All classes Brain Tumor Segmentation with U-Net Syllabus · Episode 3 of 6

Class Brain Tumor Segmentation with U-Net #3: Building U-Net from Scratch: Encoder, Decoder & Skip Connection

3 views

Tutorial building U-Net from scratch with PyTorch: encoder, bottleneck, decoder, skip connection, tensor shape inspection, and one batch overfitting test. Episode 3 of the brain tumor segmentation series.

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

Episode 3 of 6 · view syllabus

Download PDF (7 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 / 7

Class Brain Tumor Segmentation with U-Net #3: Building U-Net from Scratch: Encoder, Decoder & Skip Connection: Cover: Class U-Net Brain Tumor Segmentation #3: Building U-Net from Scratch: Encoder, Decoder & Skip Connection

Class Brain Tumor Segmentation with U-Net #3: Building U-Net from Scratch: Encoder, Decoder & Skip Connection: Before you start, steps list, and Step 1

Class Brain Tumor Segmentation with U-Net #3: Building U-Net from Scratch: Encoder, Decoder & Skip Connection: Step 2: Encoder, bottleneck, decoder

Class Brain Tumor Segmentation with U-Net #3: Building U-Net from Scratch: Encoder, Decoder & Skip Connection: Step 3: Skip connection in forward

Class Brain Tumor Segmentation with U-Net #3: Building U-Net from Scratch: Encoder, Decoder & Skip Connection: Step 4: Checking tensor shape

Class Brain Tumor Segmentation with U-Net #3: Building U-Net from Scratch: Encoder, Decoder & Skip Connection: Step 5: Overfit test one batch

Class Brain Tumor Segmentation with U-Net #3: Building U-Net from Scratch: Encoder, Decoder & Skip Connection: Done: Final Result and Further Practice

The end

Cover: Class U-Net Brain Tumor Segmentation #3: Building U-Net from Scratch: Encoder, Decoder & Skip Connection

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 Brain Tumor Segmentation with U-Net #3: Building U-Net from Scratch: Encoder, Decoder & Skip Connection”.

18 episodesFree preview

Python · Beginner–Advanced

Machine Learning Python

18 video lessons with PDF guides, 3 h 41 min in total. Starts with: Introduction to Machine Learning & First Model.

3 h 41 min 61 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

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

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

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

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

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

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, the UNET architecture we wrote ourselves with PyTorch, from 96 pixel images down to 12 pixel bottleneck, then up again to the segmentation map. And the overfit test proves our UNET can draw the tumor. Let's start. Hello, welcome to episode three of the brain tumor segmentation series with the metaheuristic optimized UNET. UNET was introduced in 2015 specifically for biomedical image segmentation, and until now it has become the base model in almost all medical segmentation research. Today we are building it from scratch, line by line, so that you truly understand how it works, then we will test it before real training.

Step 1: Double convolution block

Create a file UNET.py The basic component is the convolution block, 2x conv2d 3x3, each followed by BatchNorm and ReLU. Padding 1 keeps the image size the same. At the end of the block there is a 2D dropout which is by default 0, its value will later become one of the hyperparameters to be optimized.

Step 2: Encoder, bottleneck, decoder

Now the UNET class. Parameter F is the number of filters at the first level. The encoder has 3 blocks with filters F, 2F, and 4F, separated by Max Pooling which halves the image size. The bottleneck at the bottom of the U letter has 8F filters. The decoder is the opposite, conv transpose 2 which doubles the feature map size twice, then the convolution block. The last convolution layer 1x1 produces one channel, which is the tumor score for each pixel.

Step 3: Skip connection in forward pass

Look at the forward function. This is the key of UNET, skip connection. Before entering the decoder block, the sampled loop result is combined with the encoder features from the same level. The encoder knows what the image contains but loses location details due to pooling. Skip connection restores those location details, so the tumor boundary can be drawn sharply. Now create a file check_underscore_unit.py to check the tensor shape at each level.

Step 4: Checking tensor shape

Run. Input 3 channel 96 pixel. Encoder 1 produces 16 channels at the same size, then 32 channels at 48 pixel, 64 channels at 24 pixel, and bottleneck 128 channels at 12 pixel. Output returns to 96x96 with 1 channel. Total of 483 thousand parameters.

Step 5: Test overfitting on one batch

Before long training, a good researcher performs overfitting test. Train the model on a small 1B only. If the model cannot memorize 8 images, there must be a bug. We take 8 slices with the largest tumor, then train for 60 steps. Loss drops from 0.43 to 0.04. The image, top row is the original mask, bottom row is prediction. Location and shape of the tumor are already suitable, edges are still smooth because it's only 60 steps. It means our unit is working. Done. We have built the unit from scratch with encoder, bottleneck, decoder, and skip connection,

Done!

checking the tensor shape at each level, and passed the overfitting test. In the next episode, we will discuss the loss specific to segmentation. DAIS, IOU, and BCI DAIS, then fully train the baseline unit. 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, heuristic optimization with PSO and Grey Wolf, up to DAIS, IOU evaluation, and result writing, as well as opening deep learning courses with PyTor. Want to master the entire series, from MRI brain dataset to heuristic optimized UNET? All 6 episodes, complete with PDF step-by-step guides and source code, are available for premium tutorial subscribers. Subscribe starting from Rp69.000 per month Save 28% with the 12-month package, or pay once Rp999.000 for lifetime access. Want a single class only? Buy single classes with lifetime access. Subscribe now on the premium page 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

0:48Step 1: Double convolution block

2:04Step 2: Encoder, bottleneck, decoder

3:36Step 3: Skip connection in forward pass

4:36Step 4: Checking tensor shape

6:20Step 5: Test overfitting on one batch

9:00Done!

Ask the class material

My notes

Questions & answers

Your instructor

▶ Video Lesson

Recommended classes for you

Need more than a video?

Transcript

0:03Introduction

0:48Step 1: Double convolution block

2:04Step 2: Encoder, bottleneck, decoder

3:36Step 3: Skip connection in forward pass

4:36Step 4: Checking tensor shape

6:20Step 5: Test overfitting on one batch

9:00Done!

Ask the class material

My notes

Questions & answers

Your instructor

Introduction

Step 1: Double convolution block

Step 2: Encoder, bottleneck, decoder

Step 3: Skip connection in forward pass

Step 4: Checking tensor shape

Step 5: Test overfitting on one batch

Done!

Introduction

Step 1: Double convolution block

Step 2: Encoder, bottleneck, decoder

Step 3: Skip connection in forward pass

Step 4: Checking tensor shape

Step 5: Test overfitting on one batch

Done!