Brain Abnormal cells detection using Segmentation
When compared to tumors in other parts of the body, the brain tumor has become one of the most prevalent diseases in the world. It is defined as the unrestricted proliferation of abnormal cells in the brain, and it poses a diagnostic problem. In spite of the tumor, there are also some damaged cells
2025-06-28 16:25:43 - Adil Khan
Brain Abnormal cells detection using Segmentation
Project Area of Specialization Artificial IntelligenceProject SummaryWhen compared to tumors in other parts of the body, the brain tumor has become one of the most prevalent diseases in the world. It is defined as the unrestricted proliferation of abnormal cells in the brain, and it poses a diagnostic problem. In spite of the tumor, there are also some damaged cells in brain cells that cause different diseases, more specifically Dementia. Tumor and Dementia both are becoming common and also are dangerous. Tumors can directly affect the human nervous system. Usually, brain abnormalities are traditionally detected using MRI imaging by medical experts. But we proposed to make a system that will take MRI images of the brain and tell whether a patient has a tumor or some brain cell damage or not. This system will use segmentation as it gives more accuracy about the exact size, position, and texture of abnormal brain cells. The proposed system will be a fully automatic model-based system, that will use a Convolutional neural network, with U-net architecture as it provides more precision for medical images. And we will do things under proper medical experts.
Project ObjectivesThe main objective of the proposed system is to help doctors and patients. Early identification of tumors and Dementia is very important so that further treatment should be carried out. Traditionally, the tumor is detected normally at Stage three or four. And are not easy to cure it. Some traditional methods are also used to detect Dementia, so we want to replace these traditional ways with a system that is more efficient and accurate.
Project Implementation Method- The system will first take input in the form of MRI images.
- Preprocessing raw images will be the next step. Preprocessing removes noise by ensuring image parity, which improves the effectiveness of the subsequent segmentation and feature extraction steps.
- Then we will use different deep learning models to train our model.
- After proper model training, the model will recognize whether a particular MRI image is normal or abnormal.
- After detecting abnormality, the segmentation method will be used to find the exact position and size of the tumor and damaged cells.
- As a result, will show an MRI image with highlighted abnormal cells and damaged cells.
This project will have many benefits. Some of them are given below.
- The system will detect tumors or damaged brain cells at the very initial stage.
- The system will provide great precision, sensitivity, and specificity.
- The system will tell the doctor about the exact location, size, and texture of tumors or damaged cells.
- The main UI/UX design of the software will be so easy to understand that a doctor can easily understand it and be able to use it.
- System for detecting abnormal brain cells:
The proposed system will detect abnormal brain cells (tumors and damaged cells) with great accuracy and specificity. The system will be developed on a Convolutional neural network using deep learning algorithms. And further classification of tumors or abnormal cells will be carried out by segmentation.
- Complete system design will be delivered by the end of the project. Which will describe how the following system work and how things are carried out.
- Proper documentation of the project will be delivered.
| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 45000 | |||
| GPU | Equipment | 1 | 35000 | 35000 |
| Doctor + Resources | Miscellaneous | 1 | 10000 | 10000 |