May 5, 2022
An examination of emerging technologies would be helpful to determine if any might be leveraged across one or more of the imaging modalities in the organization. Examining existing market technologies involves understanding the various technologies’ abilities and current use in imaging areas.
If a technology is not currently being used in an R&D setting, it will be essential to assess the applicability for the imaging modalities of the biopharmaceutical R&D groups. The following are some of the categories of emerging imaging technologies that could be considered. All of these technologies may have applicability in R&D imaging in the near future.
Integrity, Authenticity & Confidentiality
In the life sciences industry it is imperative to ensure the security and confidentiality of any image and the metadata surrounding that image.
- Medical image watermarking – Watermark graphic or text indicates ownership or copyright of an image. This technology makes it difficult for someone to use the image without permission or claim ownership of the original.
- Blockchain – For secure sharing of medical images blockchain allows users (i.e., patients, physicians, radiologists and scientists) to control how and by whom healthcare data are used.
The main goal of image analysis is to help address and solve scientific questions by leveraging state of the art image analysis technology to find ways to harvest information from those images.
- Reproducible image analysis – Is a set of resources that allow any person to replicate the process of image handling and analysis. The aim is to derive the same results as the authors presented.
- Container – based image analysis – containers support the ability to deploy analysis code along with versioned run time elements across different infrastructures regardless of the host environment. Resources, libraries, and tools are shared in real time and only for the analysis needed . This provides for reproducible and repeatable image analyses. Technology providers include Docker, Singularity, Amazon Machine Images, Amazon Fargate
- AI/ML Deep Learning – AI has been deployed significantly in image analysis especially in cellular image analysis and high content analysis. There are solutions that utilize machine learning classification algorithms to evaluate multiple measures simultaneously to automatically identify classes within the data and reveal deeper information than can be discovered with manual classification in more biophysical approaches. As of today, AI uses have been limited to improving peak detection in NMR or mass spec or diagnosing potential sample related problems. Other applications are focused more on the use of AI for the curation of large structural data sets for further analysis.
Augmented Reality / Virtual Reality
Augmented reality augments your surroundings by adding digital elements to a live view, often by using the camera on a smartphone while virtual reality (VR) is a completely immersive experience that replaces a real-life environment with a simulated one.
Given the 3D structure of our cognitive and visual processes, an intuitive 3D visualization and computational tool is more than a step forward – it’s the inevitable progression of research ranging from 3D genetic analysis to 3D data visualization. When we reduce a 3D object to a 2D medium, our cognitive system has to work extra hard to comprehend the information. We can undertake scientific research more efficiently by using spatial computing to view massive data sets and analyze test findings of 3D objects.
Semantic Image Annotation (SIA) & Ontologies
The process of assigning a class or description to an unknown image is known as image annotation. The goal of automatic picture annotation, in particular, is to provide coherent visual descriptions that are as excellent as those written by humans. This will not only allow for a more rapid and thorough knowledge of the contents of image collections, but it may also be used to improve the performance of image retrievals by content. SIA is a framework for automatically annotating photos with ontologies.
Image Management & Viewing
Intended to provide access to images for viewing in various ways.
- Zero-footprint image viewing – A zero footprint viewer requires no client-side installation or download, allowing users to read documents and images in their native web browser while taking advantage of the browser’s full capabilities, built-in plugins, and interactions with the device.
- Content-based image retrieval – this is the application of computer vision techniques to the image retrieval problem. It solves the issue of searching for digital images in vast databases.
- Multi-modality enterprise Vendor Neutral Archive(VNA) – healthcare organizations are moving to VNA to manage images regardless of source or format. VNAs are intended to manage images in their native format and can be implemented alongside an existing PACS. Supports DICOM and non-DICOM images, but extent of native image format support can vary.
An evaluation of upcoming technologies would be beneficial in determining whether any could be used across one or more of the organization’s imaging modalities. Understanding the various technologies’ capabilities and present application in imaging is necessary for determining their applicability in biopharmaceutical R&D. If a technology isn’t presently in use in R&D, it’ll be necessary to evaluate its suitability for any of the imaging modalities.
Why It Matters to You
Emerging Technologies could play an important role in a Digital Transformation of the imaging modalities of biopharmaceuticals R&D groups. Performing an assessment of the current market would be a prudent step in the process of a transformation.
In this blog we discussed:
- Key areas of emerging technologies that could be considered,
- What these technologies do.
- How they impact the imaging management.