Circle Representation for Medical Object Detection

Nguyen, Ethan H.; Yang, Haichun; Deng, Ruining; Lu, Yan; Zhu, Zheyu; Roland, Joseph T.; Lü, Le; Landman, Bennett A.; Fogo, Agnes B.; Huo, Yuankai

doi:10.1109/tmi.2021.3122835

Cited by 30 publications

(17 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additionally, 120 glomerular patches were derived from five WSIs of the kidney tissue which respectively from different patients and were utilized as detection test set in our automatic detection experimentation. The details of the data selection, acquisition and clinical background were provided in [71].…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…2 and 3). First, Cir-cleNet [21], [71] is employed to perform automatic glomerular detection. After the detection, we extract image patches that contain the detected glomeruli and add padding (50 pixels) around the patches.…”

Section: Methodsmentioning

confidence: 99%

“…The glomerular detection function is implemented based on our previously proposed CircleNet method [21], [71], which introduces a new bounding circle representations for glomerular detection (Fig. 3).…”

Section: B Glomerular Detectionmentioning

confidence: 99%

“…3). Since CircleNet achieved superior performance for glomerular detection compared with current benchmarks [21], [71], we directly migrate CircleNet as the detection method in this study. The detection outcomes are saved in one XML file that contains the circle location, type, and the detection score for each detected object.…”

Section: B Glomerular Detectionmentioning

confidence: 99%

See 3 more Smart Citations

Glo-In-One: Holistic Glomerular Detection, Segmentation, and Lesion Characterization with Large-scale Web Image Mining

Yao¹,

Lu²,

Long³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

The quantitative detection, segmentation, and characterization of glomeruli from high-resolution whole slide imaging (WSI) play essential roles in the computer-assisted diagnosis and scientific research in digital renal pathology. Historically, such comprehensive quantification requires extensive programming skills in order to be able to handle heterogeneous and customized computational tools. To bridge the gap of performing glomerular quantification for non-technical users, we develop the Glo-In-One toolkit to achieve holistic glomerular detection, segmentation, and characterization via a single line of command. Additionally, we release a large-scale collection of 30,000 unlabeled glomerular images to further facilitate the algorithmic development of self-supervised deep learning. The inputs of the Glo-In-One toolkit are WSIs, while the outputs are (1) WSIlevel multi-class circle glomerular detection results (which can be directly manipulated with ImageScope), (2) glomerular image patches with segmentation masks, and (3) different lesion types. In the current version, the fine-grained global glomerulosclerosis (GGS) characterization is provided, including assessedsolidified (S-GGS, associated with hypertension-related injury), disappearing (D-GGS, a further end result of the SGGS becoming contiguous with fibrotic interstitium), and obsolescent (O-GGS, nonspecific GGS increasing with aging) glomeruli. To leverage the performance of the Glo-In-One toolkit, we introduce selfsupervised deep learning to glomerular quantification via largescale web image mining. The GGS fine-grained classification model achieved a decent performance compared with baseline supervised methods while only using 10% of the annotated data. The glomerular detection achieved an average precision of 0.627 with circle representations, while the glomerular segmentation achieved a 0.955 patch-wise Dice Similarity Coefficient (DSC). In this paper, we develop and release an open-source Glo-In-One toolkit, a software with holistic glomerular detection, segmentation, and lesion characterization. This toolkit is userfriendly to non-technical users via a single line of command.The toolbox and the 30,000 web mined glomerular images have been made publicly available at https://github.com/hrlblab/ Glo-In-One.

show abstract

Section: Experiments and Resultsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: B Glomerular Detectionmentioning

confidence: 99%

Section: B Glomerular Detectionmentioning

confidence: 99%

See 2 more Smart Citations

Glo-In-One: Holistic Glomerular Detection, Segmentation, and Lesion Characterization with Large-scale Web Image Mining

Yao¹,

Lu²,

Long³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…The input and output of the ImplantFormer is the 2D slice at tooth crown S i ∈ R H×W ×3 , (H = W = 512) and the corre-sponding implant position P os c (i), respectively. Similar to the existing keypoint regression network [25] [26], ImplantFormer is based on the Gaussian heatmap and the center of the implant position (x i , y i ) at 2D slice is set as the regression target. The network structure of ImplantFormer is given in Fig.…”

Section: B Transformer Based Implant Position Regression Network (Imp...mentioning

confidence: 99%

ImplantFormer: Vision Transformer based Implant Position Regression Using Dental CBCT Data

Yang¹,

Li²,

Li³

et al. 2022

Preprint

View full text Add to dashboard Cite

Implant prosthesis is the most optimum treatment of dentition defect or dentition loss, which usually involves a surgical guide design process to decide the position of implant. However, such design heavily relies on the subjective experiences of dentist. To relieve this problem, in this paper, a transformer based Implant Position Regression Network, ImplantFormer, is proposed to automatically predict the implant position based on the oral CBCT data. The 3D CBCT data is firstly transformed into a series of 2D transverse plane slice views. ImplantFormer is then proposed to predict the position of implant based on the 2D slices of crown images. Convolutional stem and decoder are designed to coarsely extract image feature before the operation of patch embedding and integrate multilevels feature map for robust prediction. The predictions of our network at tooth crown area are finally projected back to the positions at tooth root. As both long-range relationship and local features are involved, our approach can better represent global information and achieves better location performance than the state-of-the-art detectors. Experimental results on a dataset of 128 patients, collected from Shenzhen University General Hospital, show that our ImplantFormer achieves superior performance than benchmarks.

show abstract

LFD-CD: Peripheral Blood Cells Detection Using a Lightweight Cell Detection Model with Full-Connection and Dropconnect

Li,

You,

Liu

et al. 2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Circle Representation for Medical Object Detection

Cited by 30 publications

References 57 publications

Glo-In-One: Holistic Glomerular Detection, Segmentation, and Lesion Characterization with Large-scale Web Image Mining

Glo-In-One: Holistic Glomerular Detection, Segmentation, and Lesion Characterization with Large-scale Web Image Mining

ImplantFormer: Vision Transformer based Implant Position Regression Using Dental CBCT Data

LFD-CD: Peripheral Blood Cells Detection Using a Lightweight Cell Detection Model with Full-Connection and Dropconnect

Contact Info

Product

Resources

About