New AI system can possibly identify the type of lung cancer



NEW YORK: A new artificial intelligence (AI) system can accurately identify lung cancers by analyzing the tumors of patients, scientists say.

The system, described in the journal Nature Medicine, was able to distinguish between adenocarcinoma and squamous cell carcinoma with 97 percent accuracy – two types of lung cancer that pathologists sometimes had difficulty in analyzing without confirmatory tests.

"Our study provides strong evidence that an AI approach will be able to directly determine the cancer subtype and the mutation profile to allow patients to start faster with targeted therapies," said Aristotelis Tsirigos, associate professor at New York University (NYU). ) in the U.S.

The AI ​​tool was also able to determine whether abnormal versions of 6 genes associated with lung cancer – including EGFR, KRAS and TP53 – were present in cells, with an accuracy ranging from 73 to 86 percent, depending on the gene .

Such genetic changes or mutations often cause the abnormal growth seen in cancer, but can also change the shape and interactions of a cell with the environment, and provide visual clues for automated analysis.

Keep cancer in the bay: eat garlic, turmeric and other anti-carcinogenic foods

Anti-carcinogenic herbs & spices

Feb 3, 2018

Bollywood actress Sonali Bendre has undergone treatment in New York after the diagnosis of metastatic cancer.

Although you are never sure what the beginning of this deadly disease can cause, it is important to take the necessary precautions to stay safe. There are certain food products that have anticarcinogenic properties and can help to keep cancer at bay.

These herbs and spices from your kitchen cabinet can stimulate your immune system to control the deadly disease.

Dr. Manoj Kutteri, director of the Atmantan Wellness Center, shares a list of ingredients that can help you fight the disease.

Also read: Know about metastatic cancer of which Sonali Bendre is diagnosed

Determining which genes have changed in each tumor has become vital with the increased use of targeted therapies that only work against cancer cells with specific mutations, researchers said.

For example, about 20 percent of patients with adenocarcinoma are known to have mutations in the gene-epidermal growth factor receptor or EGFR, which can now be treated with approved drugs.

However, the genetic tests currently used to confirm the presence of mutations can take weeks to return results, researchers said.

The study found that about half of the small percentage of tumor images misclassified by the study AI study were also wrongly classified by the pathologists, showing the difficulty of differentiating between the two types of lung cancer.

On the other hand, 45 out of 54 of the images misclassified by at least one of the pathologists in the study were assigned to the right type of cancer by the machine learning program, suggesting that AI could provide a useful second opinion. offer.

"In our study, we were excited to improve accuracy at the level of the pathologist and to show that AI can detect previously unknown patterns in the visible characteristics of cancer cells and the tissues around them," said Narges Razavian, an assistant -professor at the NYU.

The team plans to continue to train its AI program with data until it can determine which genes have been mutated in a particular form of cancer with more than 90 percent accuracy. At that time they will seek permission from the government to use the technology clinically and in the diagnosis of different types of cancer.


Source link

Leave a Reply