Ping an Technology (Shenzhen) Co., Ltd.

Alibaba (China) Co., Ltd.

Dakai Jin

Le Lu

Adam Patrick Harrison

Dazhou Guo

Jinzheng Cai

Ke P Yan

Shun P Miao

Ling Zhang

Zhuotun Zhu

Jiawen Yao

Yingda Xia

Xiansheng Hua

Yirui Wang

Youbao Tang

Ke Yan

A method, an apparatus, and a non-transitory computer readable medium for training an image processing model are provided. The method includes: acquiring a sample image comprising a target object to determine an object segmentation image of the target object in the sample image; constructing an object coordinate map corresponding to the object segmentation image according to the object segmentation image; and training an image processing model comprising a self-attention mechanism layer according to the sample image, the object segmentation image, and the object coordinate map.

Method, apparatus, and non-transitory computer readable medium for training image processing model

An image processing method is provided. The method includes obtaining a to-be-processed image comprising a target object, and inputting the to-be-processed image to a convolutional layer of an image processing model, to obtain an initial feature map of the to-be-processed image, wherein the image processing model comprises an encoder and a decoder; inputting the initial feature map to a self-attention mechanism layer of the encoder, and obtaining a target feature map corresponding to the initial feature map according to position information of each feature in the initial feature map and a position relationship between the each feature and other features; and inputting the target feature map to the decoder for processing, to obtain an object segmentation map and an object label of the target object in the to-be-processed image.

Image processing method, apparatus and non-transitory computer readable medium for performing image processing

A method for performing a computer-aided diagnosis (CAD) includes: acquiring a medical image set; generating a three-dimensional (3D) tumor distance map corresponding to the medical image set, each voxel of the tumor distance map representing a distance from the voxel to a nearest boundary of a primary tumor present in the medical image set; and performing neural-network processing of the medical image set to generate a predicted probability map to predict presence and locations of oncology significant lymph nodes (OSLNs) in the medical image set, wherein voxels in the medical image set are stratified and processed according to the tumor distance map.

Device and method for detecting clinically important objects in medical images with distance-based decision stratification

The present disclosure provides a method, a device, and a computer program product using a self-supervised anatomical embedding (SAM) method. The method includes randomly selecting a plurality of images; for each image of the plurality of images, performing random data augmentation to obtain a patch pair, generating global and local embedding tensors for each patch of the patch pair, and selecting positive pixel pairs from the patch pair and obtaining positive embedding pairs; for each positive pixel pair, computing global and local similarity maps, finding global hard negative embeddings, selecting global random negative embeddings, pooling the global hard negative embeddings and the global random negative embeddings to obtain final global negative embeddings, and finding local hard negative embeddings using the global and local similarity maps, and randomly sampling final local negative embeddings from the local hard negative embeddings; and minimizing a final info noise contrastive estimation (InfoNCE) loss.

Method, device, and computer program product for self-supervised learning of pixel-wise anatomical embeddings in medical images

A method for performing a computer-aided diagnosis (CAD) for universal lesion detection includes: receiving a medical image; processing the medical image to predict lesion proposals and generating cropped feature maps corresponding to the lesion proposals; for each lesion proposal, applying a plurality of lesion detection classifiers to generate a plurality of lesion detection scores, the plurality of lesion detection classifiers including a whole-body classifier and one or more organ-specific classifiers; for each lesion proposal, applying an organ-gating classifier to generate a plurality of weighting coefficients corresponding to the plurality of lesion detection classifiers; and for each lesion proposal, performing weight gating on the plurality of lesion detection scores with the plurality of weighting coefficients to generate a comprehensive lesion detection score.

Device and method for universal lesion detection in medical images

A method and device for stratified image segmentation are provided. The method includes: obtaining a three-dimensional (3D) image data set representative of a region comprising at least three levels of objects; generating a first segmentation result indicating boundaries of anchor-level objects in the region based on a first neural network (NN) model corresponding to the anchor-level objects; generating a second segmentation result indicating boundaries of mid-level objects in the region based on the first segmentation result and a second NN model corresponding to the mid-level objects; and generating a third segmentation result indicating small-level objects in the region based on the first segmentation result, a third NN model corresponding to the small-level objects, and cropped regions corresponding to the small-level objects.

Method and device for stratified image segmentation

The present disclosure provides a clinical target volume delineation method and an electronic device. The method includes: receiving a radiotherapy computed tomography (RTCT) image; and obtaining a plurality of binary images by delineating a gross tumor volume (GTV), lymph nodes (LNs), and organs at risk (OARs) in the RTCT image. A SDMs for each of the binary images is calculated. The RTCT image and all the SDM are finally input into a clinical target volume (CTV) delineation model; and a CTV in the RTCT image is delineated by the CTV delineation model. An automatic delineation of the CTV of esophageal cancer are realized, a delineation efficiency is high and a delineation effect is good.

Clinical target volume delineation method and electronic device

An enhanced medical images processing method and a computing device includes: acquiring series of enhanced medical images and detecting a phase of each enhanced medical image in the series of enhanced medical images using a pre-trained 3D convolutional neural network model. A plurality of target enhanced medical images from the enhanced medical image are selected according to the phases. A plurality of interest images is obtained by identifying and segmenting an interest region in each of the plurality of target enhanced medical images, and finally registering the plurality of interest images. The registered images have clear phase markers and are all spatially aligned, allowing a subsequent doctor or clinician to directly use the registered interest images for diagnosis without the need to rescan the patient.

Enhanced medical images processing method and computing device

A fracture detection method executed by an electronic device is provided. The fracture detection method includes obtaining a to-be-detected image; using a Fully Convolutional Networks (FCN) model to process the to-be-detected image to obtain a fracture probability map of the to-be-detected image; performing a maximum pooling process on the fracture probability map to obtain a first fracture probability; extracting Regions of Interests (ROIs) of the to-be-detected image based on the FCN model; inputting the ROIs into a classification model to obtain a second fracture probability; calculating a product of the first fracture probability and the second fracture probability as a probability of a fracture phenomenon in the to-be-detected image. The present disclosure combines the FCN model and the ROIs to realize an automatic fracture detection, and the accuracy is higher. A device employing the method is also disclosed.

Fracture detection method, electronic device and storage medium

In a GTV segmentation method, a PET-CT image pair and an RTCT image of a human body are obtained. A PET image in the PET-CT image pair is aligned to the RTCT image to obtain an aligned PET image. A first PSNN performs a first GTV segmentation on the RTCT image to obtain a first segmentation image. The RTCT image and the aligned PET image are concatenated into a first concatenated image. A second PSNN performs a second GTV segmentation on the first concatenated image to obtain a second segmentation image. The RTCT image, the first segmentation image, and the second segmentation image are concatenated into a second concatenated image. A third PSNN performs a third GTV segmentation on the second concatenated image to obtain an object segmentation image.

Growing community of inventors

Laurel, MD, United States of America

Dakai Jin