AI-Driven Medical Imaging Platform: Advancements in Image Analysis and Healthcare Diagnosis

Document Type : Original Article

Authors

1 Research engineer, DigiBrain4, USA.

2 CRO, DigiBrain4, USA, visiting professor Shorouk Academy, EG.

Abstract

In the realm of healthcare, the integration of artificial intelligence (AI) has revolutionized medical imaging analysis [1-3]. This research paper delves into the AI-driven aspects of a comprehensive medical imaging platform, focusing on three pivotal phases: classification, object detection, and segmentation.
In the classification phase, we harnessed the potential of AI models, including ResNet50 [4], DenseNet121 [5], and VGG16 [6], to accurately categorize medical images such as CT, MRI, Sonar, and X-ray. For object detection, we employed YOLOv5l [7] to efficiently identify abnormalities in X-ray images and tumors in MRI brain images. In the segmentation phase, we developed specialized models, including U-Net [8], Attention U-Net [9], and Res50-U-Net [10], to precisely delineate tumors from MRI brain images.
what differs our platform from others is the automatic pipeline that progressively process the medical images.
Our results demonstrate the effectiveness of AI in enhancing diagnostic accuracy and streamlining medical image analysis. By focusing solely on the AI components, this research paper sheds light on the transformative impact of AI in healthcare, paving the way for more accurate diagnoses and improved patient care.

Keywords