Brain Anomaly detection from Radiographic images using deep learning

Project Title

Brain Anomaly detection from Radiographic images using deep learning

Project Area of Specialization Artificial IntelligenceProject Summary

The brain anomalies, are the most common and aggressive disease, leading to a very short life expectancy in their highest grade. Thus, treatment planning is a key stage to improve the quality of life of patients. However, experts to patient ratio is not very feasible in Pakistan. This can be mitigated with the help of Artificial intelligence (AI). 

Generally, various image techniques such as Computed Tomography (CT), Magnetic Resonance Imaging (MRI) etc. are used to evaluate the anomaly such as Brain hemorrhage, Brain tumor, Traumatic brain injury etc. in a brain. We will develop an AI based deep learning system that will take the brain radiographic images and automatically classify it to the respective anomaly of the brain with human level precision at economically acceptable computational complexity.

Project Objectives

1. Development of deep learning algorithms for the analysis of medical imagery
2. Anomaly detection techniques in the context of medical imagery analysis
3. Localization of an abnormality using deep learning algorithms
4. Contribute to the field of medical image analysis

Project Implementation Method

Methodology:

Here is a brief overview of how this research will move forward while achieving the desired results;

1. Data collection

This is a very crucial step as the annotated data is a huge problem. This step will be achieved by collecting data from following different sources.

1. Online labelled imagery
2. Local hospital imagery
3. Unannotated data will be labeled through expert

1. Data preprocessing

1. Data Cleansing
2. Data Normalization

1. Deep Neural Network (DNN) algorithm.
   Different option would be used for this step.

1. Training indigenously developed algorithm
2. Pre-Trained models can be used while benefiting from transfer learning
3. A DNN model can be deployed online that will auto-learn with each new labelled data

1. Classification
2. Validation of Results

Finally, classified results will be validated through an expert.

Tools:

Following are the different tools and deep learning libraries that will be used during this research.

Development:

1. 1. Python
      1. Matplotlib
      2. Pandas
      3. NumPy
      4. Keras
      5. TensorFlow

Data Labeling:

Hardware:

1. GPU/TPU; Self acquired or from Google Colab

Benefits of the Project

Methodology:

Here is a brief overview of how this research will move forward while achieving the desired results;

1. Data collection

This is a very crucial step as the annotated data is a huge problem. This step will be achieved by collecting data from following different sources.

1. Online labelled imagery
2. Local hospital imagery
3. Unannotated data will be labeled through expert

1. Data preprocessing

1. Data Cleansing
2. Data Normalization

1. Deep Neural Network (DNN) algorithm.
   Different option would be used for this step.

1. Training indigenously developed algorithm
2. Pre-Trained models can be used while benefiting from transfer learning
3. A DNN model can be deployed online that will auto-learn with each new labelled data

1. Classification
2. Validation of Results

Finally, classified results will be validated through an expert.

Tools:

Following are the different tools and deep learning libraries that will be used during this research.

Development:

1. Python
   1. TensorFlow
   2. Keras
   3. NumPy
   4. Pandas
   5. Matplotlib

Data Labeling:

Hardware:

1. GPU/TPU; Self acquired or from Google Colab

Technical Details of Final Deliverable

We will develop a specific environment for that which will proceed in the future to perform the task of data cleaning processing and developing a model to get high accuracy. And last but not the least deployment of the model.

Final Deliverable of the Project HW/SW integrated systemCore Industry ITOther Industries Medical Core Technology Artificial Intelligence(AI)Other TechnologiesSustainable Development Goals Good Health and Well-Being for PeopleRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	77897
Google Colab Pro+(for Seven months)	Equipment	7	9877	69139
Hospital Visit ,internet etc	Miscellaneous	1	8758	8758