President and Chief Executive Officer Accurate Dx. Passionate about personalized medicine through AI-supported pathology.
getty
The concept of “who gets what type of cancer treatment” might seem like a movie plot or question decades ago, but unfortunately, even in 2022, it’s still a daily dilemma the world faces. While underlying causes, detection, and treatment options vary by specific type of cancer, the central role of pathology is to properly diagnose and characterize cancer to determine subsequent management and treatment options. The problem is that anatomic pathology, the technique commonly used to diagnose cancer, is inherently subjective, with well-documented variability in accuracy and reproducibility. The promise of greater accuracy and reproducibility is driving scientists and commercial organizations to apply artificial intelligence (AI) platforms to examine tens of thousands of features in tissue and analyze every cell on a slide to ensure highly reliable results.
Standardization saves lives.
There will be an estimated 18.1 million cancer cases worldwide in 2020, but cancer grades cannot be determined with complete objective accuracy because they depend on how humans interpret cells under a microscope. In short, pathology to date has been largely subjective.
For example, patients with slow-growing tumors may be overly referred to as having more advanced and advanced cancers, leading to unnecessary surgery, chemotherapy, or radiation therapy. On the other hand, if the morphological features of the tumor are considered by the pathologist to be less invasive, the patient may be underestimated and thus not be given the appropriate, potentially life-saving treatment that the patient needs.
AI has already been adopted and is positively impacting radiology and patient monitoring. More recently, several organizations have emerged that combine industry knowledge and medical science expertise with cutting-edge computer science and engineering. These include PathAI, Paige, Ibex and my own company PreciseDx.
The problem is most acute in resource-limited settings.
While the subjectivity of pathology is a general problem, this problem becomes even more critical in resource-limited settings. For example, African countries have one of the worst shortages of pathologists in the world. According to Dark Daily, as of 2016: “Mozambique, with a population of 25 million, has only four pathologists. Botswana has only three pathologists serving its population of 2.1 million.” In Mexico, its estimated 131 million citizens have only 1,800 pathologists. scientist. Even the United States is not immune to a shortage of qualified pathologists, with the number of providers falling nearly 18% between 2007 and 2017.
These striking data should lead us to think: How does limited (or total lack) access to pathologists affect care, and what can we do about it? Through the adoption and adoption of artificial intelligence tools, pathologists will benefit from greater efficiency and the ability to participate in directing high-quality care in other parts of the world.
Connecting patients with personalized treatments drives the best outcomes. The first step in determining the best treatment is to be able to accurately and objectively assess the risk of disease progression, metastasis, or death.The ability to use AI-enabled algorithms to determine this risk without the need to contact a locally qualified pathologist begins the process of advancing healthcare in all regions of the world, including the United States
In resource-constrained settings, the application of AI-enhanced pathology algorithms addresses the inability of pathologists to accurately determine which patients are more critical than others. Remote access to experts through telepathology to review and analyze pathology slides, combined with AI-supported risk assessment, can provide the information needed to make better treatment decisions. Without access to specialists, local healthcare providers are often forced to make unwarranted decisions about who has access to the limited treatment options they can offer.
These providers deserve the support tools necessary to optimize the distribution of care across their patient populations.
Technology can open up healthcare across the globe.
Through technology—particularly the digitization of slides and the use of artificial intelligence—we can make pathology and pathology insights accessible to nearly everyone. AI has been successful in many areas of healthcare, and in the past five years several new technologies have been introduced specifically to help pathologists and oncologists. These new technologies include multiple oncology solutions from Philips, histology solutions from Leica Biosystems, and detection and monitoring solutions from Hamamatsu Photonics.
While still in its early stages, artificial intelligence could be trained to “data mine” millions of data points to identify and quantify key morphological and cellular characteristics of each cancer type. These techniques are being applied using whole slide imaging (WSI), so every cell on the slide can be viewed, not just the sample on the slide. This creates a statistically more robust set of data, and the AI can display the results in absolute and percentile formats. Access to these data and the resulting informative analyzes has the potential to support pathology and oncology in exciting new ways by providing highly accurate, objective patient-specific guidance.
The potential lies in improving care for everyone. This is not an instant fix; however, regions with limited resources can best take advantage of this new technology for organizations dealing with scanners by providing technicians with the necessary tools and training.
Members of the healthcare industry should keep an eye out for upcoming publications and join their colleagues to be early adopters of processes where AI will improve outcomes and increase efficiency.
To take full advantage of this process, the industry will need to employ an infrastructure with the bandwidth to send large images and image-based annotation results to fully leverage and integrate the power of AI in healthcare.
With the development of technology, the importance of geographical location for the quality of care is greatly reduced. Internet connectivity can bring higher levels of pathology accuracy and standardization to everyone. As one of the many opportunities for AI in healthcare, the incorporation of AI into cancer pathology holds great potential to support highly personalized treatments and improve patient outcomes.
The Forbes Technology Council is an invite-only community for world-class CIOs, CTOs, and technology executives. Am I eligible?
