Dongao MA

Title: Dongao MA and AI for Medical Image Classification.

Introduction

Dongao MA is an inventor associated with Arizona State University in Tempe, California, United States. His work includes a patent on training artificial intelligence models for medical image classification by accruing and reusing knowledge across tasks.

Latest Patents

The patent listed for MA is “Systems, methods, and apparatuses for accruing and reusing knowledge (ARK) for superior and robust performance by a trained AI model for use with medical image classification.” It describes systems that receive medical images from multiple publicly available datasets and identify expert annotations embedded in those images, even when the datasets use inconsistent labeling.

The disclosed approach uses task-head classifiers to learn from the annotations and generate an interim AI model. It then scales that model to additional tasks through cyclical pre-training and task-specific learning objectives. During training for a target application, task-revisitation training prompts the model to revisit prior tasks and reuse accumulated knowledge. MA is credited with 1 patent.

Career Highlights

MA’s listed patent focuses on addressing heterogeneous labels and improving how AI models transfer knowledge across medical image classification tasks. The invention’s training strategy is designed to draw on previously learned information while adapting the model to a specific target task.

Collaborations

MA is associated with Arizona State University. His listed coworker is Jianming Liang.

Conclusion

Dongao MA’s patent presents a knowledge-reuse approach to training AI models for medical image classification, with the aim of supporting robust performance across varied datasets and tasks.