Amazon Web Services takes the fight against cancer to the cloud

A subsidiary of shopping giant Amazon that provides cloud solutions to the corporate field Amazon Web Servicesbrings the fight against cancer to the cloud.

Sharing a detailed post on this subject, Amazon Web Services stated the following: Humanity has been struggling with this disease since cancer was first described in Ancient Egypt approximately 5000 years ago. Since then, the heroic efforts of researchers, healthcare providers, and caregivers have helped make major advances in understanding, diagnosis, treatment, and care of the disease. But the threat of cancer continues to grow. The disease claims nearly 10 million lives worldwide each year, and the World Health Organization predicts that there will be more than 35 million new cases of cancer by 2050. This figure represents a 77 percent increase compared to the estimated 20 million cases in 2022. The world continues to search for a cure for cancer, and technology’s ability to accelerate the pace of innovation while lowering costs is changing the dynamic of this fight in several ways.

YOU MAY BE INTERESTED IN

In particular, the scalability and relative affordability of cloud technology enables healthcare providers and patients to fight cancer more effectively at every stage, from diagnosis to remission. Artificial Intelligence and Machine Learning (AI/ML) is accelerating the speed of data analysis by generating more comprehensive insights from genetic, clinical and image data. Providing secure access to data when needed enables researchers around the world to collaborate and share important findings in real time. Remote monitoring and telemedicine enable patients to take a more active role in the fight against cancer.

AWS Chief Medical Officer and Director of Global Healthcare and Nonprofits Dr. Rowland Illing“Technology is not a magic wand that will end cancer, but a tool that paves the way for real heroes. “I have experienced firsthand how technology can make research more efficient and effective, diagnosis faster and more accurate, accelerate the discovery and development of new medicines, and increase accessibility through remote care and more inclusive clinical trials,” he said.

Research: Mapping cancer through genome sequencing

Cancer is a disease caused by the uncontrolled division of abnormal cells, and in the vast majority of cases, this is caused by some type of genetic change. This means that most cancers are a DNA disease. The challenges of cancer treatment arise from the fact that genes change as they mutate and continue to change as tumors grow. Each type of cancer may have a mutational signature, but the genes in each tumor and the cells within each tumor differ from each other.

Mapping this genetic diversity requires generating and analyzing enormous amounts of data. To equip doctors with the insights they need, this data needs to be collected in a way that protects patient privacy, stored securely, and transformed through visualization and analysis so that it can guide both patient treatment and the development of new drugs.

The Cancer Genome Atlas collected data from nearly 20,000 tumors and comparative normal tissue samples from 11,328 patients across 33 cancer types. Visualizations produced through this data show how types of cancer develop and spread, with elements such as patterns in the cell where cancer arises, the role of different virus variants in triggering mutations, and signaling pathways in the body that can be leveraged for effective treatments. Atlas, part of AWS’s Registry of Open Data, enables researchers around the world to expand their knowledge base.

Scaling this data-driven approach to decoding cancer depends on cost-effectively sequencing each potential patient’s DNA. California-based Ultima Genomics also aims for this. Ultima has developed a high-throughput, cost-effective next-generation sequencing instrument on AWS that can sequence the entire human genome for as little as $100, which would have cost $1,000 to extract a decade ago.

Accelerating and democratizing diagnostics

Genome sequencing of the kind pioneered by Ultima Genomics could significantly speed up the detection of cancer by making genetic analysis affordable. Using Amazon Elastic Compute Cloud (Amazon EC2) to accelerate genomic data processing, the Munich Leukemia Lab reduced processing time from 20 hours to three hours and increased both diagnostic speed and accuracy.

However, most patients do not encounter a cancer diagnosis through an in-depth genetic analysis. Many only visit their doctor for a regular checkup or for an ailment seemingly unrelated to cancer. Identifying the subtle visual signals of this disease, even in non-cancer-focused doctor appointments, is one of the most important contributions of artificial intelligence (AI) to improving cancer survival rates.

Collaboration with Ibex and AWS to bring AI diagnostic tools to pathology laboratories through the Navify platform by Michael Rivers, Vice President of Digital Pathology, Roche Tissue Diagnostics“Cancer diagnosis is still mostly made by pathologists looking at tissues on slides with microscopes. “Digitalization of these slides can enable artificial intelligence-based image analysis, helping us make a diagnosis and determine the patient’s treatment plan,” he said.

According to Rivers, one of the most striking ways AI can speed up diagnosis is by training Fundamental Models (FMs) under detailed instructions, allowing them to develop their own approaches to analyzing images. “Explainable AI is a really important component of the solutions we develop,” Rivers said. “We provide the pathologist with heat maps and illustrative layers to show how the algorithm interprets an image, but the pathologist always has final control over the final diagnosis,” he said.

Breast cancer is a worldwide concern, but in some populations it can be even more difficult to diagnose. For example, the relatively denser breast tissue of Asian women has historically made early diagnosis more difficult.

FathomX, a Singapore-based artificial intelligence company originating from the National University of Singapore, developed the FxMammo algorithm that addresses this problem. Using artificial intelligence technology, FxMammo is designed to increase accuracy in identifying masses and abnormalities in dense breast tissue, enabling more reliable and cost-effective early detection and reducing false positives by up to 75.5 percent.

Changing the numbers in cancer drug discovery

While research and diagnosis are key to fighting cancer, the course of an individual’s disease ultimately depends on whether the right treatments are available. According to PhRMA, it takes an average of 10-15 years and $2.6 billion to bring a new drug to market, and despite this level of investment, only 12 percent of new drug molecules receive approval from the United States Food and Drug Administration (FDA). .

Delivering inclusive and personalized cancer care at scale

The ability for healthcare providers to combine information about a patient’s cancer obtained at the DNA level with more personalized treatment options could improve treatment outcomes nationally.

Genomics England is sequencing the DNA of cancer patients and their tumors to provide more information for treatment. This process involves identifying the genes that help cancer spread, the treatments most likely to affect them, and the possible side effects for each patient. A study backed by data from Genomics England and published this year in the journal Nature Medicine found that treatment for nine out of ten brain tumors and most bowel and lung cancers could be guided by genetic information. Personalized treatments for cancer patients may include targeted therapy drugs, a form of chemotherapy that focuses on changes that differentiate cancer cells and avoids the range of side effects associated with cruder approaches that also kill healthy cells.

Genomics England is building on this work using Anthropic’s Claude models on Amazon Bedrock to help researchers identify associations between genetic variants and disorders and diseases, including cancer.

Empowering cancer patients and caregivers around the world

Fighting cancer is not just about treating the disease faster and more effectively. As the disease spreads, it will be equally important to improve access to treatment for all and ensure that future treatments are developed with all cancer patients in mind.

The limited availability of expert ultrasound operators in remote areas of China is one of the major obstacles to diagnosis. To help fill this gap, Shangyiyun led Dr. It developed a breast cancer screening AI assistant called J. Dr. J can automatically detect and label lesions and upload ultrasound video and images to the cloud for experts to analyze. This helps scale the reach of breast cancer screening and quickly identify cases that need investigation. Thanks to AWS’s computing and storage services, Dr. J is able to provide stable, efficient and accurate scanning services for institutions and users. AWS also Dr. It also provides a solid data security foundation for J’s use on a global scale.

Hurone AI uses Large Language Models (LLMs) and predictive AI built on AWS to help close gaps in cancer care in sub-Saharan Africa. By establishing a two-way messaging system on AWS, the company allows oncologists to monitor and support cancer patients even when local infrastructure is inadequate. The company is exploring how the same model can be used to identify potential candidates for clinical trials, which could play a vital role in developing treatments for everyone.

Kingsley Ndoh, Founder and CEO of Hurone AI“If you look at all the cancer drugs approved by the Federal Drug Administration (FDA) in the last 20 years, less than 5 percent of the clinical trial participants for those drugs were of African or Hispanic descent. Currently, there is only one oncologist for every 3,000 cancer patients in sub-Saharan Africa and only one oncologist for every 1,000 patients in Latin America. “These patients eventually face side effects from medications and it is difficult for them to access the ongoing care and support they need,” he said.

Expanding the reach of cancer care geographically is one important way to increase access and support more patients. Another way is to expand home care services. This enables patients to fight cancer on their own terms and in their own space, increasing their well-being, sense of dignity and autonomy.

The complexity of cancer that challenges medicine stems from the fact that each case is unique, just like the genes in each tumor. The cloud helps turn this challenge into an advantage by making the fight against cancer more personalized and inclusive. In addition to making it possible to develop more personalized, targeted and effective treatments, it also provides important psychological benefits. “Helping patients feel understood and supported not only enables more targeted treatments but significantly contributes to both patient experiences and treatment outcomes.”

lgct-tech-game