personal-web-image-assets/machine-learning/brain-tumor/preview
personal-web-image-assets/machine-learning/brain-tumor/img1
personal-web-image-assets/machine-learning/brain-tumor/img2
personal-web-image-assets/machine-learning/brain-tumor/img3
personal-web-image-assets/machine-learning/brain-tumor/img4
personal-web-image-assets/machine-learning/brain-tumor/img5
personal-web-image-assets/machine-learning/brain-tumor/img6
Introduction
This is a human brain tumor detector app that can identify 3 types of tumor. Given a medical MRI image, app will analyse the image and return an analysed result. The result indicate whether there is a tumor detected or not with probability.
The purpose of the app is to take advantage of machine learning to assist medical professional or patient in brain tumor detection

Motivation & Purpose
This tumor detector app was builded during study of computer vision in deep learning.
It usually take a period of time to waiting to see the result of MRI of brain from doctor. This waiting could take days to a week.
The tumor detector app can shorten the waiting and be used as personal assistance.

More
The tumor detector app include a deep learning model that was trained on a set of MRI human brain images. And the app can detect 3 type of brain cancers.
In addition it also report the probability of have any of these cancer in percentage.
The set of MRI images was used to train the deep learning model are in different angles of human head. Therefore new MRI image can be in any angles as below.
    🔴 top
    🔴 left
    🔴 right
    🔴 front
    🔴 back
The result tell you whether there is a tumor detected or not, what type of tumor and the probability.
The model was trained with dataset from Kaggle.
Frameworks were used to build this app:
    🧰 Tensorflow : Deep learning framework for the model
    🧰 Streamlit : Frontend app

Note
The detecotr is only for assistance and nothing is 100% certain. Don't use this detector as a personal doctor. Seeking for professional doctor is recommend.

Website