US 2023/0401706 A1Published application

Method for detecting a rib with a medical image, device, and medium

AssigneeAlibaba (China) Co., Ltd.

InventorsHeng GUO, Jianfeng ZHANG, Minfeng XU, Le LYU

Life of this patent

Estimating the term…

Loading patent…

Abstract

A method for detecting a rib with a medical image, a device and a medium are provided. Feature coding is performed, by using a neural network model, on different rib regions in the medical image to obtain multiple feature vectors. According to a learned corresponding relationship between rib query items and the different rib regions, a feature vector for respective rib query item is determined. Parallel decoding is performed on the feature vector for respective rib query item to obtain a rib detection result corresponding to respective rib query item. The rib query items are endowed with semantic information in anatomy, such that respective rib query items may focus on features of different rib regions, and thus realizing controllable rib detection in instance-level. Any rib detection result includes a rib prediction classification and a prediction value of a pose parameter, in a three-dimensional space, of a rib detection box.

Claims20

Dependent claims

A method for detecting a rib with a medical image, applied to an augmented reality (AR) device, comprising:

acquiring a three-dimensional medical image comprising ribs;
performing, by using a neural network model, feature coding on different rib regions in the medical image, to obtain a plurality of feature vectors;
determining, according to a learned corresponding relationship between rib query items and the different rib regions in the medical image, a feature vector corresponding to a respective rib query item of at least one rib query item, wherein there is a one-to-one correspondence between the at least one rib query item and at least one rib instance to be detected;
performing parallel decoding on the feature vector corresponding to the respective rib query item of the at least one rib query item, to obtain a rib detection result corresponding to the respective rib query item of the at least one rib query item, wherein any rib detection result comprises:
a rib prediction classification, and a prediction value of a pose parameter, in a three-dimensional space, of a rib detection box;
and
superimposing and displaying, in a captured real image, the rib detection result corresponding to the respective rib query item of the at least one rib query item.

4 more elements

A method for detecting a rib with a medical image, comprising:

acquiring a three-dimensional medical image comprising ribs;
performing, by using a neural network model, feature coding on different rib regions in the medical image, to obtain a plurality of feature vectors;
determining, according to a learned corresponding relationship between rib query items and the different rib regions in the medical image, a feature vector corresponding to a respective rib query item of at least one rib query item, wherein there is a one-to-one correspondence between the at least one rib query item and at least one rib instance to be detected;
and
performing parallel decoding on the feature vector corresponding to the respective rib query item of the at least one rib query item, to obtain a rib detection result corresponding to the respective rib query item of the at least one rib query item, wherein any rib detection result comprises:
a rib prediction classification, and a prediction value of a pose parameter, in a three-dimensional space, of a rib detection box.

3 more elements

A method for detecting a rib with a medical image, comprising:

acquiring, in response to a calling request for a first interface from a client, a three-dimensional medical image comprising ribs from an interface parameter of the first interface;
performing, using a neural network model, feature coding on different rib regions in the medical image, to obtain a plurality of feature vectors;
determining, according to a learned corresponding relationship between rib query items and the different rib regions in the medical image, a feature vector corresponding to a respective rib query item of at least one rib query item, wherein there is a one-to-one correspondence between the at least one rib query item and at least one rib instance to be detected;
and
performing parallel decoding on the feature vector corresponding to the respective rib query item of the at least one rib query item, to obtain a rib detection result corresponding to the respective rib query item of the at least one rib query item, wherein any rib detection result comprises:
a rib prediction classification, and a prediction value of a pose parameter, in a three-dimensional space, of a rib detection box.

3 more elements

Dependencies are inferred from the claim text.