AI helps in early detection of pancreatic cancer

Photo credit: Unsplash+

An insight into the study

Here’s some exciting news. A team of scientists from Harvard Medical School and the University of Copenhagen have developed an artificial intelligence (AI) tool.

What is special about this tool? It can identify people at high risk of developing pancreatic cancer up to three years before diagnosis.

how does it do that The patient’s medical records are used. Their findings were published May 8 in Nature Medicine.

Why is that important?

Pancreatic cancer is one of the deadliest cancers worldwide. It usually occurs in advanced stages, when treatments are no longer as effective.

This means that survival rates are low. We need better tools for early detection. This new AI tool could be game-changing.

Today we do not have the tools to screen large groups for pancreatic cancer.

We screen people with a family history and certain genetic mutations in a more targeted way. However, this approach ignores some cases.

AI: A potential game changer

The AI ​​tool developed by the researchers could help us specifically identify those most at risk.

Chris Sander, a faculty member at Harvard Medical School and co-senior investigator on this study, emphasized the potential of the tool.

He said it could improve the decision on who needs further testing. This is crucial as these tests can be invasive, expensive and come with their own risks.

Using AI could help us detect pancreatic cancer earlier. This could lead to earlier treatment, better outcomes and longer patient lifespans.

How the study was conducted

The researchers trained the AI ​​algorithm using two separate data sets. This included 9 million patient records from Denmark and the United States.

The AI ​​model was asked to look for signs of pancreatic cancer in the recordings. For this purpose, combinations of disease codes and the time of their occurrence were examined.

The researchers tested different versions of the AI ​​models. They wanted to see how well they could identify people who are at risk of developing the disease within different time periods: six months, one year, two years and three years.

Breakdown of current screening methods

Screening for common cancers like breast, cervical and prostate cancer has become easier with powerful techniques like mammography, Pap smear and blood tests.

In the case of pancreatic cancer, however, the situation is different. The detection methods are more complicated and expensive.

Currently, doctors mainly examine family history and genetic mutations. While these are important indicators, they are often overlooked by many patients.

The new AI tool could be used for all patients, not just those with known genetic risk or known family history.

The challenges of pancreatic cancer

The pancreas is located deep in the abdomen, making it difficult to access for testing. It is often referred to as “the angry organ” because of its sensitivity.

If pancreatic cancer is diagnosed early, about 44 percent of people survive at least five years.

But only 12 percent of cases are diagnosed this early. Once the cancer has spread, the survival rate drops dramatically to 2 to 9 percent.

This is how the AI ​​model works

Researchers created multiple versions of the AI ​​model. They trained them over a period of 41 years using the health records of 6.2 million patients from Denmark’s national health system.

Of these patients, 23,985 eventually developed pancreatic cancer.

The AI ​​model examined patterns based on disease progression. These are conditions that have occurred in a specific order over time.

Diagnoses such as gallstones, anemia, type 2 diabetes and other gastrointestinal problems indicated an increased risk of pancreatic cancer within 3 years.

Testing the AI ​​model

After training the AI ​​models, the researchers decided to put them to the test. They used a new set of medical records from the US Veterans Health Administration.

This dataset contained nearly 3 million records spanning 21 years and included 3,864 people diagnosed with pancreatic cancer.

However, the performance of the AI ​​tool dropped somewhat on this dataset. The researchers believe this is due to the shorter time frame of the US dataset and different patient profiles compared to the Danish dataset.

To improve accuracy, they trained the AI ​​model from scratch using only the US dataset. The results were much better.

This highlighted the importance of using high-quality and rich data, as well as the need to train AI models on local health data to account for population differences.

The future of AI in pancreatic cancer screening

The results of this study show that AI can be a powerful tool in the fight against pancreatic cancer.

By accurately identifying those at highest risk years before diagnosis, we can begin treatment earlier, potentially improving outcomes and extending patient lives.

However, it is important to remember that these AI tools are not standalone solutions. They should be used in addition to traditional screening methods and under the guidance of healthcare professionals.

Even though AI can predict risk, it doesn’t mean that everyone classified as a high-risk patient will develop pancreatic cancer.

In summary, AI represents a promising opportunity to change the course of pancreatic cancer. Healthcare use is only just beginning and this study represents a significant step forward in terms of potential applications.

As we continue to develop and refine these tools, we can only expect to see more breakthroughs in the future.

If you are interested in cancer, please read studies on how to reduce the spread of pancreatic cancer by almost 90% Green tea could help reduce the risk of death from type 2 diabetes

For more information on health, see the latest studies on new ways to extend the life expectancy of cancer survivors and their results Vitamin D supplements significantly reduce cancer death rates.

The study was published in natural medicine.

Copyright © 2023 Knowridge Science Report. All rights reserved.

Related Articles

Back to top button