What is Digital Pathology?
Digital Pathology:
Digital Pathology consists of creating, observing, sharing, analyzing and interpreting specific content in digital images of whole-slide histopathology (WSI). This approach has been increasingly implemented in clinical trials, educational settings and research. For example, in the United States, the Food and Drug Administration - FDA, since 2017 has approved the use of commercial platforms for digital histopathology whole-slide imaging, which has provided tools for primary diagnostics [1].
In this field of research, we explore how to combine experimental design, statistical pattern recognition and prognostic analysis within a unified framework that allows answering not only questions from the clinical setting, but also from a scientific approach [x], where subsequently Artificial Intelligence - AI models can be implemented that allow performing different tasks on the image, for example, segmentation or tissue classification [2].
Whole Slide Images:
These are large-scale, high-resolution digitized images (e.g., 80,000 x 80,000 pixels), rendered at various magnifications that can occupy storage capacities around Gigabytes using a common pyramid format (e.g., JPEG2000).
Compared to their radiology counterparts, the average size of a standard pathology WSI file can range from a few hundred megabytes to several gigabytes. WSIs are formatted as multi-resolution pyramids composed of hundreds of thousands of individual images optimized for fast real-time rendering and display. This encodes a multilayer pyramid model in which the specific FOV is held constant, and subsequent images are derived from adjacent image data. Several vendors use proprietary image file extensions, although there are ongoing interoperability initiatives. Various compression formats are used to generate smaller file sizes. Image files can use various compression schemes, such as TIFF, which can be lossless or lossy compression, or JPEG, which uses a lossy compression format. Digital Imaging and Communications in Medicine (DICOM) has published supplements 145 and 122 to discuss a universal WSI interchange standard, and is an ongoing effort. Specific WSI viewers are needed to view proprietary file formats, but progress has been made in cross-displaying different WSI file formats in single viewers [1].
Retrieved from: [1]
Telepathology:
The term Telepathology (TP), first used by Weinstein in 1986, is defined as “using telecommunications technology to make remote diagnoses on a computer screen rather than directly through the lens of a microscope”. Almost any type of specimen viewed with light microscopy can also be evaluated using TP. TP can also refer to sending nonimaging file types to make diagnoses, but for the purposes of this review, we refer to TP in the context of digital pathology to support remote intraoperative consultation (IOC) [1].
Telepathology is the diagnosis of surgical pathology cases at a distance using real-time video imaging or store-and-forward digitized images. The American Telemedicine Association clinical guidelines for telepathology define telepathology as: “A form of communication between medical professionals that includes the transmission of pathology images and associated clinical information for the purpose of various clinical applications including, but not limited to, primary diagnoses, rapid cytology interpretation, intraoperative and second opinion consultations, ancillary study review, archiving, and quality activities.”[2].
Computational pathology:
Computational Pathology investigates a complete probabilistic treatment of scientific and clinical workflows in general pathology, i.e. it combines experimental design, statistical pattern recognition and survival analysis within a unified framework to answer scientific and clinical questions in pathology [1].
We define computational pathology as an approach to diagnosis that incorporates multiple sources of raw data (eg, clinical electronic medical records; laboratory data, including “-omics”; and imaging); extracts biologically and clinically relevant information from those data; uses mathematical models at the levels of molecules, individuals, and populations to generate diagnostic inferences and predictions; and presents that clinically actionable knowledge to customers through dynamic and integrated reports and interfaces, enabling physicians, patients, laboratory personnel, and other health care system stakeholders to make the best possible medical decisions [2].
What is Industry 4.0?
Fourth Industrial Revolution or Industry 4.0:
It is a significant transformation of the entire industrial production by merging digital and internet technologies to conventional industry. Opinions are divided on the use of terms of revolution or evolution. In Europe, the concept was launched and it is supported by Germany government programs and leading companies like Siemens or Bosch. In America, the approach is often called "Smart Manufacturing" in China discusses the "Made in China 2025" and Japan "Innovation 25".
All aim the development of an industry to launch products faster to increase flexibility and increase resource efficiency through digitization. In intelligent factories created by Industry 4.0, modularly structured, cyber-physical systems monitor physical processes, create a virtual copy of the physical world, and make decentralized decisions. They communicate using the Internet of Things, cooperating in real time with each other and with human resources.The information storage and processing takes place using Cloud computing [1].
Retrieved from: [1]
Big Data:
The term Big Data refers to large growing data sets that include heterogeneous formats: structured, unstructured and semi-structured data. Big Data has a complex nature that requires powerful technologies and advanced algorithms. So the traditional static Business Intelligence tools can no longer be efficient in the case of Big Data applications.
Most data scientists and experts define Big Data by the following three main characteristics (called the 3Vs) (Furht and Villanustre, 2016):
Volume: Large volumes of digital data are generated continuously from millions of devices and applications (ICTs, smartphones, products’ codes, social networks, sensors, logs, etc.).
Velocity: Data is generated in a fast way and should be processed rapidly to extract useful information and relevant insights. For instance, Wallmart (an international discount retail chain) generates more than 2.5 PB of data every hour from its customers' transactions. YouTube is another good example that illustrates the fast speed of Big Data.
Variety: Big Data is generated from various sources and in multiple formats (e.g., videos, documents, comments, logs). Large data sets consist of structured and unstructured data, public or private, local or distant, shared or confidential, complete or incomplete, etc.
Emani et al. (2015) and Gandomi and Haider (2015) indicate that more Vs and other characteristics have been added by some actors to better define Big Data: Vision (a purpose), Verification (processed data conforme to some specifications), Validation (the purpose is fulfilled), Value (pertinent information can be extracted for many sectors), Complexity (it is difficult to organize and analyze Big data because of evolving data relationships) and Immutability (collected and stored Big data can be permanent if well managed) [1].
Retrieved from: [1]
Data Analytics:
Data analytics is defined as the application of computer systems to the analysis of large data sets for the support of decisions. Data analytics is a very interdisciplinary field that has adopted aspects from many other scientific disciplines such as statistics, machine learning, pattern recognition, system theory, operations research, or artificial intelligence.
Typical data analysis projects can be divided into several phases. Data are assessed and selected, cleaned and filtered, visualized and analyzed, and the analysis results are finally interpreted and evaluated. The knowledge discovery in databases (KDD) process comprises the six phases selection, preprocessing, transformation, data mining, interpretation, and evaluation [1].
Retrieved from: [1]
Computer Vision:
Computer vision is a field of artificial intelligence (AI) that enables computers and systems to derive meaningful information from digital images, videos and other visual inputs and take actions or make recommendations based on that information. If AI enables computers to think, computer vision enables them to see, observe and understand.
Computer vision works much the same as human vision, except humans have a head start. Human sight has the advantage of lifetimes of context to train how to tell objects apart, how far away they are, whether they are moving and whether there is something wrong in an image.
Computer vision trains machines to perform these functions, but it has to do it in much less time with cameras, data and algorithms rather than retinas, optic nerves and a visual cortex. Because a system trained to inspect products or watch a production asset can analyze thousands of products or processes a minute, noticing imperceptible defects or issues, it can quickly surpass human capabilities.
Computer vision is used in industries ranging from energy and utilities to manufacturing and automotive and the market is continuing to grow. It is expected to reach USD 48.6 billion by 2022 [1].
Retrieved from: [1]
Artificial Intelligence:
The concept of using computers to simulate intelligent behavior and critical thinking was first described by Alan Turing in 1950.1 In the book Computers and Intelligence, Turing described a simple test, which later became known as the “Turing test,” to determine whether computers were capable of human intelligence.2 Six years later, John McCarthy described the term artificial intelligence (AI) as “the science and engineering of making intelligent machines” [1].
Artificial intelligence (AI) is perhaps the oldest field of computer science and very broad, dealing with all aspects of mimicking cognitive functions for real-world problem solving and building systems that learn and think like people. Therefore, it is often called machine intelligence (Poole, Mackworth, & Goebel, 1998) to contrast it to human intelligence (Russell & Norvig, 2010). The field revolved around the intersection of cognitive science and computer science (Tenenbaum, Kemp, Griffiths, & Goodman, 2011). AI now raises enormous interest due to the practical successes in machine learning (ML). In AI there was always a strong linkage to explainability, and an early example is the Advice Taker proposed by McCarthy in 1958 as a “program with common sense” (McCarthy, 1960). It was probably the first time proposing common sense reasoning abilities as the key to AI. Recent research emphasizes more and more that AI systems should be able to build causal models of the world that support explanation and understanding, rather than merely solving pattern recognition problems (Lake, Ullman, Tenenbaum, & Gershman, 2017) [2].
