Project Outline for Intracranial Hemorrhage Detection
Patient mortality in cerebral hemorrhage treatment is dependent on rapid diagnosis based on a radiologist’s evaluation of CT scans in traditional method.
Intracranial hemorrhage (ICH) is a type of stroke that occurs when a blood vessel breaks open and bleeds into the space around the brain. As ICH is a life-threatening medical emergency, it is important to detect it as soon as possible.
Neural network approach to find and classify the condition based upon the CT scan. The model architecture implements a time distributed convolutional network. It was observed accuracy above 91% from such an architecture, provided enough dataset is taken for training.
Further extensions to our approach involving the deployment of federated learning would be helpful in pooling learned parameters without violating the inherent privacy of the data involved.
The use of deep learning algorithms can provide a powerful tool for the early detection of ICH in CT scans.
Methodology
Dataset:-
Collection of Intracranial Hemorrhage CT Scans images.
Data Pre-processing:-
Data preprocessing is a process of preparing the raw data and making it suitable for a machine learning model. It is the first and crucial step while creating a machine learning model. When creating a machine learning project, it is not always a case that we come across to clean and format data. And while doing any operation with data, it is mandatory to clean it and put in a formatted way. So for this, we use data preprocessing task.
Implementations
CNN:-
Convolutional Neural Networks are a complex neural network chain which work to get the features of an image from a dataset which is trained and classify them to get the required output. It trains the neural networks by using the dataset images and changing them to numerical values.
The main advantage of CNN compared to its predecessors is that it automatically detects the important features without any human supervision.
ConvNets are more powerful than machine learning algorithms and are also computationally efficient. These numerical values are then put into numerical arrays based on their categorized characteristics.
These arrays are then put into different nodes in the network and passed through multiple iterations based on the input given. The CNN models are used for geographical classification in multiple companies which require data to be classified in a quick and secure way it almost acts like a filter removing dust and separates the features .
Applications:
Develop automated systems for the detection of intracranial hemorrhage in CT scans
Used in clinical settings to quickly and accurately diagnose ICH in patients.
Develop decision support systems that can provide physicians with additional information about the patient’s condition and the best course of treatment.
Hardware Requirement
- Graphics Processing Unit (GPU)
- A minimum of 8GB RAM
- Hard disk with at least 200GB of storage
Software Requirement
- Python programming language
- Keras deep learning framework with tensorflow as backend
- OpenCV computer vision library
What You’ll Learn after doing this project
- Gain a better understanding of how these algorithms work and how to apply them to a real-world problem
- Get experience in building and training neural networks, as well as working with medical imaging data.
- Creation of a graph showing model accuracy for both training and testing
- Plot confusion matrix for different classes
- Shows the performance metrics Specificity, Sensitivity, Accuracy, respectively.
- Create a web-based or Windows application an user interface that provide visualization tools to allow users to view and interpret the CT scans.
