All classes Brain Tumor Segmentation with U-Net Syllabus · Episode 1 of 6

Class U-Net Brain Tumor Segmentation #1: Brain MRI Dataset & Tumor Mask: Loading, Visualization & Patient-wise Splitting

7 views

Medical image segmentation tutorial: brain tumor segmentation from MRI using optimized metaheuristic U-Net. Episode 1: loading LGG MRI dataset, tumor mask, overlay, and patient-wise data splitting.

Software: Deep Learning (PyTorch U-Net)
Level: Intermediate
Duration: ± 12 min
Series total: 1 h 10 min · 6 videos
Published: October 24, 2026

Episode 1 of 6 · view syllabus

Download PDF (8 PDF pages)

0 downloads

Download Project File· 95 MB Unlock All Premium Classes ▶ Watch the video lesson0 plays

▶ Video Lesson

Watch every step above performed live, with English narration.

0 plays

Class book

Page 1 / 8

Class U-Net Brain Tumor Segmentation #1: Brain MRI Dataset & Tumor Mask: Loading, Visualization & Patient-wise Splitting: Cover: Class U-Net Brain Tumor Segmentation #1: Brain MRI Dataset & Tumor Mask: Loading, Visualization & Patient-wise Split

Class U-Net Brain Tumor Segmentation #1: Brain MRI Dataset & Tumor Mask: Loading, Visualization & Patient-wise Splitting: Before you start, steps list, and Step 1

Class U-Net Brain Tumor Segmentation #1: Brain MRI Dataset & Tumor Mask: Loading, Visualization & Patient-wise Splitting: Step 2: Function to load data per patient

Class U-Net Brain Tumor Segmentation #1: Brain MRI Dataset & Tumor Mask: Loading, Visualization & Patient-wise Splitting: Step 3: Image and mask size

Class U-Net Brain Tumor Segmentation #1: Brain MRI Dataset & Tumor Mask: Loading, Visualization & Patient-wise Splitting: Step 4: Tumor pixel imbalance

Class U-Net Brain Tumor Segmentation #1: Brain MRI Dataset & Tumor Mask: Loading, Visualization & Patient-wise Splitting: Step 5–6: Patient-wise data splitting

Class U-Net Brain Tumor Segmentation #1: Brain MRI Dataset & Tumor Mask: Loading, Visualization & Patient-wise Splitting: Step 7: Tumor Area Statistics

Class U-Net Brain Tumor Segmentation #1: Brain MRI Dataset & Tumor Mask: Loading, Visualization & Patient-wise Splitting: Done: Final Result and Further Practice

Cover: Class U-Net Brain Tumor Segmentation #1: Brain MRI Dataset & Tumor Mask: Loading, Visualization & Patient-wise Split

Click a page or use ◀ ▶ to turn · swipe on mobile · to enlarge

Want to go deeper?

Join one of our courses or ask us directly on WhatsApp.

Download PDF· 0 downloads View the full 6-episode syllabus View Medical Image Segmentation Courses Ask on WhatsApp

Recommended classes for you

Picked from the software, topic, and level of “Class U-Net Brain Tumor Segmentation #1: Brain MRI Dataset & Tumor Mask: Loading, Visualization & Patient-wise Splitting”.

Deep Learning CNN-LSTM Pneumonia Detection

6 episodesFree preview

Python · Intermediate–Advanced

Deep Learning CNN-LSTM Pneumonia Detection

6 video lessons with PDF guides, 57 min in total. Starts with: Chest X-ray Dataset: Loading, Exploration & Data Splitting.

57 min 22 0

View class

6 episodesFree preview

Python · Intermediate–Advanced

Transfer Learning for Leaf Disease

6 video lessons with PDF guides, 1 h 5 min in total. Starts with: PlantVillage Dataset: Folder Structure, Class Distribution & Data Split.

1 h 5 min 9 0

View class

ANFIS Metaheuristics Time Series Forecasting

6 episodesFree preview

Python · Intermediate–Advanced

ANFIS Metaheuristics Time Series Forecasting

6 video lessons with PDF guides, 59 min in total. Starts with: Non-Linear Time Series: Dataset, Stationarity, and Patterns.

59 min 13 0

View class

50 episodesFree preview

Agentic AI · Beginner–Intermediate

Agentic AI Without Coding

50 video lessons with PDF guides, 14 h 29 min in total. Starts with: Chatbots, Automation, and AI Agents: What's the Difference?.

14 h 29 min 66 0

View class

100 episodesFree preview

Claude · Beginner–Advanced

Claude from Zero to Expert

100 video lessons with PDF guides, 15 h 1 min in total. Starts with: What Claude Is and How It Differs from ChatGPT and Gemini.

15 h 1 min 62 0

View class

EKG Arrhythmia Classification Wavelet Deep Learning

6 episodesFree preview

Python · Intermediate–Advanced

EKG Arrhythmia Classification Wavelet Deep Learning

6 video lessons with PDF guides, 56 min in total. Starts with: EKG & EEG Signals: Characteristics and Public Datasets (MIT-BIH).

56 min 2 0

View class

NSGA-II MOPSO Hybrid Multi-Objective Optimization

6 episodesFree preview

Python · Intermediate–Advanced

NSGA-II MOPSO Hybrid Multi-Objective Optimization

6 video lessons with PDF guides, 59 min in total. Starts with: Multi-Objective Optimization: Pareto Front, Dominance, and Test Functions ZDT/DTLZ.

59 min 9 0

View class

6 episodesFree preview

Python · Intermediate–Advanced

Occlusion Robust Face Recognition

6 video lessons with PDF guides, 1 h 26 min in total. Starts with: LFW Face Dataset & Creating Synthetic Occlusion (Mask, Sunglasses, Hand).

1 h 26 min 0 0

View class

Need more than a video?

Other ways we can help with Python.

Learn with a mentor

Scheduled AI & Machine Learning courses with live guidance and a certificate.

Course

Multi-Objective Optimization Course: NSGA-II, MOPSO & Hybrid Algorithms (Python)

Course

ECG Arrhythmia Classification Course with Wavelet & Deep Learning (PyTorch)

Course

ANFIS + PSO/GA Course for Non-Linear Time Series Forecasting (Python)

See all in AI & Machine Learning →

Have it built for you

No time to learn Python? Our team can do the project for you.

Service

Python Machine Learning — Prediction Models, Customer Segmentation, Fraud Detection & AI Apps

Service

AI Image Creation with Stable Diffusion — Unlimited Custom Visual Production

★ 4.9

Service

AI Image & Video Creation with Leonardo AI — All-in-One Visuals for Business

★ 4.9

See all in AI & Machine Learning →

Get notified about new Python classes

At most one email a week. Unfollow anytime with one click.

Transcript

The full narration of this video. The current sentence is highlighted while the video plays; click any sentence to jump there.

Show transcript

Introduction

This is the final result, MRI brain images from 110 patients, complete with tumor masks drawn by doctors, and yellow lines marking the tumor boundaries on top of the images. This is the first step of our series in building a metaheuristic optimized U-Net for brain tumor segmentation. Let's get started. Welcome to the AI research tutorial series, brain tumor image segmentation with a metaheuristic optimized U-Net. Unlike classification, which only answers whether there is a tumor or not, segmentation marks which pixels are part of the tumor, so doctors know the location and size. In this 6-episode series, we will discuss the MRI dataset, preprocessing and augmentation, building the U-Net from scratch, Dice loss and IOU, hyperparameter optimization with particle swarm optimization and gray wolf optimizer, all the way to evaluation. In this first episode, we will get to know the data.

Step 1: Installing library & dataset

We open the terminal through the terminal menu, new terminal, then install the required libraries, torque, open python, scikit-learn, and matplotlib. The dataset is LGGMRI segmentation from RSIB TCGA, MRI brain images of 110 patients with low-grade glioma, with manually drawn tumor masks. We have already downsized it to 128 pixels in the dataset folder, one subfolder per patient. This folder is available in the customer's premium project file.

Step 2: Function to load data per patient

Create a file named data_underscore_brain_dot_py. The load_patient function reads all MRI slices of a patient along with their mask, then resizes them to 96 pixels. Note that the images are downsized with bilinear interpolation, while the mask uses nearest neighbor interpolation so that the values remain 0 or 1. The load_all function combines all patients and saves the compressed CNPZ files, so that the next loading is much faster. The most important function is the split_patient function. The data is split into 70, 15, 15 percent based on patients, not on slices.

Step 3: Image and mask size

Now create the file exploration_underscore_brain_dot_py and run it. There are 110 patients with a total of 3,929 slices. Each image is 96x96 with 3 channels, which are 3 MRI sequences: pre-contrast, flair, and post-contrast. The mask only contains 0 and 1.

i Step 4: Tumor pixel imbalance

Let's count how many slices contain a tumor. Only 1,373 slices, about 35 percent. And among the tumor-containing slices, the average tumor area is only 2.9 percent of the image. This is the main challenge of medical segmentation, the tumor pixels are very few compared to the background. A model that always predicts no tumor will appear accurate pixel-wise, although it is not useful. Therefore, we will use the DAIS and IOU metrics later. Look at the data split.

Step 5: Data splitting per patient

77 patients for training, 16 for validation, and 17 for testing. Why per patient? Slices from the same patient are very similar. If split by slice, the model seems to see exam questions while studying, and the score becomes overly optimistic. This is a common mistake found in descriptions, so make sure your data split is per patient.

Step 6: Mask visualization and overlay

Now, visualization. The first row is the MRI image, the second row is the tumor mask in white, and the third row is the overlay, with a yellow line marking the tumor boundary over the image. The tumor appears as a bright area in the flair sequence, with highly variable shapes and sizes. This boundary line is what UNET will later need to draw. Finally, two graphs: the number of slices without and with tumors, and a histogram of tumor area.

Step 7: Tumor area statistics

Most tumors occupy less than 5% of the image, only a few are large. This distribution needs to be remembered when reading evaluation results, since small tumors are much harder to segment.

Done!

Done. We have now become familiar with the LGGMRI dataset, loaded images and masks per patient, understood the tumor pixel imbalance, and split the data without leakage. In the next episode, we will perform MRI preprocessing and image augmentation along with the mask. Currently working on a thesis, dissertation, or AI research in the field of medical image segmentation? Cerdas Inovasi Digital helps with deep learning implementation, from UNET segmentation, metaheuristic optimization with PSO and Grey Wolf, up to DAIS, IOU evaluation, result writing, and opening deep learning courses with PyTorch. Want to master the entire series, from brain MRI datasets to optimized UNET? All 6 episodes, complete with step-by-step PDF guides and source code, are available for premium tutorial subscribers. Subscribe starting at Rp69.000 per month, save 28% with the 12-month package, or pay once Rp999.000 for lifetime access. Want just one class? Buy single class with lifetime access. Subscribe now on the premium page at cerdasinovasidigital.com Contact our WhatsApp or visit cerdasinovasidigital.com Thank you for watching, see you in the next episode.

Ask the class material

Type a question or keyword; we find the part of this episode that explains it and jump the video there.

My notes

Notes are saved with the current video time and are only visible to you.

Questions & answers

Stuck on a step? Ask here; the instructor and other learners can answer.

No questions yet.

Your instructor

Cerdas Inovasi Digital Team

Software developers, BIM & 3D engineers, and AI practitioners

53 video classes1,296 episodes

We build software, BIM models, games and AI automations for clients every day, and turn that hands-on work into step-by-step classes: every step is shown on screen and comes with a PDF guide you can keep.

See our clients →

Memuat

▶ Video Lesson

Recommended classes for you

Need more than a video?

Transcript

0:03Introduction

0:59Step 1: Installing library & dataset

1:32Step 2: Function to load data per patient

4:53Step 3: Image and mask size

6:08i Step 4: Tumor pixel imbalance

7:04Step 5: Data splitting per patient

7:43Step 6: Mask visualization and overlay

9:27Step 7: Tumor area statistics

10:43Done!

Ask the class material

My notes

Questions & answers

Your instructor

▶ Video Lesson

Recommended classes for you

Need more than a video?

Transcript

0:03Introduction

0:59Step 1: Installing library & dataset

1:32Step 2: Function to load data per patient

4:53Step 3: Image and mask size

6:08i Step 4: Tumor pixel imbalance

7:04Step 5: Data splitting per patient

7:43Step 6: Mask visualization and overlay

9:27Step 7: Tumor area statistics

10:43Done!

Ask the class material

My notes

Questions & answers

Your instructor

Introduction

Step 1: Installing library & dataset

Step 2: Function to load data per patient

Step 3: Image and mask size

i Step 4: Tumor pixel imbalance

Step 5: Data splitting per patient

Step 6: Mask visualization and overlay

Step 7: Tumor area statistics

Done!

Introduction

Step 1: Installing library & dataset

Step 2: Function to load data per patient

Step 3: Image and mask size

i Step 4: Tumor pixel imbalance

Step 5: Data splitting per patient

Step 6: Mask visualization and overlay

Step 7: Tumor area statistics

Done!