	US 12,450,738 B2
	Automated multi-class segmentation of digital mammogram
Susan Ng, Villanova, PA (US); Vincent Yun Dong, Philadelphia, PA (US); Tristan Douglas Maidment, Villanova, PA (US); Lucas Rodrigues Borges, Sao Paulo (BR); Katherine Martling Hopkins, Lancaster, PA (US); Johnny Kuo, Lancaster, PA (US); Albert Milani, Merion Station, PA (US); and Peter A. Ringer, Allentown, PA (US)
Assigned to REAL TIME TOMOGRAPHY, LLC, Villanova, PA (US)
Filed by Real Time Tomography, LLC, Villanova, PA (US)
Filed on Dec. 7, 2022, as Appl. No. 18/062,712.
Claims priority of provisional application 63/265,037, filed on Dec. 7, 2021.
Prior Publication US 2023/0177686 A1, Jun. 8, 2023
Int. Cl. G06T 7/00 (2017.01); A61B 6/00 (2006.01); A61B 6/50 (2024.01); G06T 7/11 (2017.01)

CPC G06T 7/0012 (2013.01) [A61B 6/502 (2013.01); A61B 6/5235 (2013.01); G06T 7/11 (2017.01); G06T 2207/20081 (2013.01); G06T 2207/30068 (2013.01)]

20 Claims

1. A method of automatically analyzing a digital breast image, the method comprising, by a processor:

receiving an image of a breast;

using a machine learning model to process the image to generate a breast mask for an entire region of the image in which the breast appears;

using the machine learning model to process the image to generate at least a second mask for a region of the image in which a feature other than the breast in its entirety appears;

combining the breast mask and the second mask to yield a combined mask image;

using the combined mask image to generate a measurement of quality of the received image by:

generating a line from the mask of the nipple to a nearest border point on a border of the mask of the pectoralis muscle,

measuring an angle formed between an orientation of the line and another line specifying the border of the mask of the pectoralis muscle, and

generating the measurement of quality based on the measured angle; and

generating and outputting a report that includes the measurement of quality.