Dani Kiyasseh

Title: Innovations by Dani Kiyasseh in Cardiac MRI Segmentation

Introduction

Dani Kiyasseh is an accomplished inventor based in Marina del Rey, CA (US). She has made significant contributions to the field of medical imaging, particularly in the area of cardiac MRI segmentation. Her innovative work focuses on enhancing the accuracy and efficiency of image analysis through advanced machine learning techniques.

Latest Patents

Dani Kiyasseh holds a patent for "Meta-learning for cardiac MRI segmentation." This patent describes methods and systems for image segmentation, where a machine learning model is applied to a set of images to generate results. The results are obtained as a probability map for each image in the set. The model is trained by accessing a set of labeled images, each associated with a label indicating the location of a feature within the respective image. An initial set of parameters is accessed, and an encoder is initialized with these parameters. The encoder is then applied to the set of labeled images to predict feature locations. The initial parameters are updated based on the predictions and the associated labels. The updated parameters, along with an additional set generated using unlabeled images, are aggregated to improve the model's performance.

Career Highlights

Dani Kiyasseh is currently employed at Merck Sharp & Dohme Corp., where she continues to develop innovative solutions in medical imaging. Her work has the potential to significantly impact the field of cardiology by improving diagnostic accuracy and patient outcomes.

Collaborations

Dani collaborates with talented professionals in her field, including Antong Chen and Albert Joseph Swiston. These collaborations enhance her research and contribute to the advancement of technology in medical imaging.

Conclusion

Dani Kiyasseh's contributions to cardiac MRI segmentation through her innovative patent demonstrate her commitment to advancing medical technology. Her work exemplifies the intersection of machine learning and healthcare, paving the way for improved diagnostic tools.