CLOUD APP FOR PNEUMONIA DETECTION WITH X-RAY IMAGES IN DL

Inspiration

Pneumonia is a respiratory infection caused by bacteria or viruses, it affects many individuals, especially in developing and underdeveloped nations, where high levels of pollution, unhygienic living conditions, and overcrowding are relatively common, together with inadequate medical infrastructure. Pneumonia causes pleural effusion, a condition in which fluids fill the lung, causing respiratory difficulty. Early diagnosis of pneumonia is crucial to ensure curative treatment and increase survival rates. Chest X-ray imaging is the most frequently used method for diagnosing pneumonia. However, the examination of chest X-rays is a challenging task and is prone to subjective variability. In this study, we developed a computer-aided diagnosis system for automatic pneumonia detection using chest X-ray images. We develop a deep learning approach to detect pneumonia disease detection using 3 neural network algorithms CNN with vgg16, CNN with resnet50, Unet. We create a Real Time Application Pneumonia Prediction Web App using Python – Flask Framework, Deployed in Heroku Cloud Application platform.

What it does

Artificial Intelligence has been witnessing a monumental growth in bridging the gap between the capabilities of humans and machines. Researchers and enthusiasts alike, work on numerous aspects of the field to make amazing things happen. One of many such areas is the domain of Computer Vision. The agenda for this field is to enable machines to view the world as humans do, perceive it in a similar manner and even use the knowledge for a multitude of tasks such as Image & Video recognition, Image Analysis & Classification, Media Recreation, Recommendation Systems, Natural Language Processing, etc. The advancements in Computer Vision with Deep Learning have been constructed and perfected with time, primarily over one particular algorithm — a Convolutional Neural Network. A Convolutional Neural Network (ConvNet/CNN) is a Deep Learning algorithm which can take in an input image, assign importance (learnable weights and biases) to various aspects/objects in the image and be able to differentiate one from the other. The pre-processing required in a ConvNet is much lower as compared to other classification algorithms. While in primitive methods filters are hand-engineered, with enough training, ConvNets have the ability to learn these filters/characteristics.

How we built it

The architecture of a ConvNet is analogous to that of the connectivity pattern of Neurons in the Human Brain and was inspired by the organization of the Visual Cortex. Individual neurons respond to stimuli only in a restricted region of the visual field known as the Receptive Field. A collection of such fields overlaps to cover the entire visual area. Machine learning and Deep learning refers same. In machine learning the features form data should be explicitly given by user but in deep learning the algorithm itself took the main features in the data. In other words, the machine learning has lower computational power compared to the deep learning. There are some deep learning algorithms which had already made to solve the complex problem. Some deep learning algorithms are feed-forward neural network, convolutional neural network, recurrent neural network, long short-term memory networks, auto encoders, and etc. In all algorithms, there are three types of layers: input layer, hidden layer, output layer. The major computational part will be done in hidden layer. Before knowing deep learning algorithms, one should know about the perceptron, basic cell or neuron, in machine learning.

Challenges we ran into

Pneumonia remains a significant public health concern globally, contributing to substantial morbidity and mortality, particularly among vulnerable populations. Timely and accurate diagnosis of pneumonia is crucial for effective patient management and improved clinical outcomes. However, existing diagnostic methods, such as manual interpretation of X-ray images by radiologists, are often subjective, time-consuming, and prone to errors. This presents a pressing need for the development of a robust and automated system for pneumonia detection.

   The objective of this project is to develop a cloud-based application powered by deep learning algorithms capable of accurately detecting pneumonia from X-ray images. The application will enable healthcare professionals to upload X-ray images securely to the cloud platform, where they will undergo automated analysis using advanced machine learning techniques. The diagnostic results, including the presence or absence of pneumonia and any associated findings, will be provided to users in a timely manner through an intuitive and user-friendly interface.

Accomplishments that we're proud of

 Design and train a deep learning model using convolutional neural networks (CNNs) or similar architectures to accurately detect pneumonia from X-ray image  Utilize large datasets of labeled X-ray images to train the model, ensuring robust performance across a wide range of patient demographics and imaging conditions. Cloud Infrastructure Setup:  Establish a scalable and reliable cloud infrastructure to host and deploy the deep learning model.  Select a cloud service provider (e.g., AWS, Google Cloud, Microsoft Azure) and configure computing resources, storage solutions, and networking components to support the computational requirements of the application. User Interface Development:  Develop an intuitive web-based user interface that allows healthcare professionals to securely upload X-ray images and receive diagnostic results.  Implement user authentication and access control mechanisms to ensure data privacy and compliance with healthcare regulations.

What we learned

The testing approach document is designed for Information and Technology Services’ upgrades to PeopleSoft. The document contains an overview of the testing activities to be performed when an upgrade or enhancement is made, or a module is added to an existing application. The emphasis is on testing critical business processes, while minimizing the time necessary for testing while also mitigating risks. It’s important to note that reducing the amount of testing done in an upgrade increases the potential for problems after go-live.

What's next for CLOUD APP FOR PNEUMONIA DETECTION WITH X-RAY IMAGES IN DL

In the future, it is hoped that transfer learning models would be trained on this dataset that would outperform these CNN models. It is also expected that neural network models based on GAN, generative adversarial networks, would also be trained and compared with the existing models. From our work, Mask-RCNN showed better results in pneumonia detection with max accuracy. The deep learning algorithms has its benefits when comparative to machine learning algorithms. But the main problem is computational cost and is high. In each and every problem, like detection, prediction, classification, recommendation systems in medical industry adopting deep learning algorithms for better precise decisions. deep learning techniques are using in radiology: radiotherapy. Google had already started its collaboration with one of the universities in England to work in radiotherapy.

Built With

• cnn
• machine-learning
• python
• reset
• unet