A single-stage detector of cerebral microbleeds using 3D feature fused region proposal network (FFRP-Net)

Kim, Junho; Al‐masni, Mohammed A.; Lee, Haejoon; Choi, Yun-Chul; Kim, Dong-Hyun

doi:10.1109/aicas54282.2022.9869855

Cited by 2 publications

(5 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This article presents a direct comparison with other 3D detection models that were implemented on the same dataset. Even though other works in the literature presented methods of automatic CMBs detection that employed 3D networks, [7][8][9]12,16 most of the frameworks were not implemented as single networks nor end-to-end structures, which may cause difficulties when seeking to incorporate them into routine clinical use. Therefore, for comparison, this study built the 3D Faster R-CNN as an end-to-end structure for CMBs detection and compared the testing performance.…”

Section: Discussionmentioning

confidence: 99%

“…Therefore, for comparison, this study built the 3D Faster R‐CNN as an end‐to‐end structure for CMBs detection and compared the testing performance. Furthermore, this study directly compared TPE‐Det against the most recently presented deep learning detection models with two‐stage 10 and single‐stage structures 16 on the same dataset. As a result, the proposed TPE‐Det achieved significantly higher precision ( P < 0.02) with higher sensitivity and a lower FP avg .…”

Section: Discussionmentioning

confidence: 99%

“…Furthermore, this study implemented the most recently proposed fully deep learning CMB detection methods for each two‐stage 10 and single‐stage structure 16 utilizing 3D information. All three models were trained and evaluated on the same dataset (DS1) with the proposed TPE‐Det.…”

Section: Methodsmentioning

confidence: 99%

“…10 To overcome the above limitations, several studies have proposed a framework consisting of only one stage. [14][15][16] Wang et al 14 and Hong et al 15 applied a 2D sliding window technique to the entire MR data and performed classification on each sliding window for all subjects. Both works achieved high sensitivity.…”

mentioning

confidence: 99%

“…Consequently, it is burdensome to employ 3D networks that are much heavier than 2D networks in such one‐stage frameworks. In addition, most recently, Kim et al 16 proposed a single‐stage CMB detector employing a 3D feature fusion module. Due to the computational cost, they divided each 3D subject datum into multiple patches and performed detection, providing a precision of 25.64%.…”

mentioning

confidence: 99%

See 4 more Smart Citations

Detection of Cerebral Microbleeds in MR Images Using a Single‐Stage Triplanar Ensemble Detection Network (TPE‐Det)

Lee

Kim

Lee

et al. 2022

Magnetic Resonance Imaging

View full text Add to dashboard Cite

Background Cerebral microbleeds (CMBs) are microscopic brain hemorrhages with implications for various diseases. Automated detection of CMBs is a challenging task due to their wide distribution throughout the brain, small size, and visual similarity to their mimics. For this reason, most of the previously proposed methods have been accomplished through two distinct stages, which may lead to difficulties in integrating them into clinical workflows. Purpose To develop a clinically feasible end‐to‐end CMBs detection network with a single‐stage structure utilizing 3D information. This study proposes triplanar ensemble detection network (TPE‐Det), ensembling 2D convolutional neural networks (CNNs) based detection networks on axial, sagittal, and coronal planes. Study Type Retrospective. Subjects Two datasets (DS1 and DS2) were used: 1) 116 patients with 367 CMBs and 12 patients without CMBs for training, validation, and testing (70.39 ± 9.30 years, 68 women, 60 men, DS1); 2) 58 subjects with 148 microbleeds and 21 subjects without CMBs only for testing (76.13 ± 7.89 years, 47 women, 32 men, DS2). Field Strength/Sequence A 3 T field strength and 3D GRE sequence scan for SWI reconstructions. Assessment The sensitivity, FPavg (false‐positive per subject), and precision measures were computed and analyzed with statistical analysis. Statistical Tests A paired t‐test was performed to investigate the improvement of detection performance by the suggested ensembling technique in this study. A P value < 0.05 was considered significant. Results The proposed TPE‐Det detected CMBs on the DS1 testing set with a sensitivity of 96.05% and an FPavg of 0.88, presenting statistically significant improvement. Even when the testing on DS2 was performed without retraining, the proposed model provided a sensitivity of 85.03% and an FPavg of 0.55. The precision was significantly higher than the other models. Data Conclusion The ensembling of multidimensional networks significantly improves precision, suggesting that this new approach could increase the benefits of detecting lesions in the clinic. Evidence Level 1 Technical Efficacy Stage 2

show abstract