The project was a collaborative effort within the Omdena initiative, uniting a diverse team of AI enthusiasts. Together, we took on the task of utilizing deep learning methods to detect various chest diseases, with a primary focus on pneumonia. The objective was to create a precise and easily accessible diagnostic tool, especially crucial in areas with constrained healthcare resources and a scarcity of specialized medical professionals.
In essence, the project's core aim was to leverage deep learning techniques for the accurate identification of chest diseases, particularly pneumonia, in regions where access to expert healthcare practitioners, particularly radiologists, is limited, such as in Myanmar.
Click on the certificate above to view the full-size certificate.
- Implemented and fine-tuned state-of-the-art models like MobileNetV2 and ResNet9 for optimal performance.
- Trained a deep learning model to detect chest diseases, achieving a validation accuracy of up to 89.36%.