SYSTEMS AND METHODS FOR IMAGE PROCESSING

• US 20190318828A1
• Filed: 04/11/2019
• Published: 10/17/2019
• Est. Priority Date: 04/11/2018
• Status: Active Grant

• Current Assignee: Shanghai United Imaging Healthcare Co., Ltd.
• Original Assignee: Shanghai United Imaging Healthcare Co., Ltd.
• Inventors: LI, Yang, Zhao, Chunlin, Yu, Wenjun, Wang, Ce
• Granted Patent: US 10,943,699 B2
• Time in Patent Office: Days
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06T 15/205 Image-based rendering
• CPC Class Codes: G06T 2207/10072 Tomographic images
• CPC Class Codes: G06T 2207/20084 Artificial neural networks [ANN]
• CPC Class Codes: G06T 2207/30008 Bone
• CPC Class Codes: G06T 2215/06 Curved planar reformation of 3D line structures
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/0014 using an image reference approach
• CPC Class Codes: G16H 30/20 for handling medical images, e.g. DICOM, HL7 or PACS
• CPC Class Codes: G16H 30/40 for processing medical images, e.g. editing
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

The present disclosure provides computer-aided diagnosis systems and methods for detecting bone fracture. The method may include obtaining one or more medical images related to one or more bones. The method may also include obtaining a fracture detection model generated based on a machine learning model. The method may also include detecting, for at least one of the one or more medical images, one or more bone fracture regions of the one or more bones in the medical image using the fracture detection model.

Claims

1. A computer-aided diagnosis method for detecting bone fracture implemented on a computing device having one or more processors and one or more storage devices, the method comprising:

obtaining one or more medical images related to one or more bones;

obtaining a fracture detection model generated based on a machine learning model; and

for at least one of the one or more medical images, detecting one or more bone fracture regions of the one or more bones in the medical image using the fracture detection model.

7. The method of claim 1, further comprising:

reconstructing one or more bone images based on the one or more detected bone fracture regions; and

displaying a marker of the one or more detected bone fracture regions in the one or more bone images or the one or more medical images.

8. The method of claim 7, wherein the one or more bone images include at least one of a stretched curved planar reconstruction (CPR) image, a multiplanar reconstruction (MPR) image, and a three-dimensional (3D) rendering image.

10. The method of claim 8, further comprising:

displaying a management list for managing at least one of one or more bone masks related to the one or more bones and information related to the one or more detected bone fracture regions;

receiving an instruction related to selecting at least one of the one or more bones, the instruction being generated through the management list or the 3D rendering image; and

displaying at least one of the stretched CPR image and one or more MPR images related to the at least one selected bone based on the instruction.

13. The method of claim 1, wherein the one or more medical images include multiple medical images;

the method further comprising;

receiving an instruction of selecting, for display, a first location in a first medical image of the one or more medical images; and

simultaneously displaying the first medical image, or a portion thereof, including the selected first location and a second medical image, or a portion thereof, of the one or more medical image, the second medical image including a second location corresponding to the first location, wherein the displaying of the second medical image, or a portion thereof, includes displaying a marker of the second location.

2-6. -6. (canceled)

9. (canceled)

11-12. -12. (canceled)

14-30. -30. (canceled)

31. A computer-aided diagnosis system for detecting bone fracture, comprising:

at least one storage device including a set of instructions;

at least one processor in communication with the at least one storage device, wherein when executing the set of instructions, the at least one processor is directed to cause the system to perform operations including;

obtaining one or more medical images related to one or more bones;

obtaining a fracture detection model generated based on a machine learning model; and

for at least one of the one or more medical images, detecting one or more bone fracture regions of the one or more bones in the medical image using the fracture detection model.

32. The system of claim 31, wherein the fracture detection model is obtained by performing operations including:

obtaining training images in which bone fractures are marked; and

determining the fracture detection model by training a preliminary model using the training images.

33. The system of claim 31, wherein the detecting the one or more bone fracture regions of the one or more bones in the medical image using the fracture detection model including:

detecting one or more candidate fracture regions in the medical image using the fracture detection model; and

obtaining the one or more bone fracture regions by removing one or more false positive regions from the one or more candidate fracture regions using a bone mask related to the one or more bones.

34. The system of claim 31, wherein when executing the set of instructions, the at least one processor is directed to cause the system to perform additional operations including:

displaying a marker of the one or more bone fracture regions in the at least one of the one or more medical images.

35. The system of claim 31, wherein when executing the set of instructions, the at least one processor is directed to cause the system to perform additional operations including:

determining a type of bone fracture in the one or more bone fracture regions using the fracture detection model.

36. The system of claim 31, wherein the one or more medical images include multiple medical images taken at different slices of the one or more bones, and the at least one processor is directed to cause the system to perform additional operations including:

determining whether there are at least two of the multiple medical images in each of which the one or more bone fracture regions are detected;

determining a distance between the detected bone fracture regions in the at least two of the multiple medical images in response to a determination that there are at least two of the multiple medical images in each of which the one or more bone fracture regions are detected;

determining whether the distance is less than a distance threshold; and

combining the detected bone fracture regions in the at least two of the multiple medical images in response to a determination that the distance is less than the distance threshold, the detected bone fracture regions in the at least two of the multiple medical images being deemed to relate to a same bone fracture.

37. The system of claim 36, wherein when executing the set of instructions, the at least one processor is directed to cause the system to perform additional operations including:

reconstructing one or more bone images based on the one or more detected bone fracture regions or the combined bone fracture region; and

displaying a marker of the one or more detected bone fracture regions or the combined bone fracture region in the one or more bone images.

38. The system of claim 37, wherein the one or more bone images include at least one of a curved planar reconstruction (CPR) image, a multiplanar reconstruction (MPR) image, and a three-dimensional (3D) rendering image.

39. The system of claim 38, wherein to reconstruct the CPR image, the at least one processor is directed to cause the system to perform additional operations including:

extracting a centerline of at least one of the one or more bones based on the one or more medical images; and

generating a stretched CPR image based on the centerline of the bone.

40. The system of claim 39, wherein when executing the set of instructions, the at least one processor is directed to cause the system to perform additional operations including:

displaying a management list for managing at least one of one or more bone masks related to the one or more bones and information related to the one or more detected bone fracture regions.

41. The system of claim 40, wherein when executing the set of instructions, the at least one processor is directed to cause the system to perform additional operations including:

receiving an instruction related to selecting at least one of the one or more bones, the instruction being generated through the management list or the 3D rendering image; and

displaying at least one of the stretched CPR image and one or more MPR images related to the at least one selected bone based on the instruction.

42. The system of claim 31, wherein the fracture detection model is obtained based on a convolutional neural network (CNN).

43. The system of claim 31, wherein the one or more medical images include multiple medical images;

wherein when executing the set of instructions, the at least one processor is directed to cause the system to perform additional operations including;

receiving an instruction of selecting, for display, a first location in a first medical image of the one or more medical images; and

simultaneously displaying the first medical image, or a portion thereof, including the selected first location and a second medical image, or a portion thereof, of the one or more medical image, the second medical image including a second location corresponding to the first location.

44. The system of claim 43, wherein the displaying of the second medical image, or a portion thereof, includes displaying a marker of the second location.

45. The system of claim 31, wherein when executing the set of instructions, the at least one processor is directed to cause the system to perform operations including:

for at least one of the one or more medical images, generating a target image including the one or more bones by segmenting the one or more bones from the medical image.

46. (canceled)