Better Breast Cancer Prevention Rates Are on Their Air Thanks to Machine Learning

breast cancer symbol

U.S. scientists tapped into the new territory of artificial intelligence to improve the health care system. And they are on to great improvements in breast cancer treatments. They devised a machine learning tool that can help professionals identify high-risk breast lesions better. The main potential benefit of this study is a major reduction in unnecessary surgical interventions.

Doctors Resort to Over-Treatment Due to Lack of Better Diagnosing Equipment

At the moment, doctors can identify lesions at the level of the chest that carry risks of cancer development through biopsies. However, they can’t measure the extent of the threat precisely. Therefore, often times women have to go through intrusive surgeries rather for breast cancer prevention than improved health condition.

Nonetheless, women’s currently option to curb cancer risk is expensive and can lead to further health complications. However, researchers’ experience in machine learning can alleviate women of such procedures.

Through the new AI tool, doctors will be able to monitor high-risk lesions in real time. No matter how threatening some lesions may appear, there are always chances that they won’t jeopardize the patient’s life. Therefore, recording more data on these risks can help doctors avoid over-treatment.

The Machine Learning Tool Can Identify High-Risk Lesions that Might Remain Benign

The AI device received extensive training by uploading it with information regarding different types of lesions at the chest level. Bots can afterward read data to detect patterns that confirm an accurate diagnosis. The machine needs a variety of information to be precise such as family history, demographics, pathology results, and biopsies.

In their study, researchers uploaded 20,000 pieces of data to the AI system. Afterwards, the model received the patient history of two-thirds of 1,000 women with high-risk lesions. Only 4% of these cases didn’t involve surgical procedures.

In the end, the AI system took an exam where it diagnosed the last third of the documented cases of breast cancer. Out of 335 patients who presented high-risk symptoms, the machine identified 37 lesions that led to cancer. The result was one case short from the correct answer. Nonetheless, such an AI assistance would have prevented one-third of the women from undergoing unnecessary surgery as their lesions remained benign.

Image source: 1


Leave a Reply

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

shares