Rapid identification of moxa wool storage period based on hyperspectral imaging technology and machine learning.

Moxa wool (MW), derived from the dried leaves of A r t e m i s i a a r g y i , plays a significant role in traditional Chinese medicine. However, the quality of MW varies with its storage period, impacting its therapeutic efficacy. Traditional methods for quality detection are limited and destructive. To address this, we propose a non-destructive detection method using hyperspectral imaging technology and machine learning algorithms to accurately identify the storage period of MW. Nevertheless, hyperspectral data poses challenges due to its high dimensionality and redundancy, leading to increased computational complexity. To overcome this, we employed principal component analysis (PCA), competitive adaptive reweighted sampling (CARS), and successive projection algorithm (SPA) for data dimensionality reduction and wavelength selection. The results demonstrate that these techniques significantly enhance the accuracy of MW storage year identification. For Nanyang MW, the CARS+SVM model achieved the highest accuracy rates of 99.8% in the visible-near-infrared (VNIR) range and 99.55% in the shortwave infrared (SWIR) range. Similarly, for Qichun MW, the SPA+SVM model achieved identification accuracies of 99.78% and 99.47% in the VNIR and SWIR ranges, respectively. This research provides valuable insights into the rapid detection of MW quality by indication of storage years and presents a novel approach for quality control of MW in the field of traditional Chinese medicine. The combination of hyperspectral imaging and machine learning offers a promising solution for efficient and accurate MW identification, contributing to the advancement of traditional medicine practices.

Identification of chrysanthemum variety via hyperspectral imaging and wavelength selection based on multitask particle swarm optimization.

Chrysanthemum, a widely favored flower tea, contains numerous phytochemicals for health benefits. Due to the different geographical origins and processing technics, its variety has a direct influence on the phytochemical content and pharmacological effect. Accordingly, an accurate identification for chrysanthemum varieties is significant for quality detection and market supervision. In this study, the hyperspectral imaging (HSI) combined with chemometrics methods was exploited to identify the chrysanthemum varieties. First, to alleviate the problem of easily trapping into local optimum in traditional spectral variable selection methods, the multi-tasking particle swarm optimization (MTPSO) was developed to select the key wavelengths by dividing hundreds of variables into low-dimensional subtasks. Second, to enrich the feature information, the spatial texture and color features contained in hyperspectral images were extracted and applied to chrysanthemum identification for the first time. Finally, an ensemble learning model, extreme gradient boosting (XGBoost), was constructed to conduct the chrysanthemum variety classification due to its strong generalization ability. Experimental results showed that the proposed MTPSO achieved the identification accuracy of 96.89%, and increased by 1.11-5.91% than classical spectral feature selection methods. Furthermore, after the involvement of spatial image information, the classification accuracy using spatial-spectral features was improved further, and reached 98.39%. Overall, this study highlights that the feature fusion of key wavelengths and spatial information is more effective for chrysanthemum variety identification, and can also provide technical reference for other HSI-related applications.

Injectable Chitosan Hydrogels Doped with 2D Peptide Nanosheet-Drug Conjugates for Glutathione-Responsive Sustained Drug Delivery.

The development of novel and effective drug delivery systems aimed at enhancing therapeutic profile and efficacy of therapeutic agents is a critical challenge in modern medicine. This study presents an intelligent drug delivery system based on self-assembled two-dimensional peptide nanosheets (2D PNSs). Leveraging the tunable properties of amino acid structures and sequences, we design a peptide with the sequence of Fmoc-FKKGSHC, which self-assembles into 2D PNSs with uniform structure, high biocompatibility, and excellent degradability. Covalent attachment of thiol-modified doxorubicin (DOX) drugs to 2D PNSs via disulfide bond results in the peptide-drug conjugates (PDCs), which is denoted as PNS-SS-DOX. Subsequently, the PDCs are encapsulated within the injectable, thermosensitive chitosan (CS) hydrogels for drug delivery. The designed drug delivery system demonstrates outstanding pH-responsiveness and sustained drug release capabilities, which are facilitated by the characteristics of the CS hydrogels. Meanwhile, the covalently linked disulfide bond within the PNS-SS-DOX is responsive to intracellular glutathione (GSH) within tumor cells, enabling controlled drug release and significantly inhibiting the cancer cell growth. This responsive peptide-drug conjugate based on a 2D peptide nanoplatform paves the way for the development of smart drug delivery systems and has bright prospects in the future biomedicine field.

Recent advances in peptide-based bioactive hydrogels for nerve repair and regeneration: from material design to fabrication, functional tailoring and applications.

The injury of both central and peripheral nervous systems can result in neurological disorders and severe nervous diseases, which has been one of the challenges in the medical field. The use of peptide-based hydrogels for nerve repair and regeneration (NRR) provides a promising way for treating these problems, but the effects of the functions of peptide hydrogels on the NRR efficiency have been not understood clearly. In this review, we present recent advances in the material design, matrix fabrication, functional tailoring, and NRR applications of three types of peptide-based hydrogels, including pure peptide hydrogels, other component-functionalized peptide hydrogels, and peptide-modified polymer hydrogels. The case studies on the utilization of various peptide-based hydrogels for NRR are introduced and analyzed, in which the effects and mechanisms of the functions of hydrogels on NRR are illustrated specifically. In addition, the fabrication of medical NRR scaffolds and devices for pre-clinical application is demonstrated. Finally, we provide potential directions on the development of this promising topic. This comprehensive review could be valuable for readers to know the design and synthesis strategies of bioactive peptide hydrogels, as well as their functional tailoring, in order to promote their practical applications in tissue engineering, biomedical engineering, and materials science.

Assembling Ag2S quantum dots onto peptide nanosheet as a biomimetic two-dimensional nanoplatform for synergistic near infrared-II fluorescent imaging and photothermal therapy of tumor.

Fluorescent bioimaging and photothermal therapy (PTT) techniques have potential significance in cancer diagnosis and treatment and have been widely applied in biomedical and practical clinical trials. This study proposes the molecular design and biofabrication of a two-dimensional (2D) nanoplatform, exhibiting promising prospects for synergistic bioimaging and PTT of tumors. First, biocompatible 2D peptide nanosheets (PNSs) were designed and prepared through peptide self-assembly. These served as a support matrix for assembling polyethylene glycol-modified Ag2S quantum dots (PEG-Ag2SQDs) to form a 2D nanoplatform (PNS/PEG-Ag2SQDs) with unique fluorescent and photothermal properties. The designed 2D nanoplatform not only showed improved photothermal efficacy and an elevated photothermal conversion efficiency of 52.46 %, but also demonstrated significant lethality against tumors in both in vitro and in vivo cases. Additionally, it displays excellent imaging effects in the near-infrared II region, making it suitable for synergistic fluorescent imaging-guided PTT of tumors. This study not only provides a facile approach for devising and synthesizing 2D peptide assemblies but also presents new biomimetic strategies to create functional 2D organic/inorganic nanoplatforms for biomedical applications.

Nomogram for predicting venous thromboembolism after spinal surgery.

PURPOSE: This study aimed to establish a nomogram to predict the risk of venous thromboembolism (VTE), identifying potential risk factors, and providing theoretical basis for prevention of VTE after spinal surgery. METHODS: A retrospective analysis was conducted on 2754 patients who underwent spinal surgery. The general characteristics of the training group were initially screened using univariate logistic analysis, and the LASSO method was used for optimal prediction. Subsequently, multivariate logistic regression analysis was performed to identify independent risk factors for postoperative VTE in the training group, and a nomogram for predict risk of VTE was established. The discrimination, calibration, and clinical usefulness of the nomogram were separately evaluated using the C-index, receiver operating characteristic curve, calibration plot and clinical decision curve, and was validated using data from the validation group finally. RESULTS: Multivariate logistic regression analysis identified 10 independent risk factors for VTE after spinal surgery. A nomogram was established based on these independent risk factors. The C-index for the training and validation groups indicating high accuracy and stability of the model. The area under the receiver operating characteristic curve indicating excellent discrimination ability; the calibration curves showed outstanding calibration for both the training and validation groups. Decision curve analysis showed the clinical net benefit of using the nomogram could be maximized in the probability threshold range of 0.01-1. CONCLUSION: Patients undergoing spinal surgery with elevated D-dimer levels, prolonger surgical, and cervical surgery have higher risk of VTE. The nomogram can provide a theoretical basis for clinicians to prevent VTE.

Non-destructive prediction of isoflavone and starch by hyperspectral imaging and deep learning in Puerariae Thomsonii Radix.

Accurate assessment of isoflavone and starch content in Puerariae Thomsonii Radix (PTR) is crucial for ensuring its quality. However, conventional measurement methods often suffer from time-consuming and labor-intensive procedures. In this study, we propose an innovative and efficient approach that harnesses hyperspectral imaging (HSI) technology and deep learning (DL) to predict the content of isoflavones (puerarin, puerarin apioside, daidzin, daidzein) and starch in PTR. Specifically, we develop a one-dimensional convolutional neural network (1DCNN) model and compare its predictive performance with traditional methods, including partial least squares regression (PLSR), support vector regression (SVR), and CatBoost. To optimize the prediction process, we employ various spectral preprocessing techniques and wavelength selection algorithms. Experimental results unequivocally demonstrate the superior performance of the DL model, achieving exceptional performance with mean coefficient of determination (R2) values surpassing 0.9 for all components. This research underscores the potential of integrating HSI technology with DL methods, thereby establishing the feasibility of HSI as an efficient and non-destructive tool for predicting the content of isoflavones and starch in PTR. Moreover, this methodology holds great promise for enhancing efficiency in quality control within the food industry.

The Identification of Fritillaria Species Using Hyperspectral Imaging with Enhanced One-Dimensional Convolutional Neural Networks via Attention Mechanism.

Combining deep learning and hyperspectral imaging (HSI) has proven to be an effective approach in the quality control of medicinal and edible plants. Nonetheless, hyperspectral data contains redundant information and highly correlated characteristic bands, which can adversely impact sample identification. To address this issue, we proposed an enhanced one-dimensional convolutional neural network (1DCNN) with an attention mechanism. Given an intermediate feature map, two attention modules are constructed along two separate dimensions, channel and spectral, and then combined to enhance relevant features and to suppress irrelevant ones. Validated by Fritillaria datasets, the results demonstrate that an attention-enhanced 1DCNN model outperforms several machine learning algorithms and shows consistent improvements over a vanilla 1DCNN. Notably under VNIR and SWIR lenses, the model obtained 98.97% and 99.35% for binary classification between Fritillariae Cirrhosae Bulbus (FCB) and other non-FCB species, respectively. Additionally, it still achieved an extraordinary accuracy of 97.64% and 98.39% for eight-category classification among Fritillaria species. This study demonstrated the application of HSI with artificial intelligence can serve as a reliable, efficient, and non-destructive quality control method for authenticating Fritillaria species. Moreover, our findings also illustrated the great potential of the attention mechanism in enhancing the performance of the vanilla 1DCNN method, providing reference for other HSI-related quality controls of plants with medicinal and edible uses.

Identification of Turtle-Shell Growth Year Using Hyperspectral Imaging Combined with an Enhanced Spatial-Spectral Attention 3DCNN and a Transformer.

Turtle shell (Chinemys reecesii) is a prized traditional Chinese dietary therapy, and the growth year of turtle shell has a significant impact on its quality attributes. In this study, a hyperspectral imaging (HSI) technique combined with a proposed deep learning (DL) network algorithm was investigated for the objective determination of the growth year of turtle shells. The acquisition of hyperspectral images was carried out in the near-infrared range (948.72-2512.97 nm) from samples spanning five different growth years. To fully exploit the spatial and spectral information while reducing redundancy in hyperspectral data simultaneously, three modules were developed. First, the spectral-spatial attention (SSA) module was developed to better protect the spectral correlation among spectral bands and capture fine-grained spatial information of hyperspectral images. Second, the 3D convolutional neural network (CNN), more suitable for the extracted 3D feature map, was employed to facilitate the joint spatial-spectral feature representation. Thirdly, to overcome the constraints of convolution kernels as well as better capture long-range correlation between spectral bands, the transformer encoder (TE) module was further designed. These modules were harmoniously orchestrated, driven by the need to effectively leverage both spatial and spectral information within hyperspectral data. They collectively enhance the model's capacity to extract joint spatial and spectral features to discern growth years accurately. Experimental studies demonstrated that the proposed model (named SSA-3DTE) achieved superior classification accuracy, with 98.94% on average for five-category classification, outperforming traditional machine learning methods using only spectral information and representative deep learning methods. Also, ablation experiments confirmed the effectiveness of each module to improve performance. The encouraging results of this study revealed the potentiality of HSI combined with the DL algorithm as an efficient and non-destructive method for the quality control of turtle shells.