Emerging technologies and open data: introduction to data science applied to image analysis

We present a new report in the series 'Emerging Technologies and Open Data', by Alejandro Alija. The aim of these reports is to help the reader understand how various technologies work, what is the role of open data in them and what impact they will have on our society. This series includes monographs on data analysis techniques such as natural language analysis and predictive analytics.  This new volume of the series analyzes the key aspects of data analysis applied to images and, through this exercise, Artificial Intelligence applied to the identification and classification of diseases by means of medical radio imaging, delves into the more practical side of the monograph.

Image analysis adopts different names and ways of referring to it. Some of the most common are visual analytics, computer vision or image processing. The importance of this type of analysis is of great relevance nowadays, since many of the most modern algorithmic techniques of artificial intelligence have been designed specifically for this purpose. Some of its applications can be seen in our daily lives, such as the identification of license plates to access a parking lot or the digitization of scanned text to be manipulated.

The report introduces the fundamental concepts that allow us to understand how image analysis works, detailing the main application cases in various sectors. After a brief introduction by the author, which will serve as a basis for contextualizing the subject matter, the full report is presented, following the traditional structure of the series:

• Awareness. The Awareness section explains the key concepts of image analysis techniques. Through this section, readers can find answers to questions such as: how are images manipulated as data, how are images classified, and discover some of the most prominent applications in image analysis.
• Inspire. The Inspire section takes a detailed look at some of the main use cases in sectors as diverse as agriculture, industry and real estate. It also includes examples of applications in the field of medicine, where the author shows some particularly important challenges in this area.
• Action: In this case, the Action section has been published in notebook format, separately from the theoretical report. It shows a practical example of Artificial Intelligence applied to the identification and classification of diseases using medical radio imaging. This post includes a step-by-step explanation of the exercise. The source code is available so that readers can learn and experiment by themselves the intelligent analysis of images.

Below, you can download the report - Awareness and Inspire sections - in pdf and word (reusable version).