NYC Gazette

Katz School researchers have developed a mobile app utilizing AI technology to analyze mammogram images and provide real-time diagnostic predictions from smartphones. The findings, detailed in the white paper “Pink Guardian: A Gateway to Early Breast Cancer Detection,” will be presented at the 16th International Conference on Human System Interaction in Paris in July.

The app, named Pink Guardian, is built on the BC-InceptionV3 architecture within TensorFlow Lite, an open-source library from Google that supports machine learning models on mobile devices. This model excels at identifying potential signs of breast cancer from mammogram scans and offers users diagnostic predictions with confidence scores indicating the certainty of its analysis.

“Pink Guardian represents an important advancement in mobile health technology, offering a reliable and accessible tool for early breast cancer detection,” said Kanchan Subhashchandra Maurya, lead author of the paper and a student in the M.S. in Artificial Intelligence program. “By combining cutting-edge AI with a user-friendly interface, the app empowers users to take proactive steps in monitoring their breast health. This innovation promises to significantly impact healthcare practices and patient outcomes worldwide.”

The development process involved curating a diverse dataset of over 10,000 mammogram images from Kaggle's CBIS-DDSM repository for training and testing. The app uses advanced machine learning techniques such as deep learning and convolutional neural networks to classify images as benign or malignant accurately.

Key features include a user-friendly interface that allows users to upload mammogram images directly from their device’s gallery or camera. The app processes these images and provides diagnoses with confidence scores like “malignant (75%).” It aims to offer quick and accurate preliminary screenings optimized for various smartphones.

While Pink Guardian provides initial assessments, it advises users to seek professional medical evaluation for confirmation. Researchers are also focused on implementing security measures to protect user data privacy.

“The app marks a significant step forward in integrating AI into healthcare,” said Youshan Zhang, Maurya’s mentor, co-author of the paper, and assistant professor of artificial intelligence and computer science. “It not only enhances breast cancer detection accuracy but also democratizes access to advanced diagnostic tools. We’re committed to continuous improvement with plans to expand its capabilities to other medical imaging modalities such as ultrasound and CT scans.”