Potential Clinical Applications Research Articles

Development of macrophage M2 relate signature for predicting prognosis and immunotherapy response in ovarian cancer

BackgroundOvarian cancer ranks as the fifth most common cause of cancer-related deaths in women worldwide. Macrophages M2 is believed to support tumor growth by suppressing immune responses and promoting angiogenesis.MethodsA macrophage M2-related signature (MRS) was developed by applying machine learning methods that included 10 algorithms and utilized data from the TCGA, GSE14764 and GSE140082 datasets. The predictive capacity of the MRS for immunotherapy response was evaluated through various methods, including immunophenoscore, TIDE score, TMB score, immune escape score, as well as two immunotherapy cohorts (IMvigor210 and GSE91061).ResultsThe optimal MRS, developed using the lasso algorithm, served as an independent prognostic factor and demonstrated stable performance in predicting overall survival rates in ovarian cancer. In the TCGA dataset, the AUC values for the 1-, 3-, and 5-year ROC curves were 0.874, 0.808, and 0.813, respectively. The C-index of our MRS was superior to that of clinical stage, tumor grade, and several other established prognostic signatures. Ovarian cancer patients with low risk score exhibited higher ESTIMATE score, increased levels of immune cells, elevated PDI&CTLA4 immunophenoscore, higher TMB score, reduced TIDE score, and lower immune escape score. Additionally, the survival prediction nomogram displayed significant potential for clinical application in estimating the 1-, 3-, and 5-year overall survival rates of ovarian cancer patients.ConclusionOur study developed a novel MRS for ovarian cancer, which could act as an indicator for predicting the prognosis and immunotherapy response in ovarian cancer.

Explainable predictive models of short stature and exploration of related environmental growth factors: a case-control study

BackgroundShort stature is a prevalent pediatric endocrine disorder for which early detection and prediction are pivotal for improving treatment outcomes. However, existing diagnostic criteria often lack the necessary sensitivity and specificity because of the complex etiology of the disorder. Hence, this study aims to employ machine learning techniques to develop an interpretable predictive model for normal-variant short stature and to explore how growth environments influence its development.MethodsWe conducted a case‒control study including 100 patients with normal-variant short stature who were age-matched with 200 normal controls from the Endocrinology Department of Nanjing Children’s Hospital from April to September 2021. Parental surveys were conducted to gather information on the children involved. We assessed 33 readily accessible medical characteristics and utilized conditional logistic regression to explore how growth environments influence the onset of normal-variant short stature. Additionally, we evaluated the performance of the nine machine learning algorithms to determine the optimal model. The Shapley additive explanation (SHAP) method was subsequently employed to prioritize factor importance and refine the final model.ResultsIn the multivariate logistic regression analysis, children’s weight (OR = 0.92, 95% CI: 0.86, 0.99), maternal height (OR = 0.79, 95% CI: 0.72, 0.87), paternal height (OR = 0.83, 95% CI: 0.75, 0.91), sufficient nighttime sleep duration (OR = 0.48, 95% CI: 0.26, 0.89), and outdoor activity time exceeding three hours (OR = 0.02, 95% CI: 0.00, 0.66) were identified as protective factors for normal-variant short stature. This study revealed that parental height, caregiver education, and children’s weight significantly influenced the prediction of normal-variant short stature risk, and both the random forest model and gradient boosting machine model exhibited the best discriminatory ability among the 9 machine learning models.ConclusionsThis study revealed a close correlation between environmental growth factors and the occurrence of normal-variant short stature, particularly anthropometric characteristics. The random forest model and gradient boosting machine model performed exceptionally well, demonstrating their potential for clinical applications. These findings provide theoretical support for clinical identification and preventive measures for short stature.

SA-Seg: Annotation-Efficient Segmentation for Airway Tree Using Saliency-based Annotation.

Segmentation of the airway tree plays a vital role in clinical practice. However, the complex airway tree structure makes it quite challenging to annotate accurately. Although some annotation-efficient methods have shown promising results in medical image segmentation, most are developed for locally focused segmentation objects and are incompatible with the airway. In this work, we propose an annotation-efficient segmentation method to improve annotation efficiency and tree completeness. It includes a new efficient annotation way and an accompanying segmentation method. The saliency-based annotation method only needs to annotate high-saliency regions, thus greatly improving the annotation efficiency. Inspired by positive-unlabeled learning, we model the dependency relationship between key items in the annotation process to learn from biased weak annotation. The probabilistic model of the annotation process is divided into the score function and the bias function. The score function models the uniform foreground feature representation of the airway, while the bias function models the saliency bias between labeled and unlabeled airway regions. Then, the two models are implemented with convolutional neural networks and optimized by applying an EM algorithm during training. Experimental results reveal that our approach saves 89% annotation time and significantly narrows the performance gap between weak and full annotations. This highlights its potential for clinical applications.

An artificial intelligence model for nuclear grading of clear cell renal cell carcinoma using whole slide images: a retrospective, multicentre, diagnostic study.

An artificial intelligence model for nuclear grading of clear cell renal cell carcinoma using whole slide images: a retrospective, multicentre, diagnostic study.

REL/DPA/AVI method: a novel approach for rapid detection of carbapenemase-producing Enterobacterales directly from positive blood cultures based on optical density.

The relebactam, dipicolinic acid, and avibactam sodium (REL/DPA/AVI) method has demonstrated significant success in identifying and differentiating carbapenemase-producing Enterobacterales (CPE) from positive blood cultures, exhibiting superior performance compared with existing technologies. Although numerous advanced technologies such as mNGS, Filmarray, Verigene, and NG-Test CARBA 5 DetecTool have been developed for carbapenemase typing of CPE in positive blood cultures, our method is distinguished by a significant economic advantage, with a cost of less than $1 USD per test. This substantial cost-effectiveness underscores the immense potential for widespread clinical applications.

SSL-DA: Semi-and Self-Supervised Learning with Dual Attention for Echocardiogram Segmentation.

Echocardiogram analysis plays a crucial role in assessing and diagnosing cardiac function, providing essential data to support medical diagnoses of heart disease. A key task, accurately identifying and segmenting the left ventricle (LV) in echocardiograms, remains challenging and labor-intensive. Current automated cardiac segmentation methods often lack the necessary accuracy and reproducibility, while semi-automated or manual annotations are excessively time-consuming. To address these limitations, we propose a novel segmentation framework, semi-and self-supervised learning with dual attention (SSL-DA) for echocardiogram segmentation. We start with a temporal masking network for pre-training. This network captures valuable information, such as echocardiogram periodicity. It also provides optimized initialization parameters for LV segmentation. We then employ a semi-supervised network to automatically segment the left ventricle, enhancing the model's learning with channel and spatial attention mechanisms to capture global channel dependencies and spatial dependencies across annotations. We evaluated SSL-DA on the publicly available EchoNet-Dynamic dataset, achieving a Dice similarity coefficient of 93.34% (95% CI, 93.23-93.46%), outperforming most prior CNN-based models. To further assess the generalization ability of SSL-DA, we conducted ablation experiments on the CAMUS dataset. Experimental results confirm that SSL-DA can quickly and accurately segment the left ventricle in echocardiograms, showing its potential for robust clinical application.

Dual-Mode Detection of Trypsin and Screening of Trypsin Inhibitors on the Basis of a Colorimetric/Fluorescence Sensing Probe.

Trypsin is a specific and reliable diagnostic biomarker of pancreatic function and pathological changes. Thus, detecting trypsin and screening trypsin inhibitors are highly important for clinical diagnosis and disease treatment. Herein, we propose a colorimetric/fluorescence sensing probe based on carbon quantum dots (O-CDs) and cytochrome C (Cyt C) for sensitive and specific trypsin detection and trypsin inhibitor screening. The prepared O-CDs exhibit a persistent fluorescence and environmental tolerance. In the presence of trypsin, Cyt C is hydrolyzed into a heme with oxidase mimetic activity that can convert colorless TMB into blue oxTMB, which strongly absorbs at 652 nm and quenches the fluorescence of the O-CDs at 580 nm via the inner filter effect. This sensing probe realizes the precise determination of trypsin with a wide linear range (10-2000 ng/mL in both colorimetric and fluorescence modes) and a low limit of detection (0.05 ng/mL in colorimetric mode and 3.44 ng/mL in fluorescence mode). The developed method was then applied to detect trypsin in human serum and urine with good accuracy and reproducibility. Furthermore, the sensing probe was successfully applied to screen trypsin inhibitors from legumes on the basis of their ability to inhibit the trypsin activity. Thus, this dual-mode detection method has great potential for clinical application in the diagnosis of trypsin-related diseases, as well as for screening trypsin inhibitor drugs.

A machine learning-based prediction model for colorectal liver metastasis

Colorectal liver metastasis (CRLM) is a primary factor contributing to poor prognosis and metastasis in colorectal cancer (CRC) patients. This study aims to develop and validate a machine learning (ML)-based risk prediction model using conventional clinical data to forecast the occurrence of CRLM. This retrospective study analyzed the clinical data of 865 CRC patients between January 2018 and September 2024. Patients were categorized into non-CRLM and CRLM groups. The least absolute shrinkage and selection operator regression was employed to identify key clinical variables, and five ML algorithms were utilized to develop prediction models. The optimal model was selected based on performance metrics including the receiver operating characteristic curve, precision-recall curve, decision curve analysis, and calibration curve, which collectively evaluated both the predictive accuracy and clinical utility of the model. Among the five ML algorithms evaluated, Random forest demonstrated the best performance. Leveraging the Random forest algorithm, we developed the CRLM-Lab6 prediction model, which incorporates six features: LDH, CA199, ALT, CEA, TBIL, and AGR. This model exhibits robust predictive performance, achieving an area under the curve of 0.94, a sensitivity of 0.88, and a specificity of 0.93. To enhance its practical utility, the model has been integrated into an accessible web application. This study developed a novel risk prediction model by integrating ML algorithms with conventional laboratory test data to evaluate the likelihood of CRLM occurrence. The model demonstrates excellent predictive performance and has significant clinical application potential.

Deep Learning Meets Early Diagnosis: A Hybrid CNN-DNN Framework for Lung Cancer Prediction and Clinical Translation

Early detection of lung cancer significantly improves patient survival yet remains a challenge due to the subtle nature of early-stage radiological features. This study proposes a multimodal deep learning framework that combines convolutional neural networks (CNN) with dense neural networks (DNN) to enhance early-stage lung cancer prediction. A curated dataset comprising 2,000 patient records with CT scans and clinical metadata was preprocessed and used to train multiple models. The hybrid CNN-DNN model achieved the highest performance with an accuracy of 96.4%, precision of 95.8%, recall of 97.1%, F1-score of 96.4%, and AUC of 0.982, outperforming both traditional machine learning models and standalone CNNs. The integration of imaging and clinical features led to robust and accurate predictions, demonstrating strong potential for real-world clinical application. The results support the deployment of such AI-driven tools in diagnostic workflows to facilitate timely and accurate lung cancer detection.

Chemopreventive and Anticancer Activity of Selected Triterpenoids in Melanoma

Melanoma is one of the most aggressive forms of skin cancer, characterized by high metastatic potential and resistance to conventional therapies. Natural compounds, particularly terpenoids, have gained attention for their chemopreventive potential and anticancer properties. These plant-derived compounds exhibit diverse biological activities, e.g., cell viability and proliferation inhibition, apoptosis induction, cell cycle regulation, and immune system modulation. The review evaluates the current state of the art on the chemopreventive and anticancer activity of lupane- (betulinic acid), oleanane- (oleanolic acid, β-amyrin, escin, hederagenin, glycyrrhetinic acid), and ursane-type (ursolic acid, asiatic acid, madecassic acid, α-amyrin) triterpenoids in melanoma, highlighting their mechanisms of action, therapeutic potential, and challenges in clinical application.

Rapid Diagnostic Testing for Infective Endocarditis and Myocarditis Using Nanopore Targeted Sequencing

ABSTRACTBackgroundInfective endocarditis (IE) and myocarditis are serious heart diseases that can lead to life‐threatening complications. These illnesses can have infectious origins, including viral, bacterial, or fungal pathogens. Traditional detection methods, such as culture‐based methods, have limited ability to detect causative pathogens because of antibiotic use and the difficulty in cultivating intracellular and fastidious bacteria as well as viruses. In clinical settings, rapid diagnostics for pathogen identification are essential for timely treatment and appropriate antimicrobial therapy.MethodsWe successfully developed a method based on nanopore targeted sequencing (NTS) with pathogen‐specific panels for testing myocarditis and IE. As part of this pilot study, a sample‐to‐results protocol was developed with an optimized in‐house pipeline and bioinformatics analysis solution.ResultsThe performance of NTS met our expectations for sensitivity, specificity, and turnaround time. The pathogen‐specific panel testing was accomplished in a 10‐h turnaround time, achieving a detection limit of 20 copies/test for the IE target panel and 10 copies/test for the myocarditis target panel. NTS achieved a clinical performance of 85.0% sensitivity and 96.3% specificity compared with culture testing methods, using 74 clinical specimens from patients (53 male, 21 female) associated with IE.ConclusionsThe rapid turnaround time of NTS is advantageous for managing acute infections, such as IE and myocarditis. NTS is a powerful tool for rapidly diagnosing infections in IE and myocarditis with significant potential for broader clinical applications.

Engineering Escherichia coli Nissle 1917 Carrying PD1 Agonists Resolves Intestinal Inflammation via Local Immune Modulation.

Immunological imbalance is a key factor in the progression of intestinal inflammation, yet effective treatments remain elusive. Using a radiation-induced intestinal injury model, we investigated the causes of inflammation at the single-cell level and identified abnormal T-cell activation as a major contributor. To address this, we targeted the PD1 signaling pathway to suppress T-cell activation and evaluated the anti-inflammatory and intestinal repair effects of the PD1 agonist displaying probiotic Escherichia coli Nissle 1917 (EcNMP1-M) in two mouse models. Encapsulated in Eudragit L100-55 for pH-dependent release, EcNMP1-M and its bacterial outer membrane vesicles (OMVs) expressed PD1 agonists, which inhibited excessive immune activation and reduced inflammatory cytokines. EcNMP1-M promoted the expression of proteins that maintain intestinal epithelial barrier integrity, improving gut function and immune responses in colitis mice. Furthermore, 16S rDNA microbiome sequencing revealed that EcNMP1-M enhanced intestinal microbiota diversity, increased beneficial bacteria, and reduced harmful bacteria. This study proposes a localized EcN-based immunosuppressive therapy for radiation-induced enteritis and inflammatory bowel disease with promising potential for clinical applications.

Identification of programmed cell death-related genes and construction of a prognostic model in oral squamous cell carcinoma using single-cell and transcriptome analysis

BackgroundOral squamous cell carcinoma (OSCC) is characterized by poor prognosis and high mortality. Understanding programmed cell death-related genes could provide valuable insights into disease progression and treatment strategies.MethodsRNA-sequencing data from 341 OSCC tumor tissues and 31 healthy samples were analyzed from TCGA database, with validation using 76 samples from GSE41613. Single-cell RNA sequencing data was obtained from GSE172577 (6 OSCC samples). Differentially expressed genes (DEGs) were identified and intersected with 1,254 programmed cell death-related genes. A protein–protein interaction network was constructed, and key modules were identified. Univariate Cox, LASSO, and multivariate Cox regression analyses were performed to build a prognostic model. Model performance was evaluated using Kaplan–Meier analysis, ROC curves, and nomogram validation.ResultsThe study identified 200 candidate genes from the intersection of DEGs and programmed cell death-related genes, which were further refined to 57 hub genes through PPI network analysis. A prognostic signature consisting of five genes (MET, GSDMB, KIT, PRKAG3, and CDKN2A) was established and validated. The model demonstrated good predictive performance in both training and validation cohorts (AUC > 0.6 for 1-, 2-, and 3-year survival). Single-cell analysis revealed that prognostic genes were predominantly expressed in stromal and epithelial cells. Cell communication analysis indicated strong interactions between stromal and epithelial cells.ConclusionsThis study developed and validated a novel five-gene prognostic signature for OSCC based on programmed cell death-related genes. The model shows promising clinical application potential for risk stratification and personalized treatment of OSCC patients.

Preoperative total bile acid can be used as a prognostic biomarker in patients with operable biliary tract cancers

BackgroundBiliary tract cancers (BTCs) are highly invasive malignancies with poor prognoses. However, reliable biomarkers for survival prediction remain lacking. Notably, abnormal lipid metabolism has elicited increasing interest in digestive tract tumors, with the liver playing an important role in lipid metabolism.ObjectiveTo explore the relationship between hepatic lipid metabolism-related indicators, assessed through routine clinical biochemical testing and survival prognosis in patients with BTCs.MethodsOverall, 109 patients with a pathological diagnosis of BTC from 2017 to 2023 were included in this study. Univariate and multivariate Cox regression analyses were performed using R Studio software, and survival curves were plotted.ResultsUnivariate analysis revealed that tumor location and preoperative total bile acid (TBA), carcinoembryonic antigen, cancer antigen (CA)125, and CA19-9 levels were correlated with patient survival (P < 0.05). Multivariate Cox regression analysis identified increased TBA level [hazard ratio (HR) = 0.445, P = 0.004] as an independent prognostic factor for longer survival. Conversely, tumor location [intrahepatic cholangiocarcinoma (iCCA) and/or extrahepatic cholangiocarcinoma (eCCA)] (HR = 2.463, P = 0.036) and increased CA125 and CA19-9 levels (HR = 2.549, P = 0.008 and HR = 2.100, P = 0.019) were independent prognostic factors for shorter survival. Additionally, Kaplan‒Meier survival curves revealed significantly longer survival in patients with increased TBA levels than those in the normal group (P = 0.012). Conversely, patients with iCCA and/or eCCA tumor location and increased CA125 and CA19-9 levels had significantly shorter median survival (P = 0.044, P = 0.013, and P = 0.012, respectively).ConclusionTBA may be a biomarker for predicting survival in patients with operable BTC, highlighting its clinical significance and application potential.

Photodynamic therapy could serve as a promising approach to prevent posterior capsular opacification.

Posterior capsular opacification (PCO) causes the vision that has been restored after cataract surgery to become blurred again. YAG laser treatment for PCO not only incurs additional costs but also poses risks of complications, including glaucoma and retinal disorders. Effective prevention and management of PCO remain an area requiring active research. Photodynamic therapy (PDT) utilizes a photosensitizer (PS) to transform oxygen into reactive oxygen species (ROS) under specific wavelengths of light, thereby inducing apoptosis. Given its minimal invasiveness and high specificity, PDT has been extensively applied in the treatment of conditions characterized by abnormal cell proliferation, such as tumors. Considering the pathogenesis of PCO, PDT has demonstrated promising clinical application potential in ophthalmic disease treatment. This review examines the impact of photodynamic therapy on the biological behavior of lens epithelial cells (LECs) and its efficacy in treating PCO. It also discusses the advantages and disadvantages of different photosensitizers and their clinical application potential.

A QS21 + CpG-Adjuvanted Trivalent HSV-2 Vaccine and Trivalent HSV-2 mRNA Vaccine Induce a Strong Immune Response, Protect Against HSV-2 Infection, and Cross-Protect Against HSV-1 Infection in Mice

Background: HSV-2 infection continues to be a significant global health concern, as there are no approved vaccines despite numerous attempts at development. Methods: This study explored the immunogenicity and protective efficacy of aluminum- or QS21 + CpG-adjuvanted trivalent HSV-2 vaccines and a trivalent HSV-2 mRNA vaccine incorporating the gC2, gD2, and gE2 antigens. Results: Our results demonstrated that the QS21 + CpG-adjuvanted subunit vaccine and mRNA vaccines successfully induced robust antigen-specific humoral and cellular immune responses and provided significant protection against both HSV-2 and HSV-1 infection. These vaccines showed remarkable efficiency in reducing the viral load and preventing clinical symptoms in mice, highlighting their potential for clinical application. Conversely, the aluminum-adjuvanted vaccine exhibited limited effectiveness, emphasizing the superiority of the QS21 + CpG-adjuvanted and mRNA vaccines. Conclusions: These findings provide valuable insights for the continued development of effective HSV vaccines and suggest promising strategies for preventing both HSV-2 and HSV-1 infection.

The value of high-pitch scanning with Sn100kV and ADMIRE in CT examination of tuberculous destroyed lung: Identifying the optimal combination for ultra-low-dose imaging.

To investigate the application value of high-pitch scanning combined with energy spectrum purification using Sn100kV and ADMIRE in CT examinations of patients with tuberculous destroyed lung. A total of 60 patients with sputum mycobacterium tuberculosis smear positive and diagnosed with tuberculous lung damage on imaging were prospectively collected. The first CT examination utilized a conventional scanning mode with a fixed tube voltage of 120kV, CARE Dose4D activated, reference tube current set at 70mAs, and a pitch of 1.5. The interval between the initial and follow-up CT was over three months. During the follow-up CT, a high-pitch scanning mode combined with energy spectrum purification was employed, with a fixed tube voltage of Sn100kV, CARE Dose4D activated, reference tube current set at 300mAs, and a pitch of 3.2. The remaining parameters were consistent between the two CT scans. The first CT was recorded as Group A, and the follow-up CT as Group B. After the examinations, the reconstructed layer thickness was 1.00mm, and lung window and mediastinal window images with a layer spacing of 0.7mm were obtained. The image quality of the two CT examinations was scored by three physicians using a 5-point scale. Following the scoring, the consistency of the three physicians' scores was analyzed using the intraclass correlation coefficient.. A chief physician reviewed the lung window images from both CT scans, recorded the number of focal signs detected, and conducted Chi-square tests to compare these counts between the two groups. The CT values and noise levels in both the mediastinal window and pulmonary window were measured, SNR and CNR were calculated, and independent sample T-test was performed to analyze the differences in these parameters between the two groups. Motion artifacts in the two CT images were assessed and scored by three physicians using a 3-point values, and Mann-Whitney U test was applied to compare the scores between the groups. The radiation doses of two CT examinations was recorded, and the differences between the two groups were statistically analyzed using the Mann-Whitney U test. Data analysis was conducted using SPSS 26.0 software. The image quality scores of both groups were 3 points or above, meeting the requirements for clinical diagnosis. The intraclass correlation coefficient (ICC) value for the consistency analysis of the pulmonary window scores among the three physicians was 0.819 (p < 0.001), and for the mediastinal window consistency analysis, the ICC value was 0.795 (p < 0.001), indicating good consistency in the subjective score diagnosis results. The detection rate of lesion signs in Group B was higher than that in Group A, but there was no statistical difference between the two groups (p > 0.05). There was no significant difference in noise, SNR, and CNR between the two groups (p > 0.05). However, the SNR and CNR in Group B were better than those in Group A. There was no statistical difference in the CT values of the aorta and muscle between the two groups of mediastinal window images, while noise, SNR, and CNR were statistically significant (p < 0.05). Noise in Group B was higher than that in Group A, while the SNR and CNR in Group B were lower than those in Group A. The motion artifacts of the two groups were significantly different (p < 0.001), with Group A having significantly more motion artifacts than Group B. The radiation dose of the two groups was statistically different (p < 0.001), with Group B's radiation dose reduced by 76.24% compared to Group A. The combination of high-pitch scanning with Sn100kV and ADMIRE can be effectively used for ultra-low-dose CT examination of the tuberculous destroyed lung, obtaining satisfactory diagnostic images and reducing the occurrence of motion artifacts. This technique achieves conventional diagnostic outcomes at ultra-low doses and significantly reduces motion artifacts, holding significant potential and value for widespread clinical application in CT examinations for patients with tuberculous disfigured lung.

3D printed Gel/PTH@PAHA scaffolds with both enhanced osteogenesis and mechanical properties for repair of large bone defects

Abstract The repair of large bone defects continues to pose a significant challenge in clinical orthopedics. Successful repairs require not only adequate mechanical strength but also exceptional osteogenic activity for successful clinical translation. Composite materials based on polyamide 66 (PA66) and hydroxyapatite have been widely used in various clinical settings. However, existing PA66/hydroxyapatite composites often lack sufficient osteogenic stimulation despite their favorable mechanical properties, which limit their overall clinical efficacy. In this study, we fabricated a polyamide 66/nano-hydroxyapatite (PAHA) scaffold using an extruder and fused deposition modeling-based 3D printing technology. Subsequently, gelatin methacrylamide (GelMA) containing teriparatide (PTH) was incorporated into the PAHA scaffold to construct the Gel/PTH@PAHA scaffold. Material characterization results indicated that the compressive modulus of elasticity and compressive strength of the Gel/PTH@PAHA scaffold were 172.47 ± 5.48 MPa and 25.55 ± 2.19 MPa, respectively. In vitro evaluations demonstrated that the Gel/PTH@PAHA scaffold significantly enhanced osteoblast adhesion and proliferation while promoting osteogenic differentiation of BMSCs. In vivo studies further revealed that this scaffold notably promoted new bone regeneration in rabbit femoral defects. These findings suggest that the 3D-printed Gel/PTH@PAHA scaffold exhibits excellent mechanical properties alongside remarkable osteogenic activity, thereby meeting the dual requirements for load-bearing applications and bone regeneration. This innovative approach may be a promising candidate for customized orthopedic implants with substantial potential for clinical application.

Deep Learning on Misaligned Dual-Energy Chest X-ray Images Using Paired Cycle-Consistent Generative Adversarial Networks.

Dual-energy subtraction (DES) chest X-ray images (CXRs) are often affected by motion artifacts resulting from patients' voluntary or involuntary movements, even in clinical settings. Additionally, the mediastinum and upper abdominal regions in low-energy (LE) CXRs are susceptible to signal insufficiency due to inadequate input photon numbers. Current image processing techniques for removing motion artifacts and statistical noise from DES-CXRs are insufficient, and potential algorithms for these tasks remain largely unexplored. We propose a framework based on paired cycle-consistency adversarial generative networks to effectively remove motion artifacts and statistical noise from DES-CXRs. The proposed method incorporates ensemble discriminators, differentiable augmentation, anti-aliased convolution layers, and a basic 8-layer U-Net generator. This method was trained and tested using a clinical image dataset comprising data of 600 examinations of individuals who underwent dual-energy chest X-ray imaging for diagnostic purposes, using a sixfold cross-validation approach. It demonstrated a remarkable improvement in motion artifact suppression in terms of an analysis of full width at the 10-percent maximum improved from 0.216 ± 0.0720 to 0.200 ± 0.0783 for the left lung region of interests including the cardiac region. Furthermore, it outperformed the method in a previous study in terms of a peak signal-to-noise ratio of 50.7 ± 3.68, structural similarity index of 0.997 ± 0.0152 for LE images, and Fréchet inception distance of 85.0 ± 3.52 for bone-suppressed DES images. The proposed method significantly outperforms existing techniques for removing motion artifacts and statistical noise and shows strong potential for clinical applications in chest X-ray imaging.

An NAG-Activatable Glow-Type Chemiluminescent Probe for Sensitive and Early Diagnosis of Acute Kidney Injury through Urinalysis.

The current clinical diagnostic criteria for acute kidney injury (AKI) are based on the evaluation of changes in serum biomarkers (sCr, BUN) and urine output, which are insensitive and typically do not show significant differences until the middle to late stages, thereby hindering the time diagnosis of AKI. Herein, we report an N-acetyl-β-d-glucosaminidase (NAG)-activatable glow-type chemiluminescent probe (CL-NAG) for sensitive detection of urinary NAG and early diagnosis of AKI. CL-NAG was designed and synthesized by incorporating N-acetyl-β-d-glucosamine into the phenoxy-dioxetane scaffolds through a self-immolative linker. Molecular docking calculations predicted a strong interaction between CL-NAG and the catalytic cavity of the NAG enzyme. CL-NAG exhibited high detection sensitivity (LOD = 0.042 U/L) for NAG activity, low autofluorescence interference, and an extremely high signal-to-noise ratio (1628-fold). Furthermore, using this probe, we achieved not only the imaging detection of NAG in HUVEC and HK-2 cells but also the quantitative detection of NAG activity in the urine of an AKI mouse model at different time points. Most importantly, CL-NAG successfully identified 19 positive and 149 negative samples out of 168 clinical urine specimens, proving a high consistency rate of 95.24% compared with clinical biochemical methods. This research demonstrates the sensitive urinary NAG detection ability of CL-NAG and suggests the enormous clinical application potential of CL-NAG in the early diagnosis of AKI.