Active Learning Guided Discovery of High Entropy Oxides Featuring High H2-production.

High entropy oxides (HEOs) represent a class of solid solutions comprising multiple elements, offering significant scientific potential. Due to the enormous combination types of elements, the design of HEOs with desirable properties within high-dimensional composition spaces has traditionally relied heavily on knowledge and intuition. In this study, we present an active learning (AL) strategy tailored to efficiently explore the vast compositional space of HEOs. Our approach operates as a closed-loop system, iteratively cycling through "Training, Prediction, and Experiment" stages. Across multiple AL iterations, we have successfully identified four novel HEOs from a vast array of potential compositions. These newly discovered materials exhibit exceptional stability and demonstrate outstanding performance in H2 evolution rate (251 µmol gcat-1 min-1) during the water-gas shift reaction, surpassing benchmarks set by established catalysts such as Pt/γ-Al2O3 (135 µmol gcat-1 min-1) and Cu/ZnO/Al2O3 (81 µmol gcat-1 min-1). X-ray photoelectron spectroscopy and density functional theory calculations revealed a loss of elemental identity in the selected HEOs. This catalyst discovery process underscores the efficacy of Machine Learning in accelerating the identification of HEOs with unique characteristics by effectively leveraging insights from limited experimental data.

Predictive Value of a New Muscle Parameter in Patients with Resectable Gastric Cancer: A Pooled Analysis of Three Prospective Trials.

BACKGROUND: Sarcopenia is closely associated with gastric cancer (GC) prognosis. However, its exact definition remains controversial. METHODS: This study included computed tomography images and clinical data of patients from three prospective studies. The skeletal muscle index (SMI) and skeletal muscle radiation attenuation (SMRA) were analyzed, and a new muscle parameter, skeletal muscle gauge (SMG), was obtained by multiplying the two parameters. The values of the three indices for predicting the prognosis of patients with GC were compared. RESULTS: The study included 717 patients. The findings showed median values of 42 cm2/m2 (range, 36.8-48.2 cm2/m2) for SMI, 45 HU (range, 41-49 HU) for SMRA, and 1842 (range, 1454-2260) for SMG. Postoperatively, 111 patients (15.5%) experienced complications. The 3-year overall survival (OS), disease-free survival (DFS), and recurrence-free survival (RFS) were 74.3%, 68.2%, and 70%, respectively. Univariate logistic analysis showed that postoperative complications were associated with SMI (odds ratio [OR] 0.94; 95% confidence interval [CI] 0.92-0.96), SMRA (OR, 0.87; 95% CI 0.84-0.90), and SMG (OR 0.99; 95% CI 0.98-0.99). After a two-step multivariate analysis, only SMG (OR 0.98, 95% CI 0.97-0.99) was an independent protective factor of postoperative complications. Multivariate analysis showed that SMG also was an independent protective factor of OS, DFS, and RFS. The patients were divided into low-SMG (L-SMG) group and high-SMG (H-SMG) groups. Chemotherapy benefit analysis of the patients with stage II low SMG showed that the OS, DFS, and RFS of the chemotherapy group were significantly better than those of the non-chemotherapy group (p < 0.05). CONCLUSION: The prospective large sample data showed that the new muscle parameter, SMG, can effectively predict the short-term outcome and long-term prognosis of patients with resectable gastric cancer. As a new muscle parameter index, SMG is worthy of further study.

Cytotoxicity effect and transcriptome analysis of PCV3-infected cells revealed potential viral pathogenic mechanisms.

Porcine circovirus type 3 (PCV3) has become an important pathogen in the global swine industry and poses a threat to pig health, but its pathogenic mechanism remains unknown. In this study, we constructed an innovative, linear infectious clone of PCV3 for rescuing the virus, and explored the transcriptome of infected cells to gain insights into its pathogenic mechanisms. Subsequently, an in vivo experiment was conducted to evaluate the pathogenicity of the rescued virus in pig. PCV3 nucleic acid was distributed across various organs, indicating systemic circulation via the bloodstream and viremia. Immunohistochemical staining also revealed a significant presence of PCV3 antigens in the spleen, lungs, and lymph nodes, indicating that PCV3 had tropism for these organs. Transcriptome analysis of infected ST cells revealed differential expression of genes associated with apoptosis, immune responses, and cellular metabolism. Notably, upregulation of genes related to the hypoxia-inducible factor-1 pathway, glycolysis, and the AGE/RAGE pathway suggests activation of inflammatory responses, ultimately leading to onset of disease. These findings have expanded our understanding of PCV3 pathogenesis, and the interplay between PCV3 and host factors.

Management and Mechanisms of Diarrhea Induced by Tyrosine Kinase Inhibitors in Human Epidermal Growth Factor Receptor-2-Positive Breast Cancer.

Breast cancer has the highest incidence among female malignancies, significantly impacting women's health. Recently, numerous HER2-targeted therapies have achieved excellent clinical outcomes. Currently, anti-HER2 drugs are divided into three main categories: monoclonal antibodies, small-molecule tyrosine kinase inhibitors, and antibody-coupled drugs (ADCs). The main toxic side effects of small molecule TKI-based therapy are diarrhea, hand-foot syndrome, rash, nausea, and vomiting. Diarrhea is a potential predictor of tumor response, affecting up to 95% of cancer patients treated with TKIs. Severe gastrointestinal toxicity can result in the need for dose reductions and treatment interruptions. This not only compromises the efficacy of TKIs but also deteriorates human nutrition and quality of life. The majority of individuals develop diarrhea within 7 days of starting treatment, with approximately 30% developing grade 3 or higher diarrhea within 2-3 days of starting treatment. The severity of diarrhea typically correlates with the dosage of most TKIs. Current prevention and management strategies are primarily empirical, focusing on symptom alleviation rather than addressing the toxicological mechanisms underlying TKI-induced diarrhea. Consequently, anti-diarrheal drugs are often less effective in managing this condition in cancer patients receiving TKIs. Moreover, our understanding of the toxicological mechanisms responsible for such diarrhea remains limited, underscoring the urgent need to identify these mechanisms in order to develop effective anti-diarrheal medications tailored to this specific context. This review aims to elucidate management approaches and mechanisms for diarrhea induced by TKIs during HER2-positive breast cance.

Breast cancer has the highest incidence among female malignancies, significantly impacting women's health. In recent years, numerous HER2-targeted therapies have been developed, including lapatinib, neratinib, tucatinib, and pyrotinib. However, second-generation tyrosine kinase inhibitors (TKIs) often cause severe gastrointestinal toxicity, necessitating dose reduction and treatment interruption. This not only compromises the efficacy of TKIs but also deteriorates nutritional status and quality of life. Diarrhea is a potential predictor of tumor response and is the second most common adverse event, affecting up to 95% of cancer patients treated with TKIs. Approximately 30% of these patients experience grade 3 or higher diarrhea within just 2-3 days of initiating TKI treatment. The severity of diarrhea typically correlates with the dosage of most TKIs. Current prevention and management strategies are primarily empirical, focusing on symptom alleviation rather than addressing the toxicological mechanisms underlying TKI-induced diarrhea. Consequently, anti-diarrheal drugs are often less effective in managing this condition in cancer patients receiving TKIs. Moreover, our understanding of the toxicological mechanisms responsible for such diarrhea remains limited, underscoring the urgent need to identify these mechanisms to develop effective antidiarrheal medications tailored for this specific context. This review aims to elucidate management approaches for diarrhea induced by HER2 tyrosine kinase inhibitors during breast cancer treatment.

Improving Prediction of Complications Post-Proton Therapy in Lung Cancer Using Large Language Models and Meta-Analysis.

PURPOSE: This study enhances the efficiency of predicting complications in lung cancer patients receiving proton therapy by utilizing large language models (LLMs) and meta-analytical techniques for literature quality assessment. MATERIALS AND METHODS: We integrated systematic reviews with LLM evaluations, sourcing studies from Web of Science, PubMed, and Scopus, managed via EndNote X20. Inclusion and exclusion criteria ensured literature relevance. Techniques included meta-analysis, heterogeneity assessment using Cochran's Q test and I2 statistics, and subgroup analyses for different complications. Quality and bias risk were assessed using the PROBAST tool and further analyzed with models such as ChatGPT-4, Llama2-13b, and Llama3-8b. Evaluation metrics included AUC, accuracy, precision, recall, F1 score, and time efficiency (WPM). RESULTS: The meta-analysis revealed an overall effect size of 0.78 for model predictions, with high heterogeneity observed (I2 = 72.88%, P < 0.001). Subgroup analysis for radiation-induced esophagitis and pneumonitis revealed predictive effect sizes of 0.79 and 0.77, respectively, with a heterogeneity index (I2) of 0%, indicating that there were no significant differences among the models in predicting these specific complications. A literature assessment using LLMs demonstrated that ChatGPT-4 achieved the highest accuracy at 90%, significantly outperforming the Llama3 and Llama2 models, which had accuracies ranging from 44% to 62%. Additionally, LLM evaluations were conducted 3229 times faster than manual assessments were, markedly enhancing both efficiency and accuracy. The risk assessment results identified nine studies as high risk, three as low risk, and one as unknown, confirming the robustness of the ChatGPT-4 across various evaluation metrics. CONCLUSION: This study demonstrated that the integration of large language models with meta-analysis techniques can significantly increase the efficiency of literature evaluations and reduce the time required for assessments, confirming that there are no significant differences among models in predicting post proton therapy complications in lung cancer patients.

Using Advanced AI to Improve Predictions of Treatment Side Effects in Lung Cancer: This research uses cutting-edge artificial intelligence (AI) techniques, including large language models like ChatGPT-4, to better predict potential side effects in lung cancer patients undergoing proton therapy. By analyzing extensive scientific literature quickly and accurately, this approach has proven to enhance the evaluation process, making it faster and more reliable in foreseeing complications from treatments.

Long-term oncological outcomes of 3D versus 2D laparoscopic gastrectomy for gastric cancer: a randomized clinical trial.

BACKGROUND: Laparoscopy-assisted gastrectomy (LG) is rapidly gaining popularity owing to its minimal invasiveness. Previous studies have found that compared with two-dimensional (2D)-LG, three-dimensional (3D)-LG showed better short-term outcomes. However, the long-term oncological outcomes in patients with locally resectable gastric cancer (GC) remain controversial. METHODS: In this noninferiority, open-label, randomized clinical trial, a total of 438 eligible GC participants were randomly assigned in a 1:1 ratio to either 3D-LG or 2D-LG from January 2015 to April 2016. The primary endpoint was operating time, while the secondary endpoints included 5-year overall survival (OS), disease-free survival (DFS), and recurrence pattern. RESULTS: Data from 401 participants were included in the per-protocol analysis, with 204 patients in the 3D group and 197 patients in the 2D group. The 5-year OS and DFS rates were comparable between the 3D and 2D groups (5-year OS: 70.6% vs. 71.1%, Log-rank P = 0.743; 5-year DFS: 68.1% vs. 69.0%, log-rank P = 0.712). No significant differences were observed between the 3D and 2D groups in the 5-year recurrence rate (28.9% vs. 28.9%, P = 0.958) or recurrence time (mean time, 22.6 vs. 20.5 months, P = 0.412). Further stratified analysis based on the type of gastrectomy, postoperative pathological staging, and preoperative BMI showed that the 5-year OS, DFS, and recurrence rates of the 3D group in each subgroup were similar to those of the 2D group (all P > 0.05). CONCLUSIONS: For patients with locally resectable GC, 3D-LG performed by experienced surgeons in high-volume professional institutions can achieve long-term oncological outcomes comparable to those of 2D-LG. REGISTRATION NUMBER: NCT02327481 ( http://clinicaltrials.gov ).

Effect of sarcopenia on short-term and long-term outcomes of older patients with locally advanced gastric cancer: a multicenter study.

OBJECTIVE: To assess the effect of preoperative sarcopenia on the short-term and long-term outcomes in older patients with locally advanced gastric cancer (LAGC). METHODS: Clinicopathological data of older patients with LAGC who underwent radical surgery were retrospectively analyzed. Sarcopenia was defined as a skeletal muscle index of less than 36.4 cm2/m2 for men and less than 28.4 cm2/m2 for women. Comparing the postoperative complications and survival between sarcopenia and non-sarcopenia groups using multicenter data. RESULTS: A total of 406 older patients with LAGC were included in the analysis, including 145 (35.7%) with sarcopenia and 261 (64.3%) with non-sarcopenia. Multivariate logistic regression analysis showed that sarcopenia was an independent risk factor for postoperative complications with CD grade ≥ II (OR 1.616; P < 0.05). Kaplan-Meier survival curve analysis showed that the 5-year overall survival (OS) and 5-year recurrence-free survival (RFS) in the sarcopenia group were lower than those in the non-sarcopenia group (P both < 0.05). Multivariate Cox regression analyses showed that sarcopenia was an independent prognostic factor for 5-year OS and RFS (P both < 0.05). The 5-year recurrence rate in the sarcopenia group was 57.2%, which was significantly higher than that in the non-sarcopenia group (46.4%; P = 0.036). Recurrence pattern analysis showed that the incidence of distant metastases in patients with sarcopenia (42.8%) was significantly higher than non-sarcopenia (31.4%; P = 0.022). CONCLUSION: Sarcopenia serves as a valuable predictor of both short-term and long-term outcomes in older patients with LAGC. Therefore, the significance of assessing preoperative nutritional status and implementing thorough postoperative follow-up for older LAGC patients with sarcopenia should be emphasized.

Robot-assisted versus laparoscopic-assisted gastrectomy among malnourished patients with gastric cancer based on textbook outcome.

BACKGROUND: Textbook outcome (TO) has been widely employed as a comprehensive indicator to assess the short-term prognosis of patients with cancer. Preoperative malnutrition is a potential risk factor for adverse surgical outcomes in patients with gastric cancer (GC). This study aimed to compare the TO between robotic-assisted gastrectomy (RAG) and laparoscopic-assisted gastrectomy (LAG) in malnourished patients with GC. METHODS: According to the diagnostic consensus of malnutrition proposed by Global Leadership Initiative on Malnutrition (GLIM) and Nutrition Risk Index (NRI), 895 malnourished patients with GC who underwent RAG (n = 115) or LAG (n = 780) at a tertiary referral hospital between January 2016 and May 2021 were included in the propensity score matching (PSM, 1:2) analysis. RESULTS: After PSM, no significant differences in clinicopathological characteristics were observed between the RAG (n = 97) and LAG (n = 194) groups. The RAG group had significantly higher operative time and lymph nodes harvested, as well as significantly lower blood loss and hospital stay time compared to the LAG group. More patients in the RAG achieved TO. Logistic regression analysis revealed that RAG was an independent protective factor for achieving TO. There were more adjuvant chemotherapy (AC) cycles in the RAG group than in the LAG group. After one year of surgery, a higher percentage of patients (36.7% vs. 22.8%; P < 0.05) in the RAG group recovered from malnutrition compared to the LAG group. CONCLUSIONS: For malnourished patients with GC, RAG performed by experienced surgeons can achieved a higher rate of TO than those of LAG, which directly contributed to better AC compliance and a faster restoration of nutritional status.

Global incidence and mortality trends of gastric cancer and predicted mortality of gastric cancer by 2035.

OBJECTIVE: To study the historical global incidence and mortality trends of gastric cancer and predicted mortality of gastric cancer by 2035. METHODS: Incidence data were retrieved from the Cancer Incidence in Five Continents (CI5) volumes I-XI, and mortality data were obtained from the latest update of the World Health Organization (WHO) mortality database. We used join-point regression analysis to examine historical incidence and mortality trends and used the package NORDPRED in R to predict the number of deaths and mortality rates by 2035 by country and sex. RESULTS: More than 1,089,000 new cases of gastric cancer and 769,000 related deaths were reported in 2020. The average annual percent change (AAPC) in the incidence of gastric cancer from 2003 to 2012 among the male population, South Korea, Japan, Malta, Canada, Cyprus, and Switzerland showed an increasing trend (P > 0.05); among the female population, Canada [AAPC, 1.2; (95%Cl, 0.5-2), P < 0.05] showed an increasing trend; and South Korea, Ecuador, Thailand, and Cyprus showed an increasing trend (P > 0.05). AAPC in the mortality of gastric cancer from 2006 to 2015 among the male population, Thailand [3.5 (95%cl, 1.6-5.4), P < 0.05] showed an increasing trend; Malta Island, New Zealand, Turkey, Switzerland, and Cyprus had an increasing trend (P > 0.05); among the male population aged 20-44, Thailand [AAPC, 3.4; (95%cl, 1.3-5.4), P < 0.05] showed an increasing trend; Norway, New Zealand, The Netherlands, Slovakia, France, Colombia, Lithuania, and the USA showed an increasing trend (P > 0.05). It is predicted that the mortality rate in Slovenia and France's female population will show an increasing trend by 2035. It is predicted that the absolute number of deaths in the Israeli male population and in Chile, France, and Canada female population will increase by 2035. CONCLUSION: In the past decade, the incidence and mortality of gastric cancer have shown a decreasing trend; however, there are still some countries showing an increasing trend, especially among populations younger than 45 years. Although mortality in most countries is predicted to decline by 2035, the absolute number of deaths due to gastric cancer may further increase due to population growth.

Comparative outcomes of obese and non-obese patients with lumbar disc herniation receiving full endoscopic transforaminal discectomy: a systematic review and meta-analysis.

OBJECTIVE: This study aimed to assess the impact of full endoscopic transforaminal discectomy (FETD) on clinical outcomes and complications in both obese and non-obese patients presenting with lumbar disc herniation (LDH). METHODS: A systematic search of relevant literature was conducted across various primary databases until November 18, 2023. Operative time and hospitalization were evaluated. Clinical outcomes included preoperative and postoperative assessments of the Oswestry Disability Index (ODI) and visual analogue scale (VAS) scores, conducted to delineate improvements at 3 months postoperatively and during the final follow-up, respectively. Complications were also documented. RESULTS: Four retrospective studies meeting inclusion criteria provided a collective cohort of 258 patients. Obese patients undergoing FETD experienced significantly longer operative times compared to non-obese counterparts (P = 0.0003). Conversely, no statistically significant differences (P > 0.05) were observed in hospitalization duration, improvement of VAS for back and leg pain scores at 3 months postoperatively and final follow-up, improvement of ODI at 3 months postoperatively and final follow-up. Furthermore, the overall rate of postoperative complications was higher in the obese group (P = 0.02). The obese group demonstrated a total incidence of complications of 17.17%, notably higher than the lower rate of 9.43% observed in the non-obese group. CONCLUSION: The utilization of FETD for managing LDH in individuals with obesity is associated with prolonged operative times and a higher total complication rate compared to their non-obese counterparts. Nevertheless, it remains a safe and effective surgical intervention for treating herniated lumbar discs in the context of obesity.

Roles of DJ41_1407 and DJ41_1408 in Acinetobacter baumannii ATCC19606 Virulence and Antibiotic Response.

Acinetobacter baumannii is a major cause of nosocomial infections, and its highly adaptive nature and broad range of antibiotic resistance enable it to persist in hospital environments. A. baumannii often employs two-component systems (TCSs) to regulate adaptive responses and virulence-related traits. This study describes a previously uncharacterized TCS in the A. baumannii ATCC19606 strain, consisting of a transcriptional sensor, DJ41_1407, and its regulator, DJ41_1408, located adjacent to GacA of the GacSA TCS. Markerless mutagenesis was performed to construct DJ41_1407 and DJ41_1408 single and double mutants. DJ41_1408 was found to upregulate 49 genes and downregulate 43 genes, most of which were associated with carbon metabolism and other metabolic pathways, such as benzoate degradation. MEME analysis revealed a putative binding box for DJ41_1408, 5'TGTAAATRATTAYCAWTWAT3'. Colony size, motility, biofilm-forming ability, virulence, and antibiotic resistance of DJ41_1407 and DJ41_1408 single and double mutant strains were assessed against wild type. DJ41_1407 was found to enhance motility, while DJ41_1408 was found to upregulate biofilm-forming ability, and may also modulate antibiotic response. Both DJ41_1407 and DJ41_1408 suppressed virulence, based on results from a G. mellonella infection assay. These results showcase a novel A. baumannii TCS involved in metabolism, with effects on motility, biofilm-forming ability, virulence, and antibiotic response.

Patient-Reported Outcomes of Individuals with Gastric Cancer Undergoing Totally Laparoscopic Versus Laparoscopic-Assisted Total Gastrectomy: A Real-World, Propensity Score-Matching Analysis.

OBJECTIVE: Totally laparoscopic total gastrectomy (TLTG) and laparoscopic-assisted total gastrectomy (LATG) are two types of minimally invasive radical gastrectomy procedures to treat gastric cancer (GC). This study compared the long-term prognosis and postoperative health-related quality of life (HRQoL) between TLTG and LATG. METHODS: A total of 106 patients who underwent TLTG and 1,076 patients who underwent LATG at the Union Hospital of Fujian Medical University (Fuzhou, China) between January 2014 and April 2018 were included in the propensity score matching (PSM, 1:2). Patient-reported outcomes at 3, 6, and 12 months after gastrectomy were analyzed. The questionnaire referred to the European Organization for Research and Treatment of Cancer (EORTC) 30-item core QoL (QLQ-C30)and the GC module (QLQ-STO22) questionnaire. RESULTS: After PSM, there were no significant differences in clinicopathological characteristics between the TLTG (n = 104) and the LATG groups (n = 208). Operative time and volume of blood loss were significantly lower in the TLTG group than in the LATG group. Kaplan-Meier survival analysis revealed similar 3-year survival rates between the TLTG and LATG groups (83.7 vs. 80.3%, respectively; P = 0.462). Tolerance to nonliquid diet, decrease in body weight, and albumin levels were also significantly lower in the TLTG group than in the LATG group (all P < 0.05). The HRQoL scale demonstrated that the overall score in the TLTG group was better than that in the LATG group at 3, 6, and 12 months after gastrectomy (all P < 0.05). CONCLUSIONS: Patients with GC undergoing TLTG reported better HRQoL and experienced faster recovery of social function than those undergoing LATG, although the two groups demonstrated similar short-term outcomes and long-term prognosis.

Comparison of 14 respiratory pathogens among hospitalized children during and after the COVID-19 outbreak in Chaoshan area.

BACKGROUND: Since January 2020, measures has been adopted in the Chaoshan area to limit the spread of COVID-19. Restrictions were removed after August 2020. At the same time, children returned to school. We previously reported the changes of 14 main respiratory pathogens in hospitalized children before and during the COVID-19 outbreak in Chaoshan area. However, the changes of respiratory pathogen spectrum in hospitalized children after the epidemic are still unknown, which will be elucidated in this study. METHODS: There are 6201 children hospitalized with respiratory tract infection were enrolled in the study, which were divided into two groups: 2533 from outbreak group (1 January 2020-31 December 2020), and 3668 from post-outbreak group (1 January 2021-31 December 2021). Pharyngeal swab samples were collected. 14 respiratory tract pathogens were detected by liquid chip technology. RESULTS: The positive rate of pathogen detection is significantly lower in the outbreak group (65.42%, 1657/2533) than that in the post-outbreak group (70.39%, 2582/3668; χ2 = 17.15, P < 0.05). The Influenza A virus (FluA) detection rate was 1.9% (49) in 2020, but 0% (0) in 2021. The detection rates of Bordetella pertussis (BP) decreased from 1.4% (35) in 2020 to 0.5% (17) in 2021. In contrast, the detection rates of  Influenza B virus (FluB), Cytomegalovirus (CMV), Haemophilus influenzae (HI), Streptococcus pneumoniae (SP) increased from 0.3% (8), 24.7% (626), 2.0% (50) and 19.4% (491) in 2020 to 3.3% (121), 27.9% (1025), 4.6% (169), 22.8% (836) in 2021, respectively (P < 0.01). CONCLUSIONS: The detection rates of pathogens such as FluA, FluB, CMV, HI, SP, BP were statistically different between 2020 and 2021. From 2020 to 2021, the positive rates of Flu, CMV, HI and SP increased, while the positive rates of FluA and BP decreased. After the COVID-19 prevention and control measures are gradually relaxed, the positive rate of respiratory pathogens in children aged from 6 months to 6 years will increase.

Lower-extremity fatigue fracture detection and grading based on deep learning models of radiographs.

OBJECTIVES: To identify the feasibility of deep learning-based diagnostic models for detecting and assessing lower-extremity fatigue fracture severity on plain radiographs. METHODS: This retrospective study enrolled 1151 X-ray images (tibiofibula/foot: 682/469) of fatigue fractures and 2842 X-ray images (tibiofibula/foot: 2000/842) without abnormal presentations from two clinical centers. After labeling the lesions, images in a center (tibiofibula/foot: 2539/1180) were allocated at 7:1:2 for model construction, and the remaining images from another center (tibiofibula/foot: 143/131) for external validation. A ResNet-50 and a triplet branch network were adopted to construct diagnostic models for detecting and grading. The performances of detection models were evaluated with sensitivity, specificity, and area under the receiver operating characteristic curve (AUC), while grading models were evaluated with accuracy by confusion matrix. Visual estimations by radiologists were performed for comparisons with models. RESULTS: For the detection model on tibiofibula, a sensitivity of 95.4%/85.5%, a specificity of 80.1%/77.0%, and an AUC of 0.965/0.877 were achieved in the internal testing/external validation set. The detection model on foot reached a sensitivity of 96.4%/90.8%, a specificity of 76.0%/66.7%, and an AUC of 0.947/0.911. The detection models showed superior performance to the junior radiologist, comparable to the intermediate or senior radiologist. The overall accuracy of the diagnostic model was 78.5%/62.9% for tibiofibula and 74.7%/61.1% for foot in the internal testing/external validation set. CONCLUSIONS: The deep learning-based models could be applied to the radiological diagnosis of plain radiographs for assisting in the detection and grading of fatigue fractures on tibiofibula and foot. KEY POINTS: • Fatigue fractures on radiographs are relatively difficult to detect, and apt to be misdiagnosed. • Detection and grading models based on deep learning were constructed on a large cohort of radiographs with lower-extremity fatigue fractures. • The detection model with high sensitivity would help to reduce the misdiagnosis of lower-extremity fatigue fractures.

Interpretable Molecular Property Predictions Using Marginalized Graph Kernels.

Marginalized graph kernels have shown competitive performance in molecular machine learning tasks but currently lack measures of interpretability, which are important to improve trust in the models, detect biases, and inform molecular optimization campaigns. We here conceive and implement two interpretability measures for Gaussian process regression using a marginalized graph kernel (GPR-MGK) to quantify (1) the contribution of specific training data to the prediction and (2) the contribution of specific nodes of the graph to the prediction. We demonstrate the applicability of these interpretability measures for molecular property prediction. We compare GPR-MGK to graph neural networks on four logic and two real-world toxicology data sets and find that the atomic attribution of GPR-MGK generally outperforms the atomic attribution of graph neural networks. We also perform a detailed molecular attribution analysis using the FreeSolv data set, showing how molecules in the training set influence machine learning predictions and why Morgan fingerprints perform poorly on this data set. This is the first systematic examination of the interpretability of GPR-MGK and thereby is an important step in the further maturation of marginalized graph kernel methods for interpretable molecular predictions.

Efficient Exploration of Chemical Compound Space Using Active Learning for Prediction of Thermodynamic Properties of Alkane Molecules.

We introduce an exploratory active learning (AL) algorithm using Gaussian process regression and marginalized graph kernel (GPR-MGK) to sample chemical compound space (CCS) at minimal cost. Targeting 251,728 enumerated alkane molecules with 4-19 carbon atoms, we applied the AL algorithm to select a diverse and representative set of molecules and then conducted high-throughput molecular simulations on these selected molecules. To demonstrate the power of the AL algorithm, we built directed message-passing neural networks (D-MPNN) using simulation data as the training set to predict liquid densities, heat capacities, and vaporization enthalpies of the CCS. Validations show that D-MPNN models built on the smallest training set considered in this work, which consists of 313 molecules or 0.124% of the original CCS, predict the properties with R2 > 0.99 against the computational data and R2 > 0.94 against the experimental data. The advantage of the presented AL algorithm is that the predicted uncertainty of GPR depends on only the molecular structures, which renders it compatible with high-throughput data generation.

Research relating to three-dimensional (3D) printing in spine surgery: a bibliometric analysis.

PURPOSE: Although numerous publications on three-dimensional printing (3DP) in spine surgery have been published, bibliometric analysis studies are scarce. Thus, this study aimed to present a bibliometric analysis of the status, hot spots, and frontiers of 3DP in spine surgery and associated research disciplines. METHODS: All publications relating to the utilization of 3DP in spine surgery from 1999 to May 9, 2022, were retrieved from the Web of Science. The bibliometric analysis was performed using CiteSpace software, and information on the country, institution, author, journal, and keywords for each publication was collected. RESULTS: A total of 270 articles were identified. From 2016 onward, a significant increase in publications on spinal surgery was observed. China was the most productive and influential country (98 publications) and H-index (22), followed by the USA and Australia. The most productive institution was Capital Medical University (9 publications). P. S. D'urso (8 publications, 46 citations) and R. J. Mobbs (8 publications, 39 citations) were the most prolific authors. European Spine Journal contributed the highest number of publications. The eight main clusters were: "rapid prototyping" #0, "3D printed" #1, "spine fusion" #2, "scoliosis" #3, "spine surgery" #4, "patient-specific" #5, "nervous system" #6, and "neuronavigation" #7. The strongest keyword bursts in 3DP in spine surgery were "fixation," "drill template," "instrumentation," "fusion," "complication," and "atlantoaxial instability." CONCLUSION: This analysis provides information on research trends and frontiers in the application of 3DP in spine surgery, as well as research and collaboration partners, institutions, and countries.

Comparison of full-endoscopic and tubular-based microscopic decompression in patients with lumbar spinal stenosis: a randomized controlled trial.

PURPOSE: We aimed to demonstrate non-inferiority in terms of functional outcomes in patients with lumbar spinal stenosis who underwent full-endoscopic decompression compared with tubular-based microscopic decompression. METHODS: This prospective, randomized controlled, non-inferiority trial included 60 patients with single-level lumbar spinal stenosis who required decompression surgery. The patients were randomly assigned in a 1:1 ratio to the full-endoscopic group (FE group) or the tubular-based microscopic group (TM group). Based on intention-to-treat analysis, the primary outcome was the Oswestry Disability Index score at 24 months postoperative. The secondary outcomes included the visual analog scale (VAS) score for back and leg pain, European Quality of Life-5 Dimensions (EQ-5D) score, walking time, and patient satisfaction rate according to the modified MacNab criteria. Surgery-related outcomes were also analyzed. RESULTS: Of the total patients, 92% (n = 55) completed a 24-month follow-up. The primary outcomes were comparable between the two groups (p = 0.748). However, the FE group showed a statistically significant improvement in the mean VAS score for back pain at day 1 and at 6, 12, and 24 months after surgery (p < 0.05). No significant difference was observed in the VAS score for leg pain, EQ-5D score, or walking time (p > 0.05). Regarding the modified MacNab criteria, 86.7% of patients in the FE group and 83.3% in the TM group had excellent or good results at 24 months after surgery (p = 0.261). Despite the similar results in surgery-related outcomes, including operative time, radiation exposure, revision rate, and complication rate, between the two groups (p > 0.05), less blood loss and shorter length of hospital stay were observed in the FE group (p ≤ 0.001 and 0.011, respectively). CONCLUSION: This study suggests that full-endoscopic decompression is an alternative treatment for patients with lumbar spinal stenosis because it provides non-inferior clinical efficacy and safety compared with tubular-based microscopic surgery. In addition, it offers advantages in terms of less invasive surgery. Trial registration number (TRN): TCTR20191217001.

Deep Learning-Based Diagnostic System for Velopharyngeal Insufficiency Based on Videofluoroscopy in Patients With Repaired Cleft Palates.

Velopharyngeal insufficiency (VPI), which is the incomplete closure of the velopharyngeal valve during speech, is a typical poor outcome that should be evaluated after cleft palate repair. The interpretation of VPI considering both imaging analysis and perceptual evaluation is essential for further management. The authors retrospectively reviewed patients with repaired cleft palates who underwent assessment for velopharyngeal function, including both videofluoroscopic imaging and perceptual speech evaluation. The final diagnosis of VPI was made by plastic surgeons based on both assessment modalities. Deep learning techniques were applied for the diagnosis of VPI and compared with the human experts' diagnostic results of videofluoroscopic imaging. In addition, the results of the deep learning techniques were compared with a speech pathologist's diagnosis of perceptual evaluation to assess consistency with clinical symptoms. A total of 714 cases from January 2010 to June 2019 were reviewed. Six deep learning algorithms (VGGNet, ResNet, Xception, ResNext, DenseNet, and SENet) were trained using the obtained dataset. The area under the receiver operating characteristic curve of the algorithms ranged between 0.8758 and 0.9468 in the hold-out method and between 0.7992 and 0.8574 in the 5-fold cross-validation. Our findings demonstrated the deep learning algorithms performed comparable to experienced plastic surgeons in the diagnosis of VPI based on videofluoroscopic velopharyngeal imaging.

An Effectiveness Evaluation of Nucleo-Annuloplasty for Lumbar Discogenic Lesions Using Disc-FX: A Scoping Review.

Background and Objectives: Degenerative disk disease is a widespread chronic condition that causes diskogenic pain. Diskogenic pain can be treated with various therapy methods. Disc-FX is a revolutionary, minimally invasive, percutaneous nucleo-annuloplasty method that combines manual diskectomy with nuclear and annular remodeling using radiofrequency ablation to relieve diskogenic pain. In this study, the technical features, clinical outcomes, and complications of Disc-FX are summarized. Materials and Methods: A comprehensive literature review was performed. By exploring several databases, we collected studies on Disc-FX for treating diskogenic pain. The outcomes included perioperative data, clinical results, and complications. Results: In the 15 studies included, data from 570 patients were collected. L4-L5 was the most frequently operated level, and most cases underwent single-level procedures. The follow-up period for these patients ranged from 2 months to 24 months. One study reported a procedure time between 35 and 60 min, whereas the remaining studies reported a procedure time of less than 30 min. The mean visual analog scale score decreased from 7.22 preoperatively to 1.81 at the final follow-up. The mean numerical rating scale score decreased from 6.98 preoperatively to 3.9 at the final follow-up. The mean Japanese Orthopaedic Association score improved from 16.26 preoperatively to 25.88 in the final follow-up. The mean Oswestry Disability Index score decreased from 35.37 preoperatively to 14.66 at the final follow-up. The mean satisfaction rate (based on the Macnab criteria) was 87.6% (range, 78.4-95.2%). The total incidence of postoperative transient pain was 8.77% (50/570) after nucleo-annuloplasty using Disc-FX, and recurrence was 1.58% (9/570). Conclusions: According to our comprehensive evaluation, using percutaneous nucleo-annuloplasty for treating lumbar diskogenic diseases provided considerable pain alleviation and improved functional outcomes with fewer complications. Disc-FX is a safe and effective procedure that is a good treatment option for patients with diskogenic pain.