孙明华

阜外华中心血管病医院 放射科

Corrigendum to "A Collaborative Brain-Computer Interface Framework for Enhancing Group Detection Performance of Dynamic Visual Targets".

[This corrects the article DOI: 10.1155/2022/4752450.].

3区

Computational intelligence and neuroscience 2022

A Collaborative Brain-Computer Interface Framework for Enhancing Group Detection Performance of Dynamic Visual Targets.

The superiority of collaborative brain-computer interface (cBCI) in performance enhancement makes it an effective way to break through the performance bottleneck of the BCI-based dynamic visual target detection. However, the existing cBCIs focus on multi-mind information fusion with a static and unidirectional mode, lacking the information interaction and learning guidance among multiple agents. Here, we propose a novel cBCI framework to enhance the group detection performance of dynamic visual targets. Specifically, a mutual learning domain adaptation network (MLDANet) with information interaction, dynamic learning, and individual transferring abilities is developed as the core of the cBCI framework. MLDANet takes P3-sSDA network as individual network unit, introduces mutual learning strategy, and establishes a dynamic interactive learning mechanism between individual networks and collaborative decision-making at the neural decision level. The results indicate that the proposed MLDANet-cBCI framework can achieve the best group detection performance, and the mutual learning strategy can improve the detection ability of individual networks. In MLDANet-cBCI, the F1 scores of collaborative detection and individual network are 0.12 and 0.19 higher than those in the multi-classifier cBCI, respectively, when three minds collaborate. Thus, the proposed framework breaks through the traditional multi-mind collaborative mode and exhibits a superior group detection performance of dynamic visual targets, which is also of great significance for the practical application of multi-mind collaboration.

3区

Computational intelligence and neuroscience 2022

Effect of different reconstruction algorithms on coronary artery calcium scores using the reduced radiation dose protocol: a clinical and phantom study.

BACKGROUND:This study aimed to evaluate the effects of different iterative reconstruction (IR) algorithms on coronary artery calcium (CAC) score quantification using the reduced radiation dose (RRD) protocol in an anthropomorphic phantom and in patients.METHODS:A thorax phantom, containing 9 calcification inserts with varying hydroxyapatite (HA) densities, was scanned with the reference protocol [120 kv, 80 mAs, filtered back projection (FBP)] and RRD protocol (120 kV, 20-80 mAs, 5 mAs interval) using a 256-slice computed tomography (CT) scanner. Raw data were reconstructed with different reconstruction algorithms [iDose4 levels 1-7 and iterative model reconstruction (IMR) levels 1-3]. Signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and Agatston score (AS) were calculated for each image series. The correction factor was derived from linear regression analysis between the reference image series and other image series with different parameters. Additionally, 40 patients were scanned with the RRD protocol (50 mAs) and reconstructed with FBP, iDose4 level 4, and IMR level 2. AS was calculated for the 3-group image series, and was corrected by applying a correction factor for the IMR group. The agreement of risk stratification with different reconstruction algorithms was also analyzed.RESULTS:For the phantom study, the iDose4 and IMR groups had significantly higher SNR and CNR than the FBP group (all P<0.05). There were no significant differences in the total AS after comparing image series reconstructed with iDose4 (level 1-7) and FBP (all P>0.05), while AS from the IMR (level 1-3) image series were lower than the FBP group (all P<0.05). The tube current of 50 mAs was determined for the clinical study, and the correction factor was 1.14. For the clinical study, the median AS from the iDose4 and IMR groups were both significantly lower compared to the FBP image series [(112.89 (63.01, 314.09), 113.22 (64.78, 364.95) vs. 118.59 (65.05, 374.48), both P<0.05]. After applying the correction factor, the adjusted AS from the IMR group was not significantly different from that of the FBP group [126.48 (69.62, 355.85) vs. 118.59 (65.05, 374.48), P=0.145]. Moreover, the agreement in risk stratification between FBP and IMR improved from 0.81 to 0.85.CONCLUSIONS:The RRD CAC scoring scan using the IMR reconstruction algorithm is clinically feasible, and a correction factor can help reduce the AS underestimation effect.

2.8
2区

Quantitative imaging in medicine and surgery 2021