A review: Music-emotion recognition and analysis based on EEG signals.

Music plays an essential role in human life and can act as an expression to evoke human emotions. The diversity of music makes the listener's experience of music appear diverse. Different music can induce various emotions, and the same theme can also generate other feelings related to the listener's current psychological state. Music emotion recognition (MER) has recently attracted widespread attention in academics and industry. With the development of brain science, MER has been widely used in different fields, e.g., recommendation systems, automatic music composing, psychotherapy, and music visualization. Especially with the rapid development of artificial intelligence, deep learning-based music emotion recognition is gradually becoming mainstream. Besides, electroencephalography (EEG) enables external devices to sense neurophysiological signals in the brain without surgery. This non-invasive brain-computer signal has been used to explore emotions. This paper surveys EEG music emotional analysis, involving the analysis process focused on the music emotion analysis method, e.g., data processing, emotion model, and feature extraction. Then, challenging problems and development trends of EEG-based music emotion recognition is proposed. Finally, the whole paper is summarized.

Prediction of placenta accreta spectrum by combining deep learning and radiomics using T2WI: a multicenter study.

PURPOSE: To achieve prenatal prediction of placenta accreta spectrum (PAS) by combining clinical model, radiomics model, and deep learning model using T2-weighted images (T2WI), and to objectively evaluate the performance of the prediction through multicenter validation. METHODS: A total of 407 pregnant women from two centers undergoing preoperative magnetic resonance imaging (MRI) were retrospectively recruited. The patients from institution I were divided into a training cohort (n = 298) and a validation cohort (n = 75), while patients from institution II served as the external test cohort (n = 34). In this study, we built a clinical prediction model using patient clinical data, a radiomics model based on selected key features, and a deep learning model by mining deep semantic features. Based on this, we developed a combined model by ensembling the prediction results of the three models mentioned above to achieve prenatal prediction of PAS. The performance of these predictive models was evaluated with respect to discrimination, calibration, and clinical usefulness. RESULTS: The combined model achieved AUCs of 0.872 (95% confidence interval, 0.843 to 0.908) in the validation cohort and 0.857 (0.808 to 0.894) in the external test cohort, both of which outperformed the other models. The calibration curves demonstrated excellent consistency in the validation cohort and the external test cohort, and the decision curves indicated high clinical usefulness. CONCLUSION: By using preoperative clinical information and MRI images, the combined model can accurately predict PAS by ensembling clinical model, radiomics model, and deep learning model.

A multi-head self-attention deep learning approach for detection and recommendation of neuromagnetic high frequency oscillations in epilepsy.

Magnetoencephalography is a noninvasive neuromagnetic technology to record epileptic activities for the pre-operative localization of epileptogenic zones, which has received increasing attention in the diagnosis and surgery of epilepsy. As reported by recent studies, pathological high frequency oscillations (HFOs), when utilized as a biomarker to localize the epileptogenic zones, result in a significant reduction in seizure frequency, even seizure elimination in around 80% of cases. Thus, objective, rapid, and automatic detection and recommendation of HFOs are highly desirable for clinicians to alleviate the burden of reviewing a large amount of MEG data from a given patient. Despite the advantage, the performance of existing HFOs rarely satisfies the clinical requirement. Consequently, no HFOs have been successfully applied to real clinical applications so far. In this work, we propose a multi-head self-attention-based detector for recommendation, termed MSADR, to detect and recommend HFO signals. Taking advantage of the state-of-the-art multi-head self-attention mechanism in deep learning, the proposed MSADR achieves a more superior accuracy of 88.6% than peer machine learning models in both detection and recommendation tasks. In addition, the robustness of MSADR is also extensively assessed with various ablation tests, results of which further demonstrate the effectiveness and generalizability of the proposed approach.

Deep learning and radiomics analysis for prediction of placenta invasion based on T2WI.

The purpose of this study was to explore whether the Nomogram, which was constructed by combining the Deep learning and Radiomic features of T2-weighted MR images with Clinical factors (NDRC), could accurately predict placenta invasion. This retrospective study included 72 pregnant women with pathologically confirmed placenta invasion and 40 pregnant women with normal placenta. After 24 gestational weeks, all participants underwent magnetic resonance imaging. The uterus and placenta regions were segmented in magnetic resonance images on sagittal T2WI. Ninety-three radiomics features were extracted from the placenta region, and 128 deep features were extracted from the uterus region using a deep neural network. The least absolute shrinkage and selection operator (LASSO) algorithm was used to filter these 221 features and to form the combined signature. Then the combined signature (CS) and clinical factors were combined to construct a nomogram. The accuracy, sensitivity, specificity and AUC of the nomogram were compared with four machine learning methods. The model NDRC was trained on the dataset of 78 pregnant women in the training cohort. Finally, the model NDRC was compared with four machine learning methods on the independent validation cohort of 34 pregnant women. The results showed that the prediction accuracy, sensitivity, specificity and AUC of the NDRC model were 0.941, 0.952, 0.923 and 0.985 respectively, which outperforms the traditional machine learning methods which rely on radiomics features and deep learning features alone.