Single-cell analysis uncovers high-proliferative tumour cell subtypes and their interactions in the microenvironment of gastric cancer.

Gastric cancer (GC) remains a prominent malignancy that poses a significant threat to human well-being worldwide. Despite advancements in chemotherapy and immunotherapy, which have effectively augmented patient survival rates, the mortality rate associated with GC remains distressingly high. This can be attributed to the elevated proliferation and invasive nature exhibited by GC. Our current understanding of the drivers behind GC cell proliferation remains limited. Hence, in order to reveal the molecular biological mechanism behind the swift advancement of GC, we employed single-cell RNA-sequencing (scRNA-seq) to characterize the tumour microenvironment in this study. The scRNA-seq data of 27 patients were acquired from the Gene Expression Omnibus database. Differential gene analysis, Gene Ontology, Kyoto Encyclopedia of Genes and Genomes and Gene Set Enrichment Analysis were employed to investigate 38 samples. The copy number variation level exhibited by GC cells was determined using InferCNV. The CytoTRACE, Monocle and Slingshot analysis were used to discern the cellular stemness and developmental trajectory of GC cells. The CellChat package was utilized for the analysis of intercellular communication crosstalk. Moreover, the findings of the data analysis were validated through cellular functional tests conducted on the AGS cell line and SGC-7901 cell line. Finally, this study constructed a risk scoring model to evaluate the differences of different risk scores in clinical characteristics, immune infiltration, immune checkpoints, functional enrichment, tumour mutation burden and drug sensitivity. Within the microenvironment of GC, we identified the presence of 8 cell subsets, encompassing NK_T cells, B_Plasma cells, epithelial cells, myeloid cells, endothelial cells, mast cells, fibroblasts, pericytes. By delving deeper into the characterization of GC cells, we identified 6 specific tumour cell subtypes: C0 PSCA+ tumour cells, C1 CLDN7+ tumour cells, C2 UBE2C+ tumour cells, C3 MUC6+ tumour cells, C4 CHGA+ tumour cells and C5 MUC2+ tumour cells. Notably, the C2 UBE2C+ tumour cells demonstrated a close association with cell mitosis and the cell cycle, exhibiting robust proliferative capabilities. Our findings were fortified through enrichment analysis, pseudotime analysis and cell communication analysis. Meanwhile, knockdown of the transcription factor CREB3, which is highly active in UBE2C+ tumour cells, significantly impedes the proliferation, migration and invasion of GC cells. And the prognostic score model constructed with CREB3-related genes showcased commendable clinical predictive capacity, thus providing valuable guidance for patients' prognosis and clinical treatment decisions. We have identified a highly proliferative cellular subgroup C2 UBE2C+ tumour cells in GC for the first time. The employment of a risk score model, which is based on genes associated with UBE2C expression, exhibits remarkable proficiency in predicting the prognosis of GC patients. In our investigation, we observed that the knockdown of the transcription factor CREB3 led to a marked reduction in cellular proliferation, migration and invasion in GC cell line models. Implementing a stratified treatment approach guided by this model represents a judicious and promising methodology.

Mode-modulation-induced high power dual-wavelength generation in a random distributed feedback Raman fiber laser.

An all-fiberized random distributed feedback Raman fiber laser (RRFL) with mode-modulation-induced wavelength manipulation and dual-wavelength generation has been demonstrated, where an electrically controlled intra-cavity acoustically-induced fiber grating (AIFG) is employed to adjust the input modal content at the signal wavelength. The wavelength agility of both the Raman effect and the Rayleigh backscattering in RRFL benefits on broadband laser output in case of broadband pumping. The feedback modal content at different wavelengths can be adjusted by AIFG, and then the output spectral manipulation can be ultimately manifested through the mode competition in RRFL. Under the efficient mode modulation, the output spectrum can be continuously tuned from 1124.3 nm to 1133.8 nm with single wavelength, while ulteriorly the dual-wavelength spectrum can be formed at 1124.1 nm and 1134.7 nm with a signal-noise-ratio of 45 dB. Throughout, the power is beyond 47 W with good stability and repeatability. To the best of our knowledge, this is the first dual-wavelength fiber laser based on mode modulation and the highest output power ever reported for an all-fiberized continuous wave dual-wavelength fiber laser.

Treatment outcome of tardy ulnar nerve palsy associated with traumatic cubitus valgus by supracondylar shortening wedge rotary osteotomy and ulnar nerve in situ tension release.

BACKGROUND: Tardy ulnar nerve palsy is a common late complication of traumatic cubitus valgus. At present, the treatment of tardy ulnar nerve palsy associated with traumatic cubitus valgus is still controversial, whether these two problems can be corrected safely and effectively in one operation is still unclear. To investigate the supracondylar shortening wedge rotary osteotomy combined with in situ tension release of the ulnar nerve in the treatment of tardy ulnar nerve palsy associated with traumatic cubitus valgus. METHODS: Between 2012 and 2019, 16 patients who had traumatic cubitus valgus deformities with tardy ulnar nerve palsy were treated with simultaneous supracondylar shortening wedge rotary osteotomy and ulnar nerve in situ tension release. we compared a series of indicators of preoperative and postoperative follow-up for at least 24 months, (1) elbow range of motion; (2) the radiographic correction of the preoperative and postoperative humerus-elbow-wrist angles; (3) the static two-point discrimination and grip strength; and (4) the preoperative and postoperative DASH scores of upper limb function. The minimum follow-up was 24 months postoperative (mean, 33 months; range, 24 ~ 44 months). RESULTS: The mean ROM was improved from 107 ° preoperatively to 122 ° postoperatively (P = 0.001). The mean preoperative elbow wrist angle was 24.6 °, and the mean postoperative humerus-elbow wrist angle was 12.1 ° (P < 0.001). The average grip strength and static two-point discrimination improved from 21 kgf and 8 mm to 28 kgf and 4.0 mm (P < 0.001 and P < 0.001, respectively). The ulnar nerve symptoms were improved in all patients except one. The mean HASH score improved from 29 to 16 (P < 0.001). CONCLUSION: Supracondylar shortening wedge rotary osteotomy combined with in situ tension release of ulnar nerve is an effective method for the treatment of traumatic cubitus valgus with tardy ulnar nerve palsy, which restored the normal biomechanical characteristics of the affected limb and improved the elbow joint function.

Efficacy comparison of trifocal bone transport using unilateral external fixator for femoral and tibial bone defects caused by infection.

BACKGROUND: This study aimed to evaluate the clinical and functional outcomes of patients with femoral and tibial critical-sized bone defect (CSBD) treated by trifocal bone transport using the Ilizarov method. METHODS: From March 2011 and January 2017, clinical and radiographic data of patients with CSBD (> 6 cm) caused by infection were documented and analyzed. Patients were divided into the femur group (n = 18) and tibia groups (n = 21) according to the location of bone transport. The bone and functional outcomes were evaluated according to the Association for the Study and Application of the Method of the Ilizarov (ASAMI) criterion, and postoperative complications were evaluated by Paley classification. RESULTS: A total of 39 patients were managed by the trifocal bone transport for the femur (n = 18) or tibia (n = 21) bone defects with a mean follow-up time of 26.1 months (range 17-34 months). Eighteen femurs and 21 tibias with a mean distraction regenerate length (DRL) of 8.3 cm (range 6-13 cm) and 7.5 cm (range 6-11 cm) respectively. Infection was eradicated in all patients, and the total bone union was received in all cases (100%). Statistical difference of bone grade (excellent/good/fair/poor, 3/11/3/1 vs 2/13/4/2, P < 0.05), and function grade (excellent/good/fair/poor, 3/14/1/0 vs 4/13/3/1, P < 0.05) were respectively observed between the femur group and tibia group. The excellent and good rate of bone (femur vs tibia, 77.8% vs 71.4%), and function grade (femur vs tibia, 94.4% vs 80.9%) was higher in the femur group than the tibia. The rate of complication in the femur group was lower than in the tibia (femur vs tibia, 94.4% vs 76.2%). One femur and five tibias were performed additional surgery for delayed union and axial deviation. CONCLUSIONS: The trifocal bone transport using the unilateral external fixator was a practical method in the management of CSBD in the lower extremity. The BUT and EFI of the femur group were shorter than the tibia. Although the complications noted were more frequent on the femur, these were mostly minor.

Over 400 W graded-index fiber Raman laser with brightness enhancement.

The power scaling on all-fiberized Raman fiber oscillator with brightness enhancement (BE) based on multimode graded-index (GRIN) fiber is demonstrated. Thanks to beam cleanup of GRIN fiber itself and single-mode selection properties of the fiber Bragg gratings inscribed in the center of GRIN fiber, the efficient BE is realized. For the laser cavity with single OC FBG, continuous-wave power of 334 W with an M2 value of 2.8 and BE value of 5.6 were obtained at a wavelength of 1120 nm with an optical-to-optical efficiency of 49.6%. Furthermore, the cavity reflectivity is increased by employing two OC FBGs to scale the output power up to 443 W, while the corresponding M2 is 3.5 with BE of 4.2. To our best knowledge, it is the highest power in Raman oscillator based on GRIN fiber.

Kilowatt level Raman amplifier based on 100 µm core diameter multimode GRIN fiber with M2 = 1.6.

In this Letter, we demonstrate a high-power Raman fiber amplifier with excellent beam quality based on graded-index fiber. The Yb-doped fiber laser (YDFL) and bandwidth-tunable amplified spontaneous emission (ASE) source are employed as the pump source to compare the laser performance separately. When the ASE with a bandwidth of 8 nm is employed, a maximum power of 943 W at 1130 nm is achieved, which is twice that pumped by YDFL. The beam quality factor M2 at maximum output power is 1.6, with a brightness enhancement (BE) factor of 27. To the best of our knowledge, this is the best beam quality and BE factor based on pure Raman gain with output power of over 100 W.

Seeing the beam cleanup effect in a high-power graded-index-fiber Raman amplifier based on mode decomposition.

Due to the beam cleanup effect, brightness enhancement (BE) can be achieved in a Raman fiber amplifier (RFA) based on multimode (MM) graded-index (GRIN) fiber. In this Letter, a novel, to the best of our knowledge, diagnostic tool of mode decomposition (MD) based on a stochastic parallel gradient descent algorithm is demonstrated to observe the beam cleanup effect in a GRIN-fiber-based RFA for the first time, to our knowledge. During output power boosting up to 405 W at 1130 nm, the output beam quality factor M2 improves from 3.45 to 2.88, with a BE factor of 10.5. The MD results based on the near-field beam profiles from RFA indicate that the modal weight of the fundamental mode increases from 74.5% to 87%, confirming that the fundamental mode dominates with higher Raman gain. Moreover, the beam quality is found to be limited by the existence of a higher-order (Laguerre-Gaussian) LG10 mode, which is insensitive to the beam cleanup effect. The correlation coefficient reaches over 0.98 for all MD results. Thus, the accuracy of the MD method is high enough to provide further valuable insight into the physics of spatiotemporal beam dynamics in MM GRIN fiber.

Brightness enhancement in random Raman fiber laser based on a graded-index fiber with high-power multimode pumping.

A brightness-enhanced random Raman fiber laser (RRFL) with maximum power of 306 W at 1120 nm is demonstrated. A half-open cavity is built based on a graded-index (GRIN) passive fiber and single high-reflective fiber Bragg grating written in it directly. Due to the beam cleanup effect in the GRIN fiber enhanced in the half-open RRFL cavity, the output beam quality factor M2 is improved from 9.15 (pump) to 1.76-2.35 (Stokes) depending on power, while the pump-Stokes brightness enhancement (BE) factor increases proportionally to output power reaching 6.1 at maximum. To the best of our knowledge, this is the highest power GRIN RRFL with BE.

CuS Nanoparticles as a Photodynamic Nanoswitch for Abrogating Bypass Signaling To Overcome Gefitinib Resistance.

Bypass signaling activation plays a crucial role in the acquired resistance of gefitinib, the first targeted drug in the clinic to treat advanced non-small cell lung cancer. Although the inactivation of bypass signaling by small-molecule inhibitors or monoclonal antibodies may overcome gefitinib resistance, their clinical use has been limited by the complex production process and off-target toxicity. Here we show CuS nanoparticles (NPs) behaved as a photodynamic nanoswitch to specifically abrogate overactive bypass signaling in resistant tumor cells without interfering with the same signal pathways in normal cells. In representative insulin growth factor-1 receptor (IGF1R) bypass activation-induced gefitinib resistant tumors, CuS NPs upon near-infrared laser irradiation locally elevated reactive oxygen species (ROS) level in tumor cells, leading to the blockage of bypass IGF1R and its downstream AKT/ERK/NF-κB signaling cascades. Consequently, laser-irradiated CuS NPs sensitized tumors to gefitinib treatment and prolonged the survival of mice with no obvious toxicity. Laser-irradiated CuS NPs may serve as a simple and safe nanomedicine strategy to overcome bypass activation-induced gefitinib resistance in a specific and controllable manner and provide insights into the treatment of a myriad of other resistant tumors in the field of cancer therapy.

para-Aminothiophenol Radical Reaction-Functionalized Gold Nanoprobe for One-to-All Detection of Five Reactive Oxygen Species In Vivo.

Five major reactive oxygen species (ROS) are generated in diseases including H2O2, •OH, O2•-, ROO•, and 1O2. Simultaneous detection of the five ROS with a single probe is crucial for a comprehensive understanding of the development and progression of many diseases, such as cancer and inflammatory diseases. However, currently reported detection systems are limited by targeting one ROS with one probe. This one-to-one detection mode may fail to sufficiently unveil the diseased state. In this study, we achieved simultaneous detection of all the five ROS with one probe (i.e., one-to-all detection), by designing a novel para-aminothiophenol (PATP) and hemin-decorated gold (Au/PATP/Hemin) nanoprobe. The design is principled by our discovery that PATP can react with •OH, O2•-, ROO•, and 1O2 by a radical oxidative coupling mechanism to form 4,4'-dimercaptoazobenzene (DMAB). The DMAB then elicited strong characteristic surface-enhanced Raman scattering (SERS) peaks at 1142, 1386, and 1432 cm-1; which in turn enables direct detection of •OH, O2•-, ROO•, and 1O2 and indirect detection of H2O2 by hemin-catalyzed fenton reaction to convert H2O2 into •OH. In two representative ROS-elevated mice models of tumors and allergic dermatitis, the Au/PATP/Hemin nanoprobe demonstrated its robust performance of monitoring tumor development and inflammation progression in a highly sensitive and quantitative manner.

Multiparameter Analysis-Based Electrochemiluminescent Assay for Simultaneous Detection of Multiple Biomarker Proteins on a Single Interface.

Electrochemiluminescent (ECL) assay with high sensitivity has been considered as one of the potential strategies to simultaneously detect multiple biomarker proteins. However, it was essential, but full of challenges, to overcome the limitation caused by cross reactions among different ECL indicators. Herein, the multiparameter analysis of ECL-potential signals demonstrated by multivariate linear algebraic equations was first employed in the simultaneous ECL assay to realize multiple detection of biomarker proteins on a single interface. Additionally, owing to the exponential amplification of self-synthesized nucleotide dendrimer by hybridization chain reaction (HCR) and rolling circle amplification (RCA), the developed simultaneous ECL assay showed improved sensitivity and satisfactory accuracy for the detection of N-terminal of the prohormone brain natriuretic peptide (BNPT) and cardiac troponin I (cTnI). Furthermore, a self-designed magnetic beads-based flow system was also employed to improve the feasibility and analysis speed of the simultaneous ECL assay. Importantly, the proposed strategy enabled simultaneous detection of multiple biomarker proteins simply, which could be readily expanded for the multiplexed estimation of various kinds of proteins and nucleotide sequence also, revealing a new avenue for early disease diagnosis with higher efficiency.

The relationship between health insurance and economic performance: an empirical study based on meta-analysis.

Health insurance stands as a pivotal facet of social wellbeing, with profound implications for the overarching landscape of economic development. The existing research, however, lacks consensus on the relationship between health insurance and economic performance and provides no evidence about the magnitude of the correlation. This lack of information seriously impedes the high-quality development of the healthcare system. Therefore, to scientifically elucidate the relationship between the two, this study involved a meta-analysis, analyzing 479 effect values derived from 34 independent research samples. The results reveal a strongly positive correlation between health insurance and economic performance [r = 0.429, 95% CI = (0.381, 0.475)]. Findings show that health insurance in developed countries more effectively fosters economic performance than in developing countries. Moreover, public health insurance exerts a stronger promoting effect on economic performance than commercial health insurance. The relationship between health insurance and economic performance is moderated by data type, research method, country of sample origin, literature type, journal impact factor, publication year, type of health insurance, and the research populations. Based on meta-analysis, this study not only scientifically responds to the controversy of the relationship between health insurance and economic performance, and the magnitude of a correlation, but also further reveals the inner conduction mechanism between the two. Our research findings are meaningful for policymakers to choose an appropriate healthcare strategy according to their unique attributes, propelling sustainable economic development.

Asynchronous control of networked periodic piecewise linear systems under time-varying transmission delay.

This paper studies the stability of networked periodic piecewise linear systems (NPPLSs) of which communication channels are subject to a time-varying transmission delay. Under data-sampling, the NPPLS is modeled as an asynchronous controlled periodic piecewise linear system with time-varying input delay, where the interval of asynchronous control in each subsystem is uncertain but bounded. Aimed at obtaining less conservative stability and synthesis results when tackling the uncertain switching, the dwell time of each subsystem is divided into three subintervals according to the asynchronous control interval and the maximum delay assumption. In this way, it allows the construction of piecewise Lyapunov functionals for one subsystem, the piecewise Lyapunov functionals capture different dynamic characteristics in each subinterval. Using this approach, general stability conditions of NPPLSs are derived. Then, by constructing Lyapunov functionals as a set of delay-dependent time-varying functionals, tractable exponential stability conditions of NPPLSs under transmission delay are developed using a scaling technique. Due to the coupling of the decision variables in the conditions, an iterative algorithm is proposed to solve the periodic controller gains. A numerical example is provided to illustrate the merits of the obtained controllers.

Balancing Fuzzy Control Performance and Reachable-Set-Dependent Event-Triggered Communication.

This work investigates the control design and reachable set estimation problem for T-S fuzzy systems under bounded external disturbances. By designing a fuzzy controller, the reachable set of the closed-loop fuzzy system can be bounded by an ellipsoid that is minimized via two optimization algorithms. When the fuzzy control scheme is implemented under an event-triggered framework, the limited information availability may degrade desired control performance due to only the triggered state being available for the fuzzy controller. To alleviate this problem, a novel reachable-set-dependent event-triggering condition is proposed such that satisfactory control performance can still be ensured by the above fuzzy controller. Meanwhile, the bounding ellipsoid of the reachable set is explicitly given, and a relationship is established between control performance and communication cost under the event-triggered scheme. In addition, under the condition that the reachable set remains within a specified safe area, the communication rate can be reduced while still achieving optimal control performance. The proposed strategies are verified via a simulation example.

Diagnostic efficacy of combining diffusion-weighted magnetic resonance imaging with serum Mucin 1, Mucin 13, and Mucin 16 in distinguishing borderline from malignant epithelial ovarian tumors.

AIMS: To enhance ovarian tumor diagnosis beyond conventional methods, this study explored combining diffusion-weighted magnetic resonance imaging (DWI-MRI) and serum biomarkers (Mucin 1 [MUC1], MUC13, and MUC16) for distinguishing borderline from malignant epithelial ovarian tumors. METHODS: A total of 126 patients, including 71 diagnosed with borderline (BEOTs) and 55 with malignant epithelial ovarian tumors (MEOTs), underwent preoperative DWI-MRI. Region of interest (ROI) was manually drawn along the solid component's boundary of the largest tumor, focusing on areas with potentially the lowest apparent diffusion coefficient (ADC). For entirely cystic tumors, a free-form ROI enclosed the maximum number of septa while targeting the lowest ADC. Serum biomarkers were determined using enzyme-linked immunosorbent assay. RESULTS: Basic morphological traits proved inadequate for malignancy diagnosis, warranting this investigation. BEOTs had an ADC mean of (1.670 ± 0.250) × 103 mm2 /s, while MEOTs had a lower ADC mean of (1.332 ± 0.481) × 103 mm2 /s, with a sensitivity of 63.6% and specificity of 90.1%. Median MUC1 (167.0 U/mL vs. 87.3 U/mL), MUC13 (12.44 ng/mL vs. 7.77 ng/mL), and MUC16 (180.6 U/mL vs. 36.1 U/mL) levels were higher in MEOTs patients. The biomarker performance was: MUC1, sensitivity 50.9%, specificity 100%; MUC13, sensitivity 56.4%, specificity 78.9%; MUC16, sensitivity 83.64%, specificity 100%. Combining serum biomarkers and ADC mean resulted in a sensitivity of 96.4% and specificity of 100%. CONCLUSION: The integration of DWI-MRI with serum biomarkers (MUC1, MUC13, and MUC16) achieves exceptional diagnostic accuracy, offering a powerful tool for the precise differentiation between borderline and malignant epithelial ovarian tumors.

Thyroid hormone enhances efficacy of cisplatin in lung cancer patients via down-regulating GLUT1 expression and reversing the Warburg effect.

Cisplatin (CDDP) is a standard non-small cell lung cancer (NSCLC) chemotherapy, but its efficacy is hampered by resistance, partly due to the Warburg effect. This study investigates how thyroid hormones enhance the Warburg effect, increasing sensitivity to cisplatin in lung cancer. Clinical data from advanced NSCLC patients were analyzed based on thyroid hormone levels, categorizing patients into high and low groups. Cellular experiments involved Control, 10uM CDDP, 10uM CDDP + 0.1uM T3, and 10uM CDDP + 0.1uM T4 categories. Parameters were measured in A549 and PC9 lung cancer cells, including proliferation, apoptosis, mitochondrial membrane potential, ROS production, glycolysis enzyme activity, lactic acid level, and ATP content. Gene and protein expressions were assessed using qPCR and Western Blot. Analysis revealed higher FT3 levels correlated with prolonged progression-free survival before chemotherapy (median PFS: high FT3 group = 12.67 months, low FT3 group = 7.03 months, p = 0.01). Cellular experiments demonstrated that thyroid hormones increase lung cancer cell sensitivity to cisplatin, inhibiting proliferation and enhancing efficacy. The mechanism involves thyroid hormones and cisplatin jointly down-regulating MSI1/AKT/GLUT1 expression, reducing lactic acid and glycolysis. This Warburg effect reversal boosts ATP levels, elevates ROS, and decreases MMP, enhancing cisplatin effectiveness in A549 and PC9 cells. In conclusion, elevated free T3 levels in advanced NSCLC patients correlate with prolonged progression-free survival under cisplatin chemotherapy. Cellular experiments reveal that thyroid hormones enhance lung cancer cell sensitivity to cisplatin by reversing the Warburg effect, providing a mechanistic basis for improved therapeutic outcomes.

The Relationship between Health Insurance and Pharmaceutical Innovation: An Empirical Study Based on Meta-Analysis.

The growing research interest in the relationship between health insurance and pharmaceutical innovation is driven by their significant impact on healthcare optimization and pharmaceutical development. The existing literature, however, lacks consensus on this relationship and provides no evidence of the magnitude of a correlation. In this context, this study employs meta-analysis to explore the extent to which health insurance affects pharmaceutical innovation. It analyzes 202 observations from 14 independent research samples, using the regression coefficient of health insurance on pharmaceutical innovation as the effect size. The results reveal that there is a strong positive correlation between health insurance and pharmaceutical innovation (r = 0.367, 95% CI = [0.294, 0.436]). Public health insurance exhibits a stronger promoting effect on pharmaceutical innovation than commercial health insurance. The relationship between health insurance and pharmaceutical innovation is moderated by the country of sample origin, data range, journal type, journal impact factor, type of health insurance, and research perspective. Our research findings further elucidate the relationship mechanism between health insurance and pharmaceutical innovation, providing a valuable reference for future explorations in pharmaceutical fields.

Output Reachable Set-Based Leader-Following Consensus of Positive Agents Over Switching Networks.

This work addresses the output reachable set-based leader-following consensus problem, focusing on a group of positive agents over directed dwell-time switching networks. Two types of non-negative disturbances, namely, 1) L1 -norm bounded disturbances and 2) L∞,1 -norm bounded disturbances are studied. Meanwhile, a class of directed dwell-time switching networks for modeling the communication protocol of positive agents is investigated. To deal with the presence of disturbances, an output-feedback control protocol is developed to achieve a robust consensus with positivity preserved based on the output reachable set. By exploiting the positive characteristics, switched linear copositive Lyapunov functions are adopted to establish output reachable set-based consensus conditions. These conditions can facilitate the control protocol design by solving a bilinear programming problem, and also generate hyperpyramidal regions to confine the output consensus error. A particle swarm optimization-based (PSO-based) algorithm is applied to compute the controller gains and optimize the volume of the hyperpyramids. The proposed methods are verified by the presented numerical case studies.

Graph Reasoning Module for Alzheimer's Disease Diagnosis: A Plug-and-Play Method.

Recent advances in deep learning have led to increased adoption of convolutional neural networks (CNN) for structural magnetic resonance imaging (sMRI)-based Alzheimer's disease (AD) detection. AD results in widespread damage to neurons in different brain regions and destroys their connections. However, current CNN-based methods struggle to relate spatially distant information effectively. To solve this problem, we propose a graph reasoning module (GRM), which can be directly incorporated into CNN-based AD detection models to simulate the underlying relationship between different brain regions and boost AD diagnosis performance. Specifically, in GRM, an adaptive graph Transformer (AGT) block is designed to adaptively construct a graph representation based on the feature map given by CNN, a graph convolutional network (GCN) block is adopted to update the graph representation, and a feature map reconstruction (FMR) block is built to convert the learned graph representation to a feature map. Experimental results demonstrate that the insertion of the GRM in the existing AD classification model can increase its balanced accuracy by more than 4.3%. The GRM-embedded model achieves state-of-the-art performance compared with current deep learning-based AD diagnosis methods, with a balanced accuracy of 86.2%.

Comparing Multi-Dimensional fNIRS Features Using Bayesian Optimization-Based Neural Networks for Mild Cognitive Impairment (MCI) Detection.

The diagnosis of mild cognitive impairment (MCI), a prodromal stage of Alzheimer's disease (AD), is essential for initiating timely treatment to delay the onset of AD. Previous studies have shown the potential of functional near-infrared spectroscopy (fNIRS) for diagnosing MCI. However, preprocessing fNIRS measurements requires extensive experience to identify poor-quality segments. Moreover, few studies have explored how proper multi-dimensional fNIRS features influence the classification results of the disease. Thus, this study outlined a streamlined fNIRS preprocessing method to process fNIRS measurements and compared multi-dimensional fNIRS features with neural networks in order to explore how temporal and spatial factors affect the classification of MCI and cognitive normality. More specifically, this study proposed using Bayesian optimization-based auto hyperparameter tuning neural networks to evaluate 1D channel-wise, 2D spatial, and 3D spatiotemporal features of fNIRS measurements for detecting MCI patients. The highest test accuracies of 70.83%, 76.92%, and 80.77% were achieved for 1D, 2D, and 3D features, respectively. Through extensive comparisons, the 3D time-point oxyhemoglobin feature was proven to be a more promising fNIRS feature for detecting MCI by using an fNIRS dataset of 127 participants. Furthermore, this study presented a potential approach for fNIRS data processing, and the designed models required no manual hyperparameter tuning, which promoted the general utilization of fNIRS modality with neural network-based classification to detect MCI.