Aakrit Dongol
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
36 views2 pages

RA Selection Task

This project involves bone segmentation in 3D CT images, specifically focusing on the knee region. Tasks include segmenting the femur and tibia, expanding the segmented mask, randomizing the contour, and detecting key anatomical landmarks on the tibia. Submissions must be made via GitHub with a complete codebase, results folder, README file, and adherence to software development practices.

Uploaded by

Aakrit Dongol
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Overview of task.

This project focuses on bone segmentation in 3D CT images volume. Your task is to process a specific
bone region as outlined in the task description and ultimately identify a key anatomical landmark. We have
provided a CT scan (data) of the knee region (refer to Figure 1).

You are required to upload your work to a GitHub repository and share the link with us. Submissions in PDF
format will not be accepted. Please ensure the repository includes:

• The complete codebase, organized in a modular format.


• A results folder containing:
o Output images
o A text file with relevant details
• A well-written README file
• GitHub commit history
• Any other necessary files to support reproducibility and understanding

Please note that we will evaluate not only the final solution but also the overall quality of your work,
including code structure, readability, and good software development practices. Be sure to follow proper
coding standards and documentation guidelines throughout your implementation.

Links

Data link:
https://drive.google.com/file/d/1NR7OEboARP_fpseIZOY0Wy8Ir1NEYfL5/view?usp=drive_link

Submission Form: https://forms.gle/NyUZrbRFshFvYd4y7


Task 1.1 – Bone Segmentation
Your first task is to segment the femur and tibia regions from the provided CT image, as illustrated in Figure
1.B and save as nii.gz format. Apply only image processing techniques.

Task 1.2 – Contour Expansion


The second task requires you to expand the segmented mask by 2 mm uniformly outward. 2mm should be
parameter.

Task 1.3 – Randomized Contour Adjustment


The third task involves randomizing the expanded segmented mask such that:

• The new mask lies between the original segmentation mask and the 2 mm expanded mask.
• The randomized mask must not exceed the 2 mm expansion limit.
• The contour must not shrink below the original segmentation contour.
• 2 mm and random value should be parameter.

Task 1.4 – Landmark Detection on Tibia


The final task focuses on the tibia. You are required to:

• You are required to save the following segmentation masks in .nii.gz file:
o Original Mask
o 2 mm Expanded Mask
o 4 mm Expanded Mask
o Randomized Mask 1
o Randomized Mask 2
• Identify the medial and lateral lowest points on the tibial surface, as demonstrated in Figure 1.D for
all the above 5 masks. Please note that Figure 1.D is provided for illustration purposes only. You are
not required to generate this image as part of your submission. Our team has tools for this
visualization.
• Submit the coordinates of these points for evaluation we will compare your results with the ground
truth.

You might also like