Pectus excavatum (PE) is one of the most common chest wall defects. Accurate assessment of PE deformities is critical for effective surgical intervention. Index-based evaluations have become the ...standard for objectively estimating PE, however, these indexes cannot represent the whole information of chest CT images and may associated with significant error due to the individual differences. To overcome these limitations, this paper developed a computer-aided diagnosis (CAD) system based on the convolutional neural network (CNN) to automatically learn discriminative features and classify PE images. We also adopted block-wise fine-tuning methods based on the transfer learning strategy to reduce the potential risk of overfitting caused by limited data and experimentally explored the best fine-tuning degree. Our method achieved a high level of classification accuracy with 94.76% for PE diagnosis. Furthermore, we proposed a majority rule-based voting method to provide a comprehensively diagnostic results for each patient, which integrated the classification results of the whole thorax. The promising results support the feasibility of our proposed CNN-based CAD system for automatic PE diagnosis, which paves a way for comprehensive assessments of PE in clinics.
Humans show a remarkable perceptual ability to select the speech stream of interest among multiple competing speakers. Previous studies demonstrated that auditory attention detection (AAD) can infer ...which speaker is attended by analyzing a listener's electroencephalography (EEG) activities. However, previous AAD approaches perform poorly on short signal segments, more advanced decoding strategies are needed to realize robust real-time AAD. In this study, we propose a novel approach, i.e., cross-modal attention-based AAD (CMAA), to exploit the discriminative features and the correlation between audio and EEG signals. With this mechanism, we hope to dynamically adapt the interactions and fuse cross-modal information by directly attending to audio and EEG features, thereby detecting the auditory attention activities manifested in brain signals. We also validate the CMAA model through data visualization and comprehensive experiments on a publicly available database. Experiments show that the CMAA achieves accuracy values of 82.8, 86.4, and 87.6% for 1-, 2-, and 5-s decision windows under anechoic conditions, respectively; for a 2-s decision window, it achieves an average of 84.1% under real-world reverberant conditions. The proposed CMAA network not only achieves better performance than the conventional linear model, but also outperforms the state-of-the-art non-linear approaches. These results and data visualization suggest that the CMAA model can dynamically adapt the interactions and fuse cross-modal information by directly attending to audio and EEG features in order to improve the AAD performance.
Compensations are commonly observed in patients with stroke when they engage in reaching without supervision; these behaviors may be detrimental to long-term functional improvement. Automatic ...detection and reduction of compensation cab help patients perform tasks correctly and promote better upper extremity recovery.
Our first objective is to verify the feasibility of detecting compensation online using machine learning methods and pressure distribution data. Second objective was to investigate whether compensations of stroke survivors can be reduced by audiovisual or force feedback. The third objective was to compare the effectiveness of audiovisual and force feedback in reducing compensation.
Eight patients with stroke performed reaching tasks while pressure distribution data were recorded. Both the offline and online recognition accuracy were investigated to assess the feasibility of applying a support vector machine (SVM) based compensation detection system. During reduction of compensation, audiovisual feedback was delivered using virtual reality technology, and force feedback was delivered through a rehabilitation robot.
Good classification performance was obtained in online compensation recognition, with an average F1-score of over 0.95. Based on accurate online detection, real-time feedback significantly decreased compensations of patients with stroke in comparison with no-feedback condition (p < 0.001). Meanwhile, the difference between audiovisual and force feedback was also significant (p < 0.001) and force feedback was more effective in reducing compensation in patients with stroke.
Accurate online recognition validated the feasibility of monitoring compensations using machine learning algorithms and pressure distribution data. Reliable online detection also paved the way for reducing compensations by providing feedback to patients with stroke. Our findings suggested that real-time feedback could be an effective approach to reducing compensatory patterns and force feedback demonstrated a more enviable potential compared with audiovisual feedback.
Robot-assisted rehabilitation is a growing field that can provide an intensity, quality, and quantity of treatment that exceed therapist-mediated rehabilitation. Several control algorithms have been ...implemented in rehabilitation robots to develop a patient-cooperative strategy with the capacity to understand the intention of the user and provide suitable rehabilitation training. In this paper, we present an upper-limb motion pattern recognition method using surface electromyography (sEMG) signals with a support vector machine (SVM) to control a rehabilitation robot, ReRobot, which was built to conduct upper-limb rehabilitation training for post-stroke patients. For poststroke rehabilitation training using the ReRobot, the upper-limb motion of the patient's healthy side is first recognized by detecting and processing the sEMG signals; then, the ReRobot assists the impaired arm in conducting mirror rehabilitation therapy. To train and test the SVM model, five healthy subjects participated in the experiments and performed five standard upper-limb motions, including shoulder flexion, abduction, internal rotation, external rotation, and elbow joint flexion. Good accuracy was demonstrated in experimental results from the five healthy subjects. By recognizing the model motion of the healthy side, the rehabilitation robot can provide mirror therapy to the affected side. This method can be used as a control strategy of upper-limb rehabilitation robots for self-rehabilitation training with stroke patients.
Compensatory movements are commonly employed by stroke survivors during seated reaching and may have negative effects on their long-term recovery. Detecting compensation is useful for coaching the ...patient to reduce compensatory trunk movements and improving the motor function of the paretic arm. Sensor-based and camera-based systems have been developed to detect compensatory movements, but they still have some limitations, such as causing object obstructions, requiring complex setups and raising privacy concerns. To overcome these drawbacks, this paper proposes a compensatory movement detection system based on pressure distribution data and is unobtrusive, simple and practical. Machine learning algorithms were applied to classify compensatory movements automatically. Therefore, the purpose of this study was to develop and test a pressure distribution-based system for the automatic detection of compensation movements of stroke survivors using machine learning algorithms.
Eight stroke survivors performed three types of reaching tasks (back-and-forth, side-to-side, and up-and-down reaching tasks) with both the healthy side and the affected side. The pressure distribution data were recorded, and five features were extracted for classification. The k-nearest neighbor (k-NN) and support vector machine (SVM) algorithms were applied to detect and categorize the compensatory movements. The surface electromyography (sEMG) signals of nine trunk muscles were acquired to provide a detailed description and explanation of compensatory movements.
Cross-validation yielded high classification accuracies (F1-score>0.95) for both the k-NN and SVM classifiers in detecting compensation movements during all the reaching tasks. In detail, an excellent performance was achieved in discriminating between compensation and noncompensation (NC) movements, with an average F1-score of 0.993. For the multiclass classification of compensatory movement patterns, an average F1-score of 0.981 was achieved in recognizing the NC, trunk lean-forward (TLF), trunk rotation (TR) and shoulder elevation (SE) movements.
Good classification performance in detecting and categorizing compensatory movements validated the feasibility of the proposed pressure distribution-based system. Reliable classification accuracy achieved by the machine learning algorithms indicated the potential to monitor compensation movements automatically by using the pressure distribution-based system when stroke survivors perform seated reaching tasks.
Anti-PD-(L)1 immunotherapy has been recommended for non-small cell lung cancer (NSCLC) patients with lymph node metastases (LNM). However, the exact functional feature and spatial architecture of ...tumor-infiltrating CD8 + T cells remain unclear in these patients.
Tissue microarrays (TMAs) from 279 IA-IIIB NSCLC samples were stained by multiplex immunofluorescence (mIF) for 11 markers (CD8, CD103, PD-1, Tim3, GZMB, CD4, Foxp3, CD31, αSMA, Hif-1α, pan-CK). We evaluated the density of CD8 + T-cell functional subsets, the mean nearest neighbor distance (mNND) between CD8 + T cells and neighboring cells, and the cancer-cell proximity score (CCPS) in invasive margin (IM) as well as tumor center (TC) to investigate their relationships with LNM and prognosis.
The densities of CD8 + T-cell functional subsets, including predysfunctional CD8 + T cells (T
) and dysfunctional CD8 + T cells (T
), in IM predominated over those in TC (P < 0.001). Multivariate analysis identified that the densities of CD8 + T
cells in TC and CD8 + T
cells in IM were significantly associated with LNM OR = 0.51, 95%CI (0.29-0.88), P = 0.015; OR = 5.80, 95%CI (3.19-10.54), P < 0.001; respectively and recurrence-free survival (RFS) HR = 0.55, 95%CI (0.34-0.89), P = 0.014; HR = 2.49, 95%CI (1.60-4.13), P = 0.012; respectively, independent of clinicopathological factors. Additionally, shorter mNND between CD8 + T cells and their neighboring immunoregulatory cells indicated a stronger interplay network in the microenvironment of NSCLC patients with LNM and was associated with worse prognosis. Furthermore, analysis of CCPS suggested that cancer microvessels (CMVs) and cancer-associated fibroblasts (CAFs) selectively hindered CD8 + T cells from contacting with cancer cells, and were associated with the dysfunction of CD8 + T cells.
Tumor-infiltrating CD8 + T cells were in a more dysfunctional status and in a more immunosuppressive microenvironment in patients with LNM compared with those without LNM.
This study aims to investigate location-specific functional remodeling following ischemic stroke in pons and corona radiata.
This study was approved by the local Institutional Review Board. Written ...consent was obtained from each of the participants prior to the MRI examination. Thirty six subjects with first ever acute ischemic stroke in pons (PS,
= 15, aged 62.8 ± 11.01 years) or corona radiata (CRS,
= 21, aged 59.33 ± 13.84 years) as well as 30 age and sex matched healthy controls (HC,
= 30, aged 60 ± 6.43 years) were examined with resting state functional magnetic resonance imaging (rs-fMRI). Regional homogeneity (ReHo) and degree centrality (DC) were calculated using a voxel-based approach. Intergroup differences in ReHo and DC were explored using a permutation test with a threshold-free cluster enhancement (PT TFCE, number of permutations = 1,000, family-wise error rate (FWER) < 0.05).
ReHo and DC alterations were identified in distributed anatomies for both PS and CRS groups. DC mainly increased in the bilateral anterior and posterior cingulate cortex, the inferior frontal-orbital gyrus, and decreased in the bilateral cuneus, calcarine, and the precuneus, while ReHo mainly decreased in the precentral and the postcentral gyri, inferior parietal lobules, precuneus, posterior cingulate cortex, and the superior occipital gyrus. PS and CRS groups were not significantly different in ReHo or DC (FWER > 0.05).
Focal ischemic stroke in pons or corona radiata leads to extensive alterations in the functional network centrality. IS-induced network remodeling is more anatomy-specific than pathway-specific, which may underpin the clinicotopographical profiles during the disease dynamic. Approaches targeting neural pathway and functional connectivity may shed light on a better characterization and management innovation of ischemic stroke.
To effectively defend against microbial pathogens, the host cells mount antiviral innate immune responses by producing interferons (IFNs), and hundreds of IFN-stimulated genes (ISGs). Upon ...recognition of cytoplasmic viral or bacterial DNAs and abnormal endogenous DNAs, the DNA sensor cGAS synthesizes 2',3'-cGAMP that induces STING (stimulator of interferon genes) undergoing conformational changes, cellular trafficking, and the activation of downstream factors. Therefore, STING plays a pivotal role in preventing microbial pathogen infection by sensing DNAs during pathogen invasion. This review is dedicated to the recent advances in the dynamic regulations of STING activation, intracellular trafficking, and post-translational modifications (PTMs) by the host and microbial proteins.
Stroke patients often use trunk to compensate for impaired upper limb motor function during upper limb rehabilitation training, which results in a reduced rehabilitation training effect. Detecting ...trunk compensations can improve the effect of rehabilitation training. This study investigates the feasibility of a surface electromyography-based trunk compensation detection (sEMG-bTCD) method. Five healthy subjects and nine stroke subjects with cognitive and comprehension skills were recruited to participate in the experiments. The sEMG signals from nine superficial trunk muscles were collected during three rehabilitation training tasks (reach-forward-back, reach-side-to-side, and reach-up-to-down motions) without compensation and with three common trunk compensations lean-forward (LF), trunk rotation (TR), and shoulder elevation (SE). Preprocessing like filtering, active segment detection was performed and five time domain features (root mean square, variance, mean absolute value (MAV), waveform length, and the fourth order autoregressive model coefficient) were extracted from the collected sEMG signals. Excellent TCD performance was achieved in healthy participants by using support vector machine (SVM) classifier (LF: accuracy = 94.0%, AUC = 0.97, F1 = 0.94; TR: accuracy = 95.8%, AUC = 0.99, F1 = 0.96; SE: accuracy = 100.0%, AUC = 1.00, F1 = 1.00). By using SVM classifier, TCD performance in stroke participants was also obtained (LF: accuracy = 74.8%, AUC = 0.90, F1 = 0.73; TR: accuracy = 67.1%, AUC = 0.85, F1 = 0.71; SE: accuracy = 91.3%, AUC = 0.98, F1 = 0.90). Compared with the methods based on cameras or inertial sensors, better detection performance was obtained in both healthy and stroke participants. The results demonstrated the feasibility of the sEMG-bTCD method, and it helps to prompt the stroke patients to correct their incorrect posture, thereby improving the effectiveness of rehabilitation training.
Tissue-resident memory T (T
) cells can reside in the tumor microenvironment and are considered the primary response cells to immunotherapy. Heterogeneity in functional status and spatial ...distribution may contribute to the controversial role of T
cells but we know little about it.
Through multiplex immunofluorescence (mIF) (CD8, CD103, PD-1, Tim-3, GZMB, CK), the quantity and spatial location of T
cell subsets were recognized in the tissue from 274 patients with NSCLC after radical surgery. By integrating multiple machine learning methods, we constructed a T
-based spatial immune signature (T
-SIS) to predict the prognosis. Furthermore, we conducted a CD103-related gene set enrichment analysis (GSEA) and verified its finding by another mIF panel (CD8, CD103, CK, CD31, Hif-1α).
The density of T
cells was significantly correlated with the expression of PD-1, Tim-3 and GZMB. Four types of T
cell subsets was defined, including T
(PD-1
Tim-3
T
), T
(PD-1
Tim-3
T
), T
(PD-1
Tim-3
T
) and T
(PD-1
Tim-3
T
). The cytotoxicity of T
was the strongest while that of T
was the weakest. Compare with T
and T
, T
and T
had better infiltration and stronger interaction with cancer cells. The T
-SIS was an independent prognostic factor for disease-free survival HR = 2.43, 95%CI (1.63-3.60), P < 0.001 and showed a better performance than the TNM staging system for recurrence prediction. Furthermore, by CD103-related GSEA and mIF validation, we found a negative association between tumor angiogenesis and infiltration of T
cells.
These findings reveal a significant heterogeneity in the functional status and spatial distribution of T
cells, and support it as a biomarker for the prognosis of NSCLC patients. Regulating T
cells by targeting tumor angiogenesis may be a potential strategy to improve current immunotherapy.