AI Offers Accurate Prediction of Chemotherapy Effectiveness Prior to Breast Cancer Surgery | Be Korea-savvy

AI Offers Accurate Prediction of Chemotherapy Effectiveness Prior to Breast Cancer Surgery


(image: Korea Bizwire)

(image: Korea Bizwire)

SEOUL, Dec. 7 (Korea Bizwire)South Korean researchers have developed new technology that uses artificial intelligence (AI) to anticipate the effects of chemotherapy for breast cancer patients.

The Korea Institute of Radiological Medical Sciences announced recently that an AI deep learning mechanism can be used to anticipate the effects of chemotherapy for breast cancer patients before they begin treatment.

Local advanced breast cancer, in which cancer cells have spread to the lymph nodes or other tissues inside the breasts, poses a high risk of recurrence and spread.

Chemotherapy, therefore, is carried out prior to surgery to reduce the size of tumors and maximize the chance of a successful surgery and retaining the breast.

However, repeated PET/CT and MRI scans and delayed assessments thereof have been dragging down the whole process.

The research team conducted PET/CT and MRI scans on 56 breast cancer patients prior to chemotherapy, and consulted both medical specialists and deep learning AI to anticipate the effects of the treatment.

Prior to the test, the AI studied the PET/CT and MRI scans of both groups where chemotherapy was either effective or ineffective.

The results showed that the AI demonstrated higher accuracy than medical specialists.

Human specialists demonstrated 84 percent accuracy in anticipating the effects of chemotherapy by studying PET/CT scans, and 61 percent for MRI scans. The AI, in contrast, demonstrated 97 percent accuracy for PET/CT scans and 85 percent for MRI scans.

Experts see this new technology will expedite the treatment process and maximize convenience for patients.

Ashley Song (ashley@koreabizwire.com)

Leave a Reply

Your email address will not be published.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

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>