Cloud Hosting

A recent study published in The Lancet Oncology suggests that artificial intelligence (AI) could be as effective as two human radiologists when it comes to reviewing breast cancer scans. The trial involved 80,000 women between the ages of 40 and 80, with an average age of 54 years old.

During the trial, a combination of AI detection and human radiologists conducting reviews was used to screen 39,996 patients, while the standard process of two human radiologists was used for the remaining 40,024 patients. The AI review detected 41 more cases of breast cancer compared to the radiologists, with 19 classified as invasive and 22 as in situ cancers.

However, the study also raised concerns about potential overdiagnosis, as AI-supported screening found 60 in situ cancers compared to the 38 found in standard screening. On the positive side, the AI-supported screening did not result in more false-positive results than standard screening, with an average false-positive rate of 1.5% in both groups.

In addition to its detection capabilities, AI could significantly reduce radiologists workload. The study estimates that incorporating AI into the screening process could reduce the need for two radiologists to review mammograms to just one. This becomes crucial as the field of radiology is expected to face a shortage of professionals by 2024.

Lead author Kristina Lng, Ph.D., emphasized that more research is needed before AI can be fully implemented in mammography screening. Further trials and evaluations are necessary to assess the technologys effectiveness, cost-efficiency, and the extent to which it can assist radiologists.

While the trial is ongoing, the potential impact of AI in reducing the number of missed breast cancer cases and improving efficiency in screenings is being closely monitored. Breast cancer is the most prevalent cancer worldwide, underscoring the importance of early detection and treatment.

In conclusion, the trial results indicate the promise of AI in breast cancer screening, suggesting that it has the potential to enhance detection rates and reduce radiologists workload. However, further research and evaluation are required before widespread implementation can occur.

Read the rest here:
Artificial Intelligence Shows Promise in Breast Cancer Screening - Fagen wasanni