Medical Image Analysis or Medical Imaging is technique and process used to capture images of various parts of human organs for diagnostic and treatment purpose in digital healthcare system. The term medical images include various imaging techniques such as computed Tomography (CT) scan (by series of x-ray images), Magnetic Resonance Imaging (MRI) (by radio waves), PET (position emission tomography) scan, x-ray, ultrasound etc. These medical images are analysed in order to diagnose the diseases of patients. Medical image analysis play a vital role in healthcare sector by assisting the expert doctor in diagnosing various diseases like Diagnosis of Hypertension, Glaucoma, brain tumour, Lung cancer, Breast Cancer, corneal diseases, Jaundice and so on. The image processing techniques especially segmentation is a crucial technique in analysis. In order to save the precious human lives, it is necessary to diagnose diseases of patients more accurately and more efficiently. Diagnosis of diseases by using earlier radiological screening techniques are very time consuming, the results of patient may vary between laboratories and sometimes the patient may have to wait hours, days or even weeks. So, to reduce the time consuming screening time and to improve the efficiency of diagnosis Deep Learning(DL) is used in medical field because DL algorithms are very much accurate and automatically learn features from the medical images and performs various tasks like classification of medical images, object detection, Pattern recognition, and lots of other tasks in computer vision areas. Deep learning techniques have proved to be most effective tool for the early prediction of diseases.

Deep Learning is a subfield of machine learning which intern is a subfield of Artificial Intelligence (AI). In deep learning Convolutional neural network (CNN) is a popular algorithm and it has achieved promising performance on lots of computer vision and pattern recognition tasks due to its strong ability of self-learning and dealing with large scale data. Recent investigations have validated that deep learning strategies are effectively connected to medical domain as deep learning is totally founded on artificial neural networks (ANN), ANN is a technology which tries to emulate the pattern/way in which human brain works. One type of neural network is feed forward neural network that consists of various number of hidden layers and is exactly a case of model with deep architecture. The concept of standard back propagation algorithm that became familiar in 1980 is yet a powerful method of training the neural systems. The techniques in deep learning appear to be more powerful for classifying the biomedical images into different categories and accordingly provide higher image analysis. Among other techniques, Convolutional neural network (CNN) is considered to be more suitable so far as medical image analysis is considered which avoid handcrafted features and meet requirements of biomedical image classification with the improved accuracy rate and reliable to meet requirements for classification of biomedical images. However in view of the immense use of data particularly images generated by multiple sources in healthcare industries the necessity for design ,development and implementation of more efficient deep learning techniques has further increased. But deep learning requires large amount of data in order to improve the performance so one of the challenge in medical image analysis is lack of large training datasets.

Advantages of Medical Imaging

It helps in evaluating the impact of drugs in the treatment procedure of patients.
Medical imaging is a non-invasive technique that captures the images of internal organs and helps in regular analysis and follows up of patients in diagnosing different kinds of diseases.
Medical imaging helps in understanding and characterizing the different levels of human body.
Current Artificial Intelligence (AI) trends in Medical Field

Patient centric Approach

AI assisted high quality care to patients
Electronic healthcare reports and Smartphone apps offer at home health solutions.

Data Driven Healthcare

AI allows data compilation and for clinical purpose- medical research and review
Detecting effective treatments based on patient's relevant history data.

Medical Diagnostic Imaging

AI assisted imaging useful for detecting and screening vision threatening eye diseases, serious heart and brain strokes, lesions in lungs and liver.

Enhanced Healthcare Communication

AI assisted apps aiding patients to chat with doctors, appointment booking Decreased medical errors.

Writer is Ph.D. Scholar, IUST
Email:-------- novsheenarasool624@gmail.com