Scribd
Upload
6 views

Lecture0

The document outlines the course BIO3603: Medical Pattern Recognition, taught by Dr. Lamees Nasser, covering topics such as machine learning, pattern recognition techniques, and a course project focused on medical image classification using deep learning. It includes grading criteria, required textbooks, and details about the course project involving classification of various medical images. Students are expected to work in groups to adapt pretrained models and compare their performance using Python in Google Colaboratory.

Uploaded by

mohamedalbialy312
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
6 views

Lecture0

The document outlines the course BIO3603: Medical Pattern Recognition, taught by Dr. Lamees Nasser, covering topics such as machine learning, pattern recognition techniques, and a course project focused on medical image classification using deep learning. It includes grading criteria, required textbooks, and details about the course project involving classification of various medical images. Students are expected to work in groups to adapt pretrained models and compare their performance using Python in Google Colaboratory.

Uploaded by

mohamedalbialy312
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/ 13

BIO3603:Medical Pattern Recognition

Dr. Lamees Nasser


E-mail: [email protected]
Third Year– Biomedical Engineering Department
Academic Year 2024- 2025

10/8/2024 1
Expected Topics
• Introduction to machine learning and pattern recognition
• Pattern recognition techniques:
• Statistical methods
• Artificial neural network models
• Deep learning models
• Evaluation and validation
• Course project

10/8/2024 2
Textbooks:
• Geoff Dougherty, Pattern Recognition and Classification: An Introduction,
2012.

• Jianxin Wu, Essentials of Pattern Recognition: An Accessible Approach, 2020.

• Elgendy, Mohamed, Deep learning for vision systems. Simon and Schuster, 2020.

10/8/2024 3
Course Grading Criteria

Course basic information


Assignments 10
Final project presentation and report 20
Midterm exam 30
Final exam 90
Total 150

10/8/2024 4
Course Project: Medical Image Classification Using
Deep Learning

Chest X-ray classification


Normal

Pneumonia
Covid-19

10/8/2024 Giełczyk, Agata, et al. "Pre-processing methods in chest X-ray image classification." Plos one 17.4 (2022): e0265949. 5
Course Project Cont’d
Skin lesion classification
Benign Malignant

Harangi, Balazs, Agnes Baran, and Andras Hajdu. "Assisted deep learning framework for multi-class skin lesion classification
considering a binary classification support." Biomedical Signal Processing and Control 62 (2020): 102041.
10/8/2024 6
Course Project Cont’d
Breast MRI tumor classification
Benign Malignant

10/8/2024 7
UBC Ovarian Cancer subtypE clAssification
and outlier detectioN (UBC-OCEAN)
High-Grade Serous Carcinoma: HGSC
Low-Grade Serous Carcinoma: LGSC
Endometrioid Carcinoma: EC
Clear Cell Carcinoma: CC
Mucinous Carcinoma: MC
Outliers: Other

10/8/2024 *UBC: University of British Columbia 8


UBC Ovarian Cancer Subtype Classification and Outlier Detection (UBC-OCEAN) | Kaggle
Pretrained deep neural networks
• Pretrained networks have been trained on more than a million images and can classify
images into 1000 object categories, such as keyboard, coffee mug, pencil, and many
animals.
• AlexNet and SqueezeNet are successful and effective examples of pre-trained CNN models
used recently in many machine learning applications, such as classification, clustering, and
object recognition.
• Using a pre-trained network is typically much faster and easier than training a network
from scratch.
• Transfer learning is a machine learning method where we reuse a pre-trained model as
the starting point for a new model to accomplish a new task. The knowledge gained from
the first model is thus transferred to the second model which focuses on a new task.
• For instance, using the knowledge gained from cat/dog detection to detect buildings.

10/8/2024 9
Pretrained models and cost of each model

10/8/2024 10
Compare pretrained networks

https://medium.com/analytics-vidhya/cnns-architectures-lenet-alexnet-vgg-googlenet-resnet-
10/8/2024 11
and-more-666091488df5
Requirements
• Form groups of 3 students.

• Adapt two pretrained models such as ResNet18 and ResNet101to classify the image by assigning it to a
specific label.

• Present a comparison between the two pretrained models in terms of accuracy and computation time.

• Python will be used throughout the course

• Use Kaggle or Google Colaboratory to write and run your Python script
(https://colab.research.google.com/)

• Google Colab is a free online cloud-based Jupyter notebook environment that allows us to train our
machine learning and deep learning models on CPUs, GPUs, and TPUs.
10/8/2024 12
Thank you!

10/8/2024 13

You might also like