Google creates an AI that helps in the detection of lung cancer

lung_cancer_model

The Artificial Intelligence researchers (AI) from Google working with Northwestern University Hospital have created an AI model that can detect lung cancer. According to data from the World Health Organization, lung cancer (malignant tissue in the lungs) is one of the most common causes of death worldwide, killing more than two million people a year and killing as many people. like breast cancer.

To help healthcare professionals, algorithms and computers can help develop advanced methods in the healthcare field.

However, for these tools to be useful, they must be accessible and understandable to everyone, doctors and patients, even without any technological or computer knowledge.

In fact, it should be known that the operation of all digital devices is based on computer programs and data.

The term "artificial intelligence" implies that these devices are capable of thinking for themselves. If programmed correctly, smart devices can evaluate the data provided and change processes or parameters "on the fly." Given enough information, they can 'learn' and modify their own code based on these new parameters.

For the past three years, teams at Google have been applying AI to problems in healthcare, from diagnosing eye diseases to predicting patient outcomes in medical records.

Today we are sharing new research showing how AI can predict lung cancer in ways that increase the chances of survival for many people at risk around the world.

Artificial Intelligence to improve the quality of life

Detailed in research published May 20 in Nature Medicine, the deep learning model was used to predict whether a patient has lung cancer, generating the lung cancer risk score and identifying the location of the lung cancer.

"By showing that deep learning can increase specificity without sacrificing sensitivity, we hope to generate more research and discussion about the role AI can play in changing the cost-benefit scale of cancer detection." , we can read on the Google blog.

"The artificial intelligence system uses 3D volumetric deep learning to analyze the entire anatomy of the chest scanner, as well as patches based on object detection techniques that identify regions with malignant lesions," says Shreeva Shetty. , technical manager of Google.

By analyzing a single scan, the model detected cancer (on average 5%) more frequently than a group of six human experts and was 11% more likely to reduce false positives (a false positive is the result of a decision in a bidirectional election, declared positive, where it is actually negative)

Radiologists often see hundreds of 2D images in a single CT scan, and cancer can be tiny and difficult to detect. We created a model that can not only generate the general prediction of lung cancer malignancy (viewed in 3D volume) but also identify subtle malignant tissue in the lungs (lung nodules). 

The model can also take into account information from previous scans, useful in predicting the risk of lung cancer because the growth rate of suspicious lung nodules can be indicative of malignancy.

lung_cancer_scan

These initial results are encouraging, but further studies will evaluate the impact and utility in clinical practice.

In our research, we tapped 45,856 null chest CT detection cases (some in which cancer was detected) from the NIH research data set from the National Lung Screening Trial and Northwestern University. We validated the results with a second data set and also compared our results with 6 US board certified radiologists.

Google announces that it will make the model available through the Google Cloud Healthcare API as it continues further testing and testing with partner organizations.

Source: https://www.blog.google/


Be the first to comment

Leave a Comment

Your email address will not be published. Required fields are marked with *

*

*

  1. Responsible for the data: Miguel Ángel Gatón
  2. Purpose of the data: Control SPAM, comment management.
  3. Legitimation: Your consent
  4. Communication of the data: The data will not be communicated to third parties except by legal obligation.
  5. Data storage: Database hosted by Occentus Networks (EU)
  6. Rights: At any time you can limit, recover and delete your information.