Shaoyan Pan

Title: Shaoyan Pan: Innovator in Machine Learning for Medical Imaging

Introduction

Shaoyan Pan is a prominent inventor based in Decatur, GA (US). He has made significant contributions to the field of machine learning, particularly in medical imaging. His innovative work focuses on enhancing the accuracy and efficiency of lesion segmentation through advanced transformer-based models.

Latest Patents

Shaoyan Pan holds a patent for a system titled "Multi-dimension unified swin transformer for lesion segmentation." This invention outlines a method for multi-stage training of a transformer-based machine-learning model. The system operates through three distinct stages of training. In the first stage, a transformer encoder is pre-trained using an unlabeled 3D image dataset. The second stage involves fine-tuning this pre-trained encoder with a labeled 2D image dataset. Finally, the third stage further fine-tunes the encoder using a labeled 3D image dataset.

Career Highlights

Shaoyan Pan is currently employed at Merck Sharp & Dohme LLC, where he continues to develop innovative solutions in the realm of healthcare technology. His work is instrumental in advancing the capabilities of machine learning applications in medical diagnostics.

Collaborations

He collaborates with talented colleagues, including Yiqiao Liu and Antong Chen, to push the boundaries of research and development in their field.

Conclusion

Shaoyan Pan's contributions to machine learning and medical imaging exemplify the impact of innovation in healthcare. His patent and ongoing work at Merck Sharp & Dohme LLC highlight his commitment to improving medical technologies.