Akshat111111 / Hedging-of-Financial-Derivatives

This strategy works for every market condition irrespective of the movement
BSD 3-Clause "New" or "Revised" License
74 stars 100 forks source link

💡[FEATURE]: Predicting COVID-19 from Chest X-Ray Images #406

Closed ashis2004 closed 3 months ago

ashis2004 commented 3 months ago

Is your feature request related to a problem? Please describe. The prediction of COVID-19 from chest X-ray images using Convolutional Neural Networks (CNNs) has several critical use cases in the healthcare sector. Firstly, it enhances diagnostic accuracy and speed in clinical settings, especially in regions with limited access to PCR testing facilities. By rapidly analyzing X-ray images, CNNs can assist radiologists in identifying COVID-19-related abnormalities, enabling timely isolation and treatment of infected individuals. Secondly, CNN-based models can serve as valuable tools in emergency departments, where quick decision-making is essential. They help in triaging patients, determining the severity of lung involvement, and prioritizing those needing immediate medical attention.

Describe the solution you'd like Built and trained a convolutional neural network in Keras with TensorFlow as backend from scratch to predict patients if they were infected with COVID-19 or not using their chest X-ray images. Matplotlib was used for data visualization. Data preprocessing and data augmentation was carried out using tensorflow 2.0 The model used was sequential with a combination of convolutional layers, pooling layers, dropout layers, dense layers with relu activation and output layer with sigmoid activation. The dataset contained the lungs X-ray images of both groups i.e non-covid and covid infected patients. The dataset was obtained from kaggle. Metrics chosen for model evaluation were Training set, test set and validation set accuracy. Adam optimizer with learning rate of 0.001 was choosed for gradient descent The entire project was carried out on the Google Colab environment Results of the model were following:

Training Set Accuracy : 98.41 %

Training Set Loss : 0.0490

Validation Set Accuracy : 97.51 %

Validation Set Loss: 0.0759

Test Set Accuracy : 94.83 %

Test Set Loss : 0.1141

Describe alternatives you've considered A clear and concise description of any alternative solutions or features you've considered.

Additional context

image image

github-actions[bot] commented 3 months ago

Hi there! Thanks for opening this issue. We appreciate your contribution to this open-source project. We aim to respond or assign your issue as soon as possible.

ashis2004 commented 3 months ago

@Akshat111111 please assign it to me under Gssoc

Akshat111111 commented 3 months ago

this is a financial repo