Combining Digital Pathology with Artificial Intelligence

The buzz phrase “digital due diligence” has seen a rise within the world of venture capital of late. From an investor’s perspective, the world of business has changed so drastically through technology such that any business failing to adapt quickly enough may be doomed in the future – even if everything looks good in the present. No matter how great the finances look, if a business fails to invest in technology then they will soon find it impossible to compete with competitors that are moving forward quickly.

The same applies to an extent in healthcare. Every part of the landscape is being transformed by digital technology. Innovations are being delivered by private companies, governments are actively supporting uptake, and digital platforms are being accessed at every opportunity. In all these areas, there are similarities with obstacles and opportunities. Efficiency gains, cost-saving and quicker turn-around times are all advantages of digital innovation, while the initial start-up costs and the difficulty of integrating new systems with old remain obstacles.

Digital pathology is not without its challenges, but in large, these are short-term and practical. It is a question of being able to afford the technology and manage the integration with legacy workflow processes. It appears that the future is very digital and the further this goes, the wider the possibilities become.

Image Quality

2D digital imaging was the first step into digital pathology. In most major territories, these systems were recently approved. They involve viewing slides in a traditional two-dimensional view.

The idea with 3D digital pathology is to add new layers by stacking slides on top of one another to produce images with much more depth and detail. The ability to section tissue mass at a micron level, stain it and then examine it under a digital microscope is promised with new solutions. The entire tissue mass is being examined, meaning that huge amounts of data can be stored.

Much more important data and information is in the hands of pathology teams. This will lead to improved decision-making, analysis, and even improved artificial intelligence and machine learnings, all of which will help to process all this data.

Artificial Intelligence

Within the UK’s NHS, a report into the use of life science technology states that pathology is ripe for innovation as data creates opportunities to change the way care is managed. The report argues that tools like artificial intelligence (AI) will be ‘increasingly important’1.

“Systematic digitization of pathology images could be readily established providing substantial efficiencies in the pathology service within the NHS,” states the report, “allowing the system to become increasingly virtual and reducing the need for every hospital to have the full on-site set of pathologists.”

Paving the way for AI algorithms, digitization and data are crucial to providing insights not currently available with conventional approaches.

While in its development stage, the technology is becoming increasingly viable within healthcare systems with the growing amount of data. AI used to be a great futuristic idea with potential to unlock hidden information, but the capacity was not there. A shortage of data has been the big stumbling block, but this is changing.

Society has become incredibly adept at producing data. The past few years have provided the vast majority of data ever produced and studies suggest that the amount of captured data could increase ten-fold in the next ten years2. Digital technology is making vast amounts of data viewable and shareable in the realm of microscopy.

Clinicians at the University of Pittsburgh Medical Center are collecting a library of digital images that could prove an increasingly useful resource in the creation of AI. In the development of effective algorithms able to support pathologists in their diagnoses, these images will be vital.

Already being pioneered is an algorithm that can identify potential problems within images. This can be used either before screening or post-screening. With the former, pathologists will have access to insights that can guide them in their observations. With the latter, pathologists will be able to either confirm or correct their initial assessment.

As it starts to demonstrate its effectiveness, momentum is gathering behind the technology. At a recent conference, speakers pointed to several studies, all of which showed that AI could exceed the performance of experts.

Researchers at the Massachusetts Institute of Technology (MIT) are using a machine learning system to detect whether breast lesions identified in biopsies are potentially cancerous. The study reported that the machine correctly identified 97% as malignant. It is believed that such levels of accuracy could lead to a reduction of over 30% in unnecessary surgeries3.

Evolutions for the Future

While there have been several projects demonstrating the potential of AI in digital microscopy, this technology is still developing and unproven. Many studies have been conducted in ideal lab conditions to achieve realistic metrics that could be repeated in a daily clinical setting.

To deliver effective and accurate insights, AI will need robust and reliable algorithms. The University of Toronto has recently raised US$1.5million to develop such improved algorithms. Cells that may be cancerous and those that are growing normally could be distinguished between these algorithms. Building upon the effectiveness of readily available devices, they can be enhanced analytics tools.

Further developments in advanced concepts such as AI are being facilitated by innovation, which comes back to the concept of digital due diligence. Businesses risk becoming uncompetitive if they do not consider making their move, even if an executive doesn’t see the need for digital technology.

Regardless of all the impediments to digital adoption, this is where pathology is going. Having the best technology available, be it in a hospital or academic institution, not only impacts the quality of treatment but also its attractiveness to potential students and employees. Places at the cutting-edge of development, delivering the very best care possible, will attract people. For the best and brightest talent institutes that have progressed further with digital microscopy present a more attractive destination.

This has enormous potential to address some of the key challenges impacting pathology departments. Studies have already shown that diagnosis can be improved and the number of costly operations reduced. Furthermore, the limitations in pathologist numbers can only be overcome with automating processes and more mobile images.

References

  1. Life Sciences Industrial Strategy  https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/650447/LifeSciencesIndustrialStrategy_acc2.pdf
  2. What Will We do When the World’s Data Hits 163 Zettabytes? https://www.forbes.com/sites/andrewcave/2017/04/13/what-will-we-do-when-the-worlds-data-hits-163-zettabytes-in-2025/#2b442da8349a
  3. Artificial Intelligence Helping to Detect Breast Cancer https://healthcare-in-europe.com/en/news/artificial-intelligence-helping-to-detect-breast-cancer.html

About Grundium

The Grundium Ocus does away with physical slide transfers between the clinic and the lab and enables live telepathology consultations between surgeons and pathologists.

The Grundium Ocus is a monumental leap in digital pathology. It is a precision tool, and is small and affordable enough to be on every medical professional’s desk. It is truly portable and can be brought anywhere. Wireless connectivity means telepathology is now possible practically anywhere on the planet.

Key Benefits

The Grundium Ocus is a game changing enabler of personal digital microscopy. It brings the global pool of pathology experts to you and it makes looking at sample slides convenient even off-lab.

The Grundium Ocus is your personal microscope scanner to make daily work faster and easier.


Sponsored Content Policy: News-Medical.net publishes articles and related content that may be derived from sources where we have existing commercial relationships, provided such content adds value to the core editorial ethos of News-Medical.Net which is to educate and inform site visitors interested in medical research, science, medical devices and treatments.

Last updated: Oct 18, 2019 at 8:32 AM

Citations

Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Grundium. (2019, October 18). Combining Digital Pathology with Artificial Intelligence. News-Medical. Retrieved on December 15, 2019 from https://www.news-medical.net/whitepaper/20191010/Combining-Digital-Pathology-with-Artificial-Intelligence.aspx.

  • MLA

    Grundium. "Combining Digital Pathology with Artificial Intelligence". News-Medical. 15 December 2019. <https://www.news-medical.net/whitepaper/20191010/Combining-Digital-Pathology-with-Artificial-Intelligence.aspx>.

  • Chicago

    Grundium. "Combining Digital Pathology with Artificial Intelligence". News-Medical. https://www.news-medical.net/whitepaper/20191010/Combining-Digital-Pathology-with-Artificial-Intelligence.aspx. (accessed December 15, 2019).

  • Harvard

    Grundium. 2019. Combining Digital Pathology with Artificial Intelligence. News-Medical, viewed 15 December 2019, https://www.news-medical.net/whitepaper/20191010/Combining-Digital-Pathology-with-Artificial-Intelligence.aspx.

Other White Papers by this Supplier