This repository contains a computer vision case study focused on the classification of chest CT scans for cancer identification. The objective of this project is to develop and compare the performance of three different models: EfficientNetV3, ResNet50, and VGG19. The results of the study indicate that the EfficientNetV3 model achieved the highest accuracy and overall performance.
The dataset used in this case study consists of a collection of chest CT scans labeled with cancer identification labels such as Adenocarcinoma, Large cell carcinoma, Squamous cell carcinoma and normal. The dataset is divided into training, validation, and test sets.
The link to the dataset: https://www.kaggle.com/datasets/mohamedhanyyy/chest-ctscan-images
Three different models were implemented and evaluated in this case study:
-
EfficientNetV3: EfficientNet is a family of convolutional neural network architectures that have achieved state-of-the-art performance on various computer vision tasks. The EfficientNetV3 model was chosen for its impressive results on similar image classification problems.
-
ResNet50: ResNet is a widely adopted deep learning architecture that introduced residual connections to alleviate the vanishing gradient problem. ResNet50, with 50 layers, was selected as a representative model from the ResNet family.
-
VGG19: VGG (Visual Geometry Group) is a classic convolutional neural network architecture known for its simplicity and effectiveness. VGG19, with 19 layers, was chosen for its strong performance on image classification tasks.
This project is licensed under the Apache License 2.0.