Edit/Keich
With the number of cancer cases worldwide likely to increase by 60% over the next 20 years, and the number of pathologists is decreasing relative to this diagnostic need, there is an urgent need to accelerate the advancement of pathology. The fact that pathology – albeit the basis of diagnosis – has not yet been digitized further exacerbates the problem.
Cancer diagnostic technology company Paige uses machine learning to help pathologists make faster and more accurate cancer diagnoses based on images of tissue samples. Paige technology also helps life sciences companies evaluate and design personalized treatments for patients.
<h1 class="pgc-h-arrow-right" data-track="4" > machine learning to make faster and more accurate cancer diagnoses</h1>
In 2018, Paige spun off from Memorial Sloan Kettering Cancer Center and announced its official opening. At the same time, Paige also announced the completion of a $25 million Series A funding round. The startup focuses on detecting prostate and breast cancers, with plans to expand to other diseases.
Last December, the company received two SOFTWARE CE certifications for breast and prostate cancer, including the ability to rate tumor samples, provide prognosis and guide treatment plans.
Paige's prostate cancer diagnostic technology, the first AI product designed for pathology or oncology, was awarded the U.S. Food and Drug Administration's (FDA) Breakthrough Device designation.
In January, the startup raised $100 million in Series C funding to facilitate its development as a biomarker for clinical applications.
"This investment reaffirms the enormous potential of the Paige platform in clinical and biopharmaceutical drug development applications," Leo Grady, CEO of Paige, said in a statement. "These funds will allow us to build more AI-based products both inside and outside of oncology, deliver these products to labs and clinicians around the world, and invest in our talent in engineering and commercial functions."
Paige wants to help doctors detect cancer earlier, come up with effective treatment options, and prevent unnecessary surgery due to false positives. In May, the company announced a partnership with Quest Diagnostics to develop software products that can uncover previously potentially unknown cancer markers, helping pathologists and oncologists better diagnose prostate, breast and other types of cancer.
In June, Paige, the University of Oxford and the UK's National Health Service (NHS) regional partners won the prestigious Phase 4 Healthcare Artificial Intelligence Award from the NHS Accelerated Access Collaborative for evaluating the Paige Prostate Prostate Cancer Detection System.
Paige Prostate is a clinical-grade AI-based diagnostic software system that helps pathologists detect, grade, and measure prostate tumors in biopsy tissue obtained from patients at risk of developing prostate cancer,1 and other similar systems. These adoption guidelines will enable AI technologies and advanced algorithms to be further replicated in the NHS to help diagnose complex diseases.
"Paige is proud to work with this multidisciplinary team of experts to demonstrate the impact of digital pathology tools in routine clinical use," said Dr. Leo Grady, CEO of Paige. "Together with our partners, we look forward to the possibility of ushering in a new era of clinical diagnostics driven by AI technology that benefits patients and cellular pathology labs across the NHS."
<h1 class="pgc-h-arrow-right" data-track="14" > a strong AI technical team</h1>
Paige's machine learning team consists of PhDs, software engineers, and high-quality computing experts with deep learning expertise. Its founder, Dr. Thomas Fuchs, is known as the "father of computational pathology".
Thomas Fuchs, founder of Paige.
"Oncology is a complex, specialized medical discipline, and only detailed, massive amounts of data coupled with decades of experience can ensure that computer models accomplish the daunting task," said Dr. Fuchs, "and we will enable computational pathology to scale to the scale needed to achieve intelligent quantitative clinical models and promote the widespread use of digital pathology."
Dr. David Klimstra, one of the founders of Paige.
Dr. David Klimstra, co-founder of Paige, says cancer patients should get an accurate diagnosis as quickly as possible, but our current approach is time-consuming, expensive and subjective. "I believe that Paige will help pathologists find diseases better and faster." With computational pathology, pathologists can shift their work to more complex tasks, such as combining histological examination results with other diagnostic analyses," he adds.
Paige CEO Leo Grady.
Leo Grady is now the CEO of Paige. Paige wants to help doctors detect cancer earlier, come up with effective treatment options, and prevent unnecessary surgery due to false positives.
"Our technology is really transformative," Grady says. "This will help pathologists improve efficiency, make higher quality decisions, and return results to patients more quickly." It will also ultimately reduce the cost of the healthcare system."
Paige is now looking to use its technology beyond diagnostics. Grady said the company hopes to be able to predict a patient's treatment plan by training AI to understand the correlation between certain types of tumors and the efficacy of certain drugs.
Grady joined Paige in 2019 and said, "Paige has given him the opportunity to impact the world through artificial intelligence."
Cancer pathology today involves examining tissue samples under a microscope for diagnosis. But tissue and disease markers can vary widely, so pathologists often seek a second opinion or more testing.
Paige's technology simplifies this process through digitalization. The company has exclusive rights to tens of thousands of analyzed pathology slides from Memorial Sloan Kettering and scans them into its system to create a database of high-resolution images. The company's proprietary machine learning system is trained to detect patterns in images that are associated with disease prognosis.
When a new tissue scan enters, the system can classify it according to its training. Instead of calling colleagues or doing additional tests, Grady said, pathologists can use AI-based systems to make decisions faster and easier.
The findings, https://pubmed.ncbi.nlm.nih.gov/33904171/, recently published by him and his colleagues, show that Paige's prostate test reduced the diagnosis time by about 65 percent and found prostate cancer in four patients who were not initially diagnosed by three experienced histopathologists.
"Our vision is to create a world where patients have a deep and comprehensive understanding of their biology, translating diagnosis into effective personalized treatment," said Grady.
Reference: https://spectrum.ieee.org/ai-diagnosis-cancer
Official website: https://www.paige.ai/