Hyderabad researchers develop AI model for breast cancer detection

Hyderabad: Researchers in Hyderabad have developed an Artificial Intelligence model that detects breast cancer using mammogram images.

According to a report by Times of India, the model, developed by Kumari Jelli and Pavan Kumar Pagadala of the Department of Computer Science and Engineering at Koneru Lakshmaiah Education Foundation, has shown 95 percent accuracy while requiring less computing power than comparable models. Their research is titled “Fuzzy-based residual shufflenet-based breast cancer detection using mammogram images.”

The researchers named their model Fuzzy RS-Net, designed to improve image analysis for identifying signs of breast cancer in mammograms. It was tested on the Curated Breast Imaging Subset of the Digital Database, where it recorded 95 per cent accuracy, 93 per cent specificity and 95 per cent sensitivity. Sensitivity measures how well the model detects cancer cases, while specificity measures how well it detects non-cancer cases.

After testing across multiple mammography datasets, the researchers found that Fuzzy RS-Net required less processing time and memory than several other AI models used for comparison.

The model works by first reducing noise in mammogram images, then identifying sensitive areas and assessing patterns within them to classify the images. Statistical tests showed the improvement over comparison models was significant, with p-values below 0.05.