Patent · US Active

Multi-scale deep reinforcement machine learning for N-dimensional segmentation in medical imaging

US10032281B1 · kind B1 · utility

36Cited by

0References

20Claims

0Family size

Assignee

Siemens Healthcare Limited · DE

Inventors

Florin-Cristian Ghesu · Erlangen, DE
Bogdan Georgescu · Plainsboro, US
Dorin Comaniciu · Princeton Junction, US

Key dates

Filing date	Jul 27, 2017
Grant date	Jul 24, 2018
Priority date	—
Expiry date	Jul 27, 2037

Classification

Technology area (CPC G)Physics
CPC primaryG06T2207/30004
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

Multi-scale deep reinforcement learning generates a multi-scale deep reinforcement model for multi-dimensional (e.g., 3D) segmentation of an object. In this context, segmentation is formulated as learning an image-driven policy for shape evolution that converges to the object boundary. The segmentation is treated as a reinforcement learning problem, and scale-space theory is used to enable robust and efficient multi-scale shape estimation. By learning an iterative strategy to find the segmentation, the learning challenges of end-to-end regression systems may be addressed.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.