Explore the vast range of titles available.

This paper aims to address the problem of supervised monocular depth estimation. We start with a meticulous pilot study to demonstrate that the long-range correlation is essential for accurate depth estimation. Moreover, the Transformer and convolution are good at long-range and close-range depth estimation, respectively. Therefore, we propose to adopt a parallel encoder architecture consisting of a Transformer branch and a convolution branch. The former can model global context with the effective attention mechanism and the latter aims to preserve the local information as the Transformer lacks the spatial inductive bias in modeling such contents. However, independent branches lead to a shortage of connections between features. To bridge this gap, we design a hierarchical aggregation and heterogeneous interaction module to enhance the Transformer features and model the affinity between the heterogeneous features in a set-to-set translation manner. Due to the unbearable memory cost introduced by the global attention on high-resolution feature maps, we adopt the deformable scheme to reduce the complexity. Extensive experiments on the KITTI, NYU, and SUN RGB-D datasets demonstrate that our proposed model, termed DepthFormer, surpasses state-of-the-art monocular depth estimation methods with prominent margins. The effectiveness of each proposed module is elaborately evaluated through meticulous and intensive ablation studies.

To address the issues of low perception accuracy and poor robustness in traditional motion recognition methods within complex walking environments for visually impaired individuals, this study utilizes multi-modal data, including ECG, PPG, and IMU, for classification. Regarding the low filtering efficiency of multi-modal data, an improved wavelet filtering algorithm based on LSTM is proposed. To further enhance classification accuracy, this paper introduces a motion recognition method for the blindfolded mobility simulation based on an Attention-based Two-Stream Deep Fusion Convolutional Neural Network (ATS-DFCNN). The proposed method constructs a two-stream heterogeneous feature extraction architecture by synchronously collecting tri-axial motion signals and physiological signals from subjects. A 1D-CNN is employed to capture the spatial geometric features of limb movements, while a hybrid CNN-GRU network is utilized to mine the temporal evolution patterns of physiological stress. Furthermore, an attention mechanism is introduced to achieve dynamic weighted fusion at the feature level, which strengthens critical motion features and suppresses environmental noise. Experiments were conducted with 10 subjects simulating the movements of visually impaired individuals, covering typical actions such as walking, standing, climbing stairs, descending stairs, and falling. The results demonstrate that the proposed adaptive filtering algorithm achieves an AUC of 0.942, significantly improving feature distinctiveness compared to traditional algorithms. The ATS-DFCNN model achieved an average recognition accuracy of 92.2% across five activity categories, representing a 4.8% performance increase over single IMU modal classification. Particularly in fall detection, the model effectively reduces false alarms through physiological feedback and accurately infers motion intentions, providing reliable technical support for the safety monitoring of intelligent walking-aid systems.

The operational health state of motor bearings is critical to the operational safety of harbor portal slewing cranes. However, in harsh industrial environments with strong noise and time-varying rotational speeds, existing bearing fault diagnosis methods still suffer from the problems of incomplete fault feature extraction from single-sensor signals and the excessively large size of multi-source fusion models, which makes them unable to adapt to edge deployment. To address these issues, this paper proposes a Multi-source Feature Fusion Lightweight Network (MTFL-Net) integrated with targeted structured channel pruning. First, vibration and current signals are preprocessed via differentiated time-frequency transformation and converted into 2D time-frequency images, to fully preserve transient impact and spectral fault features. Second, a multi-branch feature extraction architecture embedded with residual connections, multi-scale convolution and channel attention gating is designed, to alleviate feature degradation and adaptively enhance fault-sensitive features. Third, targeted structured channel pruning is performed on the feature extraction branches, to remove redundant channels while retaining the multi-source fusion logic and core feature extraction structure. Experiments on two public bearing datasets show that the original model achieves 99% diagnostic accuracy, and the pruned model still maintains an accuracy of 95%. The results demonstrate that MTFL-Net can significantly reduce model size and computational cost while retaining high diagnostic precision.

Primary gastrointestinal non-Hodgkin lymphoma (PGIL) is a rare hematopoietic malignancy with limited data to guide management. We analyzed the clinical characteristics and survival of 219 newly diagnosed PGIL patients. Our single-center data showed that the incidence rate of primary gastric lymphoma (PGL) was higher than that of primary intestine lymphoma (PIL). Most PGIL was B-cell originated and DLBCL was the most common pathological type both in PGL and PIL group. Univariate and multivariate analysis showed that IPI score and pathology were independent prognostic factors. The overall survival (OS) and progression-free survival (PFS) of patients with MYC rearrangement were much shorter compared to patients without MYC rearrangement indicating that MYC translocation was related to decreased survival. Neither OS nor PFS differed between patients who received chemotherapy with or without surgery. However, patients who received surgery alone had a poor prognosis. Chemotherapy is the front-line treatment for PGIL while surgery was conducted to relieve tumor-related complications or make diagnosis. MYC rearrangement predicted poor prognosis of PGIL patients.

In aviation safety, pilots must efficiently process dynamic visual information and maintain a high level of attention. Any missed judgment of critical information or delay in decision-making may lead to mission failure or catastrophic consequences. Therefore, accurately detecting pilots’ attention states is the primary prerequisite for improving flight safety and performance. To better detect the attention state of pilots, this paper takes flight trainees as the research object and the simulated flight environment as the experimental background. It proposes a method for detecting the attention state of flight trainees based on a multi-modal dynamic depth network (M3D-Net). The M3D-Net architecture is a lightweight neural network architecture that integrates temporal image features, visual information features, and flight operation data features. It aligns image and text features through an attention mechanism to enhance the semantic association between modalities; it utilizes the Depth-wise Separable Convolution and LSTM (DSC-LSTM) module to model temporal information, dynamically capturing the contextual dependencies within the sequence, and achieving six-level attention state classification. This paper conducted ablation experiments to comparatively analyze the classification effects of the model and also evaluates the effectiveness of our proposed method through model evaluation metrics. Experiments show that the classification effect of the model architecture proposed in this paper reaches 97.56%, with a model size of 18.6 M. Compared with traditional algorithms, the M3D-Net architecture has better performance prospects in terms of application.

Primary bone marrow diffuse large B cell lymphoma (DLBCL) is an independent pathologic type with a poor prognosis when treated with standard chemoimmunotherapy. Generally, rituximab-based high-dose chemotherapy regimens such as dose-adjusted etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin (DA-EPOCH) can be administered to young patients, followed by autologous stem cell transplantation. For elderly patients, the rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisolone (R-CHOP) regimen is well tolerated, but it is an insufficient induction therapy for this group. Herein, we reported an elderly patient diagnosed with primary bone marrow DLBCL, germinal center B-cell-like subtype. Considering tolerance, the R-CHOP regimen was administered. However, his disease progressed after two treatment cycles. Then, the rituximab, gemcitabine, dexamethasone, cisplatin, lenalidomide regimen was administered, but the patient still experienced disease progression. Subsequently, the histone deacetylase (HDAC) inhibitor chidamide and Bruton’s tyrosine kinase (BTK) inhibitor ibrutinib were concurrently administered, and the patient achieved complete remission. We found that the response of primary bone marrow DLBCL to chemotherapy was poorer than that of de novo DLBCL. High-dose chemotherapy regimens such as DA-EPOCH should be administered to young patients in combination with rituximab. For elderly patients, new targeted drugs such as HDAC and BTK inhibitors appear to produce favorable outcomes.

High-dose chemotherapy followed by autologous hematopoietic stem cell transplantation (ASCT) is still a consolidation treatment choice for relapsed/refractory B-cell non-Hodgkin’s lymphoma (NHL) patients and some aggressive B-cell NHL as frontline therapy. Due to the shortage of carmustine, we switched to idarubicin-substituted BEAC (IEAC) conditioning regimen. We retrospectively compared the outcomes of 72 aggressive B-cell NHL patients treated with IEAC or BEAC regimens followed by ASCT as upfront consolidative treatment. The median time to neutrophil and platelet reconstitution showed no difference between IEAC and BEAC groups. IEAC regimen was well tolerated without increase of adverse events. Transplant-related mortality didn’t occur. The overall survival (OS) and progression-free survival (PFS) of IEAC group (33 and 23 months) were a little longer than that of BEAC group (30 and 18 months). However, due to the small sample numbers, there’s no significant difference in OS and PFS between IEAC and BEAC group with DLBCL or MCL. Multivariate analysis showed that AnnArbor staging, IPI score, lactate dehydrogenase level, remission of disease, modified regimen were related with PFS and OS. In conclusion, IEAC regimen was well tolerated and replacement with idarubicin could be an alternative when carmustine was not available.

In this paper, a multi-stage deep learning blood pressure prediction model based on imaging photoplethysmography (IPPG) signals is proposed to achieve accurate and convenient monitoring of human blood pressure. A camera-based non-contact human IPPG signal acquisition system is designed. The system can perform experimental acquisition under ambient light, effectively reducing the cost of non-contact pulse wave signal acquisition while simplifying the operation process. The first open-source dataset IPPG-BP for IPPG signal and blood pressure data is constructed by this system, and a multi-stage blood pressure estimation model combining a convolutional neural network and bidirectional gated recurrent neural network is designed. The results of the model conform to both BHS and AAMI international standards. Compared with other blood pressure estimation methods, the multi-stage model automatically extracts features through a deep learning network and combines different morphological features of diastolic and systolic waveforms, which reduces the workload while improving accuracy.

Global digitalization leads to a new digital inequality, jointly faced by urban and rural areas. Diagnosing these digital challenges and seeking common coping strategies are crucial in the digital era. Yet, the state of Urban-Rural Digitalization (URD) remains ambiguous, and effective solutions are still lacking. Here we propose an integrative approach and introduce a novel Development-Gap-Integration framework to assess URD. Utilizing authoritative data from official government sources and institutes, we conduct a comprehensive analysis of 30 provinces in China from 2000 to 2020. Our findings depict a transformative journey of URD, evolving from divide to inclusion. The increasing integration accompanies by advancing development and diminishing gaps. However, we identify three significant challenges: 1) high development coinciding with high gaps, 2) inadequate integration of digital applications, and 3) widening disparity between provinces. Thus, we suggest tailored policy recommendations focused on urban-rural integration, enhancing digital literacy, and promoting regional coordinated development.

Palmoplantar pustulosis (PPP) is a chronic inflammatory dermatosis with limited therapeutic options. Traditional Chinese Medicine (TCM) formulations may benefit PPP, yet microbiome-immune mechanisms underlying clinical response remain unclear. To evaluate whether an 8-week standardized Qingpao Chushi Jiedu Fang (QCF) regimen is associated with coordinated changes in clinical severity, oral microbiota, and circulating cytokines in PPP. Thirty PPP patients received an 8-week standardized Qingpao Chushi Jiedu Fang (QCF) regimen. Clinical severity (PPPASI, Palmoplantar Pustulosis Area and Severity Index; Dermatology Life Quality Index, DLQI; pruritus/pain Visual Analogue Scale, VAS), oral microbiota (16S rDNA sequencing), and serum cytokines (IL-1β, IL-4, IFN- , IFN- ) were assessed before and after treatment. Subgroup analyses were performed by smoking status. PPPASI significantly decreased from week 2 onward (  < 0.001), with further improvement at weeks 4 and 8, while week-6 vs. week-4 changes were non significant (  > 0.05). DLQI declined significantly at weeks 4-8 (  < 0.05). Pain scores showed improvement only at week 6 vs. week 2 (  < 0.05), and itch scores improved at week 8 (  < 0.05). Oral microbial - and -diversity shifted significantly after treatment (  < 0.05), with clear changes in community structure and taxa; smokers exhibited more pronounced restructuring. Cytokine levels changed concordantly, with IL-1β, IL-4, and IFN- decreasing and IFN- increasing after treatment (all  < 0.05). QCF treatment was associated with significant clinical improvement accompanied by oral microbiota remodeling and modulation of inflammatory cytokines. These findings support a potential microbiome-immune axis in PPP and warrant further controlled studies. http://www.itmctr.org, ITMCTR2025001530.