2D Brain MRI segmentation using UNet - s4641238

[s4684]

Student ID: 46412384
Project: 1 - Segment the 2D OASIS brain data set with an Improved UNet

Pull Summary

This pull implements a 2D Improved UNet model to segment OASIS brain MRI data slices.

Key Additions

dataset.py - Loads and processes the OASIS dataset
modules.py - Defines the UNet model architecture
train.py - Performs training of the model
predict.py - Evaluates the trained model

[24msingh24]

This is an initial inspection, no action is required at this point

2D UNet – Prostate MRI Segmentation → Easy

Category		Marks	Comments
Algorithm solves the problem	5	4	Implementation looks fine. Dice score seems low.
Implementation functions as intended	3	3	Training and test pipeline works as intended.
Good design	1	1	UNet blocks and encoder-decoder modularity are fine
Commenting	1	0.5	Code has some comments, but they are often verbose or generic. Some comments are unclear (e.g., “compromise on computational efficiency and model complexity while trying to replicate the sigmoid function” is misleading because ReLU does not replicate sigmoid). Missing comments on key operations in dataset.py and evaluation functions.
Algorithm above Normal Difficulty	5	0
Algorithm is Hard difficulty	5	0	Easy
Section IV : Max mark 10 from 20	8.50

• Discussion: None provided. This is needed.

Suggestions/Notes:

• Dice Score can be improved.

[gayanku]

Marking

Good/OK/Fair Practice (Design/Commenting, TF/Torch Usage)
	No design and implementation.	-2
	Spacing and comments.
	No Header blocks.	-1
Recognition Problem
	Good solution to problem.
	Driver Script present.
	File structure present.
	Good Usage & Demo & Visualisation & Data usage.
	Module present.
	Commenting missing.	-1
	No Data leakage found.
	Difficulty : Easy. Easy. Unet 2D	-10
Commit Log
	Good Meaningful commit messages.
	Good Progressive commits.
Documentation
	Readme :Good.
	Model/technical explanation :Good.
	Description and Comments :Good.
	Markdown used. PDF NOT submitted.	-2
Pull Request
	Successful Pull Request (Working Algorithm Delivered on Time in Correct Branch).
	No Feedback required.
	Request Description is good.
TOTAL		-16

Marked as per the due date and changes after which aren't necessarily allowed to contribute to grade for fairness.
Subject to approval from Shakes