Review: Deep Learning & CNN in Imaging of Medical Surgeries

The paper discusses the recent trends in 3D deep learning-based imaging used at the time of medical imaging viz. before / after or during the surgeries. Deep learning is a concept of Machine Learning (ML) & Artificial Intelligence (AI) in which a machine gains knowledge in a way similar to the human being, unlike conventional ML algorithms whose nature is linear that is they only learn in a specific way. But in case of the deep learning, there is hierarchical learning, and the complexity increases with each step. A few examples of deep learning are autonomous vehicles, face recognition, virtual assistants used by various companies in order to help customers, etc. On the other hand, imaging is a process of visually representing the features of any object or material's form and the characteristics such as size, shape, color, texture, boundaries etc. Most of the images known to us are usually captured using cameras but the word ‘image' is associated with more than just the cameras. Images can be captured by various other techniques like Ultrasound, X-rays, Magnetic resonance Imaging (MRI), Computed Tomography (CT), used in medical fields Infrared (IR) and Radio Imaging in case of astronomical research, RADAR and LIDAR imaging in case of driverless vehicles and many more applications can be found where images are taken and processed. Deep learning in this field could prove a very effective tool in diagnosing and predicting the causes of the disease in an optimal way with reasonable speeds and most important with greater accuracy once the model is trained properly with a large amount of data. So, this paper is an effort toward summarizing a few important types of research in this field and getting acquainted with these disruptive technologies of the upcoming future.