detecTify-Brain Tumor Segmentation

Source: Deep Learning on Medium

detecTify-Brain Tumor Segmentation

Team Members: Abhimanyu Banerjee, Aniket Chowdhury, Aditya Varshney, Abhinav Robinson, Aaryan Kapur

A brain tumor is defined as abnormal growth of cells within the brain or central spinal canal. Some tumors can be cancerous thus they need to be detected and cured in time. The exact cause of brain tumors is not clear and neither is exact set of symptoms defined, thus, people may be suffering from it without realizing the danger.

The symptom having of brain tumor depends on the location, size and type of the tumor. It occurs when the tumor compressing the surrounding cells and gives out pressure. Besides, it is also occurs when the tumor block the fluid that flows throughout the brain. The common symptoms are having headache, nausea and vomiting, and having problem in balancing and walking. Brain tumor can be detected by the diagnostic imaging modalities such as CT scan and MRI. The reason we have chosen to work with MRI scans is the they are highly adept at capturing images that help doctors determine if there are abnormal tissues within the body. MRIs are more detailed in their images.
We are developing a Java Based Application which would help us detect Brain Tumors based on Brain MRI Images

Why this project ?

We as a team has always believed that technology could take health care far beyond its current stage.Technology could put modern healthcare and diagnosis services into the hands of every man.People today don’t have the resources and time to go to a quality doctor to diagnose themselves of even influenza cold or a disease as serious as tumor . this inspired us to create this project.

Intended Audience and Objective

The objective of this work is to bring some useful information in simpler form in front of the users.We are working on this project, with the intention and ambition of making disease detection and specifically tumor detection easily and readily available.We are intending this project for people who don’t have access to reliable or readily available healthcare regarding tumors and their detection .

Project Snaps

Project Workflow

  1. The User would be required to put feed in an image of the MRI Scan into the Java Application where the image would be sent to the server for processing.
  2. Our Trained model which is contained in the Server will accept this image as an input.These images will be treated as the input data for the model which will clean and augment the MRI Scan.
  3. These Scans will be put through our convolutional neural network (CNN) which will be written using Deep Learning libraries like TensorFlow and Keras.

4. The Results Generated will be Analyzed and sent back to the Application in order to apprise the user about the Generated Medical Report.

5. These Generated results can also be accessed by the User from a Database which is present on the Server.

Project Limitations

As of now we cant promise a complete replacement of the doctor but this project can surely be an aide to the doctor.

No matter how accurate our results maybe, at this point of time a professionally qualified doctor’s advice is a must before drawing any conclusions about the tumor.

Future Scope

We plan to not just limit our application to tumor but extend it to Cover an even larger variety and range of diseases. We aim to make it an Ultimate medical diagnosis app in future

View the code here: