Automated unruptured cerebral aneurysms detection in TOF MR angiography images using dual-channel SE-3D UNet: a multi-center research.

OBJECTIVES: Time of flight magnetic resonance angiography (TOF-MRA) is the primary non-invasive screening method for cerebral aneurysms. We aimed to develop a computer-aided aneurysm detection method to improve the diagnostic efficiency and accuracy, especially decrease the false positive rate. METHODS: This is a retrospective multicenter study. The dataset contained 1160 TOF-MRA examinations composed of unruptured aneurysms (n = 1096) and normal controls (n = 166) from six hospitals. A total of 1037 examinations acquired from 2013 to 2019 were used as training set; 123 examinations acquired from 2020 to 2021 were used as external test set. We proposed an equalized augmentation strategy based on aneurysm location and constructed a detection model based on dual channel SE-3D UNet. The model was trained with a 5-fold cross-validation in the training set, then tested on the external test set. RESULTS: The proposed method achieved 82.46% sensitivity on patient-level, 73.85% sensitivity on lesion-level, and 0.88 false positives per case in the external test set. The performance did not show significant differences in subgroups according to the aneurysm site (except ACA), aneurysm size (except smaller than 3 mm), or MRI scanners. The performance preceded the basic SE-3D UNet by increasing 15.79% patient-level sensitivity and decreasing 4.19 FPs/case. CONCLUSIONS: The proposed automated aneurysm detection method achieved acceptable sensitivity while controlling fairly low false positives per case. It might provide a useful auxiliary tool of cerebral aneurysms MRA screening. KEY POINTS: • The need for automated cerebral aneurysms detecting is growing. • The strategy of equalized augmentation based on aneurysm location and dual-channel input could improve the model performance. • The retrospective multi-center study showed that the proposed automated cerebral aneurysms detection using dual-channel SE-3D UNet could achieve acceptable sensitivity while controlling a low false positive rate.

Multi-task motor imagery electroencephalogram classification based on adaptive time-frequency common spatial pattern combined with convolutional neural network / 生物医学工程学杂志

The effective classification of multi-task motor imagery electroencephalogram (EEG) is helpful to achieve accurate multi-dimensional human-computer interaction, and the high frequency domain specificity between subjects can improve the classification accuracy and robustness. Therefore, this paper proposed a multi-task EEG signal classification method based on adaptive time-frequency common spatial pattern (CSP) combined with convolutional neural network (CNN). The characteristics of subjects' personalized rhythm were extracted by adaptive spectrum awareness, and the spatial characteristics were calculated by using the one-versus-rest CSP, and then the composite time-domain characteristics were characterized to construct the spatial-temporal frequency multi-level fusion features. Finally, the CNN was used to perform high-precision and high-robust four-task classification. The algorithm in this paper was verified by the self-test dataset containing 10 subjects (33 ± 3 years old, inexperienced) and the dataset of the 4th 2018 Brain-Computer Interface Competition (BCI competition Ⅳ-2a). The average accuracy of the proposed algorithm for the four-task classification reached 93.96% and 84.04%, respectively. Compared with other advanced algorithms, the average classification accuracy of the proposed algorithm was significantly improved, and the accuracy range error between subjects was significantly reduced in the public dataset. The results show that the proposed algorithm has good performance in multi-task classification, and can effectively improve the classification accuracy and robustness.

Automatic segmentation of kidney tumor based on cascaded multiscale convolutional neural networks / 生物医学工程学杂志

The background of abdominal computed tomography (CT) images is complex, and kidney tumors have different shapes, sizes and unclear edges. Consequently, the segmentation methods applying to the whole CT images are often unable to effectively segment the kidney tumors. To solve these problems, this paper proposes a multi-scale network based on cascaded 3D U-Net and DeepLabV3+ for kidney tumor segmentation, which uses atrous convolution feature pyramid to adaptively control receptive field. Through the fusion of high-level and low-level features, the segmented edges of large tumors and the segmentation accuracies of small tumors are effectively improved. A total of 210 CT data published by Kits2019 were used for five-fold cross validation, and 30 CT volume data collected from Suzhou Science and Technology Town Hospital were independently tested by trained segmentation models. The results of five-fold cross validation experiments showed that the Dice coefficient, sensitivity and precision were 0.796 2 ± 0.274 1, 0.824 5 ± 0.276 3, and 0.805 1 ± 0.284 0, respectively. On the external test set, the Dice coefficient, sensitivity and precision were 0.817 2 ± 0.110 0, 0.829 6 ± 0.150 7, and 0.831 8 ± 0.116 8, respectively. The results show a great improvement in the segmentation accuracy compared with other semantic segmentation methods.

Intelligence-aided diagnosis of Parkinson's disease with rapid eye movement sleep behavior disorder based on few-channel electroencephalogram and time-frequency deep network / 生物医学工程学杂志

Aiming at the limitations of clinical diagnosis of Parkinson's disease (PD) with rapid eye movement sleep behavior disorder (RBD), in order to improve the accuracy of diagnosis, an intelligent-aided diagnosis method based on few-channel electroencephalogram (EEG) and time-frequency deep network is proposed for PD with RBD. Firstly, in order to improve the speed of the operation and robustness of the algorithm, the 6-channel scalp EEG of each subject were segmented with the same time-window. Secondly, the model of time-frequency deep network was constructed and trained with time-window EEG data to obtain the segmentation-based classification result. Finally, the output of time-frequency deep network was postprocessed to obtain the subject-based diagnosis result. Polysomnography (PSG) of 60 patients, including 30 idiopathic PD and 30 PD with RBD, were collected by Nanjing Brain Hospital Affiliated to Nanjing Medical University and the doctor's detection results of PSG were taken as the gold standard in our study. The accuracy of the segmentation-based classification was 0.902 4 in the validation set. The accuracy of the subject-based classification was 0.933 3 in the test set. Compared with the RBD screening questionnaire (RBDSQ), the novel approach has clinical application value.

Automated computer-assisted detection system for cerebral aneurysms in time-of-flight magnetic resonance angiography using fully convolutional network.

BACKGROUND: As the rupture of cerebral aneurysm may lead to fatal results, early detection of unruptured aneurysms may save lives. At present, the contrast-unenhanced time-of-flight magnetic resonance angiography is one of the most commonly used methods for screening aneurysms. The computer-assisted detection system for cerebral aneurysms can help clinicians improve the accuracy of aneurysm diagnosis. As fully convolutional network could classify the image pixel-wise, its three-dimensional implementation is highly suitable for the classification of the vascular structure. However, because the volume of blood vessels in the image is relatively small, 3D convolutional neural network does not work well for blood vessels. RESULTS: The presented study developed a computer-assisted detection system for cerebral aneurysms in the contrast-unenhanced time-of-flight magnetic resonance angiography image. The system first extracts the volume of interest with a fully automatic vessel segmentation algorithm, then uses 3D-UNet-based fully convolutional network to detect the aneurysm areas. A total of 131 magnetic resonance angiography image data are used in this study, among which 76 are training sets, 20 are internal test sets and 35 are external test sets. The presented system obtained 94.4% sensitivity in the fivefold cross-validation of the internal test sets and obtained 82.9% sensitivity with 0.86 false positive/case in the detection of the external test sets. CONCLUSIONS: The proposed computer-assisted detection system can automatically detect the suspected aneurysm areas in contrast-unenhanced time-of-flight magnetic resonance angiography images. It can be used for aneurysm screening in the daily physical examination.

Prediction of seizures in sleep based on power spectrum / 生物医学工程学杂志

Seizures during sleep increase the probability of complication and sudden death. Effective prediction of seizures in sleep allows doctors and patients to take timely treatments to reduce the aforementioned probability. Most of the existing methods make use of electroencephalogram (EEG) to predict seizures, which are not specific developed for the sleep. However, EEG during sleep has its characteristics compared with EEG during other states. Therefore, in order to improve the sensitivity and reduce the false alarm rate, this paper utilized the characteristics of EEG to predict seizures during sleep. We firstly constructed the feature vector including the absolute power spectrum, the relative power spectrum and the power spectrum ratio in different frequencies. Secondly, the separation criterion and branch-and-bound method were applied to select features. Finally, support vector machine classifier were trained, which is then employed for online prediction. Compared with the existing method that do not consider the characteristics of sleeping EEG (sensitivity 91.67%, false alarm rate 9.19%), the proposed method was superior in terms of sensitivity (100%) and false alarm rate (2.11%). This method can improve the existing epilepsy prediction methods and has important clinical value.

Computer-aided diagnosis of Parkinson's disease based on the stacked deep polynomial networks ensemble learning framework / 生物医学工程学杂志

Feature representation is the crucial factor for the magnetic resonance imaging (MRI) based computer-aided diagnosis (CAD) of Parkinson's disease (PD). Deep polynomial network (DPN) is a novel supervised deep learning algorithm, which has excellent feature representation for small dataset. In this work, a stacked DPN (SDPN) based ensemble learning framework is proposed for diagnosis of PD, which can improve diagnostic accuracy for small dataset. In the proposed framework, SDPN was performed on each subset of extracted features from MRI images to generate new feature representation. The support vector machine (SVM) was then adopted to perform classification task on each subset. The ensemble learning algorithm was then performed on all the SVM classifiers to generate the final diagnosis for PD. The experimental results on the Parkinson's Progression Markers Initiative dataset (PPMI) showed that the proposed algorithm achieved the classification accuracy, sensitivity and specificity of 90.15%, 85.48% and 93.27%, respectively, with the brain network features, and it also got the classification accuracy of 87.18%, sensitivity of 86.90% and specificity of 87.27% on the multi-view features extracted from different brain regions. Moreover, the proposed algorithm outperformed other algorithms on the MRI dataset from PPMI. It suggests that the proposed SDPN-based ensemble learning framework has the feasibility and effectiveness for the CAD of PD.