-
Encoding Clinical Priori in 3D Convolutional Neural Networks for Prostate Cancer Detection in bpMRI
We hypothesize that anatomical priors can be viable mediums to infuse do...
read it
-
Computer Aided Detection for Pulmonary Embolism Challenge (CAD-PE)
Rationale: Computer aided detection (CAD) algorithms for Pulmonary Embol...
read it
-
Accurate and Robust Pulmonary Nodule Detection by 3D Feature Pyramid Network with Self-supervised Feature Learning
Accurate detection of pulmonary nodules with high sensitivity and specif...
read it
-
Coarse-to-Fine Classification via Parametric and Nonparametric Models for Computer-Aided Diagnosis
Classification is one of the core problems in Computer-Aided Diagnosis (...
read it
-
Fully-Automated Liver Tumor Localization and Characterization from Multi-Phase MR Volumes Using Key-Slice ROI Parsing: A Physician-Inspired Approach
Using radiological scans to identify liver tumors is crucial for proper ...
read it
-
An Unsupervised Method for Detection and Validation of The Optic Disc and The Fovea
In this work, we have presented a novel method for detection of retinal ...
read it
-
A Locating Model for Pulmonary Tuberculosis Diagnosis in Radiographs
Objective: We propose an end-to-end CNN-based locating model for pulmona...
read it
End-to-end Prostate Cancer Detection in bpMRI via 3D CNNs: Effect of Attention Mechanisms, Clinical Priori and Decoupled False Positive Reduction
We present a novel multi-stage 3D computer-aided detection and diagnosis (CAD) model for automated localization of clinically significant prostate cancer (csPCa) in bi-parametric MR imaging (bpMRI). State-of-the-art attention mechanisms drive its detection network, which aims to accurately discriminate csPCa lesions from indolent cancer and the wide range of benign pathology that can afflict the prostate gland. In parallel, a decoupled residual classifier is used to achieve consistent false positive reduction, without sacrificing high detection sensitivity or computational efficiency. Furthermore, a probabilistic anatomical prior, which captures the spatial prevalence of csPCa and its zonal distinction, is computed and encoded into the CNN architecture to guide model generalization with domain-specific clinical knowledge. For 486 institutional testing scans, the 3D CAD system achieves 83.69±5.22% and 93.19±2.96% detection sensitivity at 0.50 and 1.46 false positive(s) per patient, respectively, along with 0.882 AUROC in patient-based diagnosis -significantly outperforming four baseline architectures (USEResNet, UNet++, nnU-Net, Attention U-Net). For 296 external testing scans, the ensembled CAD system shares moderate agreement with a consensus of expert radiologists (76.69%; kappa=0.511) and independent pathologists (81.08%; kappa=0.559); demonstrating a strong ability to localize histologically-confirmed malignancies and generalize beyond the radiologically-estimated annotations of the 1950 training-validation cases used in this study.
READ FULL TEXT
Comments
There are no comments yet.