Score Matching Research Articles

Interpretation knowledge extraction for genetic testing via question-answer model

BackgroundSequencing-based genetic testing is widely used in biomedical research, including pathogenic microorganism detection with metagenomic next-generation sequencing (mNGS). The application of sequencing results to clinical diagnosis and treatment relies on various interpretation knowledge bases. Currently, the existing knowledge bases are primarily built through manual knowledge extraction. This method requires professionals to read extensive literature and extract relevant knowledge from it, which is time-consuming and costly. Furthermore, manual extraction unavoidably introduces subjective biases. In this study, we aimed to automatically extract knowledge for interpreting mNGS results.MethodWe propose a novel approach to automatically extract pathogenic microorganism knowledge based on the question-answer (QA) model. First, we construct a MicrobeDB dataset since there is no available pathogenic microorganism QA dataset for training the model. The created dataset contains 3,161 samples from 618 published papers covering 224 pathogenic microorganisms. Then, we fine-tune the selected baseline model based on MicrobeDB. Finally, we utilize ChatGPT to enhance the diversity of training data, and employ data expansion to increase training data volume.ResultsOur method achieves an Exact Match (EM) and F1 score of 88.39% and 93.18%, respectively, on the MicrobeDB test set. We also conduct ablation studies on the proposed data augmentation method. In addition, we perform comparative experiments with the ChatPDF tool based on the ChatGPT API to demonstrate the effectiveness of the proposed method.ConclusionsOur method is effective and valuable for extracting pathogenic microorganism knowledge.

Advancements in Diffusion Models for Image Generation: A Comparative Analysis of DDPM, LDM, and DDIM

Abstract. This research provides a thorough exploration of diffusion models in image generation, comparing various methodologies to assess their efficacy and efficiency. The study begins with an introduction to foundational technologies and key concepts, progressing through an analysis of basic and advanced models, including Latent Diffusion Models (LDMs), Denoising Diffusion Implicit Models (DDIMs), and control models. The research evaluates these models based on their performance, computational efficiency, and future development potential. The review details the evolution of diffusion models from early stochastic processes to their current status as advanced generative models. Key principles, such as iterative noise addition and removal, are examined to understand the transformation from simple distributions to complex data representations. Innovations enhancing model efficiency, including advancements in score matching and neural network integration, are discussed. A thorough comparative analysis highlights the strengths and limitations of each model. The study identifies ongoing challenges such as interpretability and computational cost and proposes future research directions to address these issues. The findings aim to guide researchers and practitioners in advancing diffusion model technologies, offering insights into their impact on image generation and potential future developments.

Risk of Cardiovascular Disease in Patients With Classical Hodgkin Lymphoma: A Danish Nationwide Register-Based Cohort Study.

Risk of cardiovascular disease (CVD) in patients with classical Hodgkin lymphoma (cHL) undergoing contemporary treatment is unclear. cHL patients ≥ 18 years at diagnosis treated with doxorubicin-containing chemotherapy between 2000 and 2022 were matched 1:5 with comparators on birth year, sex, and Charlson Comorbidity Index at time of matching (score of 0 or ≥ 1). Cause-specific cumulative incidence of a composite of CVDs with corresponding 95% confidence intervals (CIs) were computed with death and lymphoma relapse as competing events (i.e., by censoring individuals at such occurrences) using the Aalen-Johansen estimator. A total of 1905 patients and 9525 comparators with a median follow-up of 10 years (interquartile range, [IQR]: 5.9-17.4). Median age was 39 years (IQR: 27-56), median cumulative doxorubicin dose was 250 mg/m2 (IQR: 200-300). The CVD cumulative incidences were 4.7% (95% CI: 3.6-5.7) for patients versus 2.6% (95% CI: 2.3-2.9) for comparators at 5 years, 8.9% (95% CI: 7.2-10.5) versus 5.5% (95% CI: 4.9-6.0) at 10 years, and 17.0% (95% CI: 14.1-19.9) versus 8.2% (95% CI: 7.4-9.0) at 15 years. CVD remains a substantial effect after contemporary treatment for cHL, suggesting that awareness of symptoms and a low threshold for referral to diagnostic examination are still important measures during survivorship.

Expected goals under a Bayesian viewpoint: uncertainty quantification and online learning

Abstract While the use of expected goals (xG) as a metric for assessing soccer performance is increasingly prevalent, the uncertainty associated with their estimates is often overlooked. This work bridges this gap by providing easy-to-implement methods for uncertainty quantification in xG estimates derived from Bayesian models. Based on a convenient posterior approximation, we devise an online prior-to-posterior update scheme, aligning with the typical in-season model training in soccer. Additionally, we present a novel framework to assess and compare the performance dynamics of two teams during a match, while accounting for evolving match scores. Our approach is well-suited for graphical representation and improves interpretability. We validate the accuracy of our methods through simulations, and provide a real-world illustration using data from the Italian Serie A league.

A propensity score-matched cost-effectiveness analysis of magnetic resonance-guided laser interstitial thermal therapy versus craniotomy for brain tumor radiation necrosis.

Radiation necrosis is becoming an increasingly prevalent complication in patients with brain tumors given the growing utility of stereotactic radiosurgery in modern treatment paradigms. Magnetic resonance-guided laser interstitial thermal therapy (MRgLITT) is a new minimally invasive modality that has exhibited an efficacy comparable to craniotomy in treating radiation necrosis. No studies to date have compared their cost-effectiveness despite the significant additional expenses associated with MRgLITT use. This study aimed to evaluate the cost-effectiveness of MRgLITT versus craniotomy in patients with comparable presentations of radiation necrosis. The National Inpatient Sample (NIS) was queried from 2011 to 2020 for patients with radiation necrosis and treated using craniotomy or MRgLITT. Admission charges and costs were inflation adjusted to 2020 $US. Surgical cohorts were propensity score-matched according to demographic, clinical, and admission characteristics. Multivariable linear and logistic regression analyses identified associations between type of intervention and outcomes. A semi-Markov model was created to simulate treatment with craniotomy versus MRgLITT. Cost, transition probabilities, and health state utilities were derived from the NIS, individual patient outcomes from multiple institutions, and prospectively collected quality-of-life data from a single institution and verified against other studies. Monte Carlo simulation and probabilistic sensitivity analysis were used to evaluate the cost-effectiveness between the two modalities. In the designated study period, 2869 patients had been admitted with brain tumor radiation necrosis and were managed with neurosurgical intervention. After propensity score matching, MRgLITT, relative to craniotomy, was independently associated with a shorter length of stay (LOS; β = -1.81, p = 0.002), lower odds of complications (OR 0.18, p = 0.033), and higher odds of home discharge (OR 3.05, p = 0.041), but there was no difference in total admission costs between the two modalities (β = $6229, p = 0.081). On Monte Carlo simulation, patients treated with MRgLITT had a lower probability of disease (radiation necrosis or tumor) recurrence (13.5% vs 22.0%, p < 0.001) but an equivalent mortality risk (22.8% vs 22.3%, p = 0.429) compared to the patients treated with craniotomy at the 1-year follow-up. Over a 4-year time horizon, MRgLITT had an incremental cost of -$25,685 and incremental effectiveness of 0.14 quality-adjusted life-year (QALY), resulting in an incremental cost-effectiveness ratio of -$183,464 per QALY relative to craniotomy. MRgLITT was a more cost-effective treatment strategy than craniotomy in the management of patients with brain tumor radiation necrosis. The cost-effectiveness of MRgLITT may be attributed to a shorter LOS, lower complication odds, and higher home discharge odds in the immediate postoperative period and a lower risk of disease recurrence over the long-term follow-up.

Sodium-glucose cotransporter 2 inhibitors and glaucoma in patients with type 2 diabetes

Purpose: Pleiotropic cardiovascular benefits of sodium glucose co-transporter 2 inhibitors (SGLT2i) have been demonstrated in patients with type 2 diabetes mellitus due to vascular remodeling effects. It is unclear whether a similar benefit may be seen for glaucoma. The purpose of this study is to assess the effect of SGLT2i on the risk of glaucoma in patients with type 2 diabetes. DesignTarget trial emulation using a population-based, propensity score-matched clinical cohort approach. MethodsSetting: Population-based, propensity score-matched clinical cohort study. Study Population: Adults with type 2 diabetes in the United States who newly initiated treatment with SGLT2i, dipeptidyl peptidase 4 inhibitors (DPP4i), or glucagon-like peptide-1 receptor agonists (GLP1RA) between 2013 and 2023. After propensity score matching, 722,446 patients were included in the SGLT2i arm and the DPP4i arm, respectively. Participants were matched based on age at index, race and sex, comorbidities, and concomitant use of medications. ExposureTreatment with SGLT2i for type 2 diabetes. Main Outcome Measure(s)Incidence of new-onset glaucoma and its subtypes after initiation of SGLT2i, DPP4i, or GLP1RA. Subgroup analyses were performed to demonstrate the effect of individual SGLT2i on incident glaucoma. ResultsPatients on SGLT2i compared to those on DPP4 had a lower risk of glaucoma (HR: 0.815, 95% confidence interval [CI]: 0.794, 0.837), including open-angle glaucoma (HR: 0.755, 95%CI: 0.729, 0.781) and primary angle-closure glaucoma (HR: 0.592, 95%CI: 0.540, 0.650). Among all SGLT2i, ertugliflozin (HR: 0.668, 95%CI: 0.512, 0.871) was associated with the lowest risk of glaucoma, followed by empagliflozin (HR: 0.727, 95%CI: 0.696, 0.759), then dapagliflozin (HR: 0.814, 95%CI: 0.774, 0.855). The protective effect of SGLT2i on glaucoma was validated when compared with GLP1RA (HR: 0.932, 95%CI: 0.906, 0.959). Conclusions: Patients on SGLT2i, especially ertugliflozin and empagliflozin, had a significantly lower risk of incident glaucoma compared to those on DPP4i, an association that was less robust but significant in a sensitivity analysis using GLP1RA as the active comparator. SGLT2i had a protective effect for both open-angle glaucoma and angle-closure glaucoma.

Closure of gastric mucosal defects using the reopenable-clip over the line method to decrease the risk of bleeding after endoscopic submucosal dissection: a multicenter propensity score-matched case-control study (with video)

Background and AimsEndoscopic submucosal dissection (ESD) is a minimally invasive treatment for early gastric cancer. However, post-ESD bleeding presents significant risks. Closing mucosal defects following ESD may reduce the incidence of post-ESD bleeding. Currently, no optimal closure method exists. Therefore, we invented clip-line closure using the reopenable-clip over the line method (ROLM) and evaluated its efficacy in preventing post-ESD bleeding. MethodsWe retrospectively reviewed data from patients who underwent gastric ESD between January 2012–March 2024. Patients were categorized into two groups: the non-closure group (mucosal defect remained unclosed) and the ROLM group (defect was closed using ROLM). Baseline characteristics of patients, distribution of bleeding risk factors, and incidence and timing of post-ESD bleeding were compared between the groups. Propensity score matching was employed to minimize potential bias. ResultsFollowing propensity score-matching, 162 ESDs were performed for 168 lesions in the non-closure group, whereas 160 ESDs were performed for 168 lesions in the ROLM group. The mean long diameter of the mucosal defects, procedure time for ROLM, and number of clips required for ROLM were 45.9 mm, 35.6 min, and 33.2, respectively. All mucosal defects resulting from the ESD were fully closed using ROLM. The post-ESD bleeding rate in the ROLM group was significantly lower (1.8%, 3/168 lesions) than in the non-closure group (7.7%, 13/168 lesions, p = 0.02). ConclusionsROLM is feasible for mucosal defect closure following gastric ESD and effectively prevents post-ESD bleeding in high-risk patients, addressing a significant gap in existing methods.

Challenges in Raman Spectroscopy of (micro)Plastics: The interfering role of colourants

Rising plastic consumption leads to widespread microplastic (MP) contamination. Raman spectroscopy is widely used for MP identification due to its ability to analyse particles as small as 1 μm. However, it faces challenges such as interference from pigments and additives. In this study, we aim to assess the accuracy of Raman micro-spectroscopy in identifying coloured plastic samples by applying various oxidative treatments to eliminate the possible interference effect caused by colourants associated with the sample. Standard and coloured microplastics were analysed using a Raman imaging microscope. Coloured plastics were treated with H2O2 30%, Sodium hypochlorite 5%, and Fenton reagent ( H2O2 30% and Ferrous sulphate 0.2 M) for 24, 48, and 72 hours. The Raman spectra were acquired after treatment to assess the impact of the treatment procedure on the polymer identification. Our results revealed that colourants significantly impact Raman spectra by peak broadening and/or fluorescence effects, which reduces identification accuracy and match scores Red pigments particularly obscure polymer identification. Treatments like oxidation and Fenton's reagent showed limited effectiveness. Additives in plastic samples can affect the accuracy of polymer identification by the Raman spectroscopy technique. Common treatment procedures do not improve the accuracy of identification. In order to improve the reliability of Raman analysis, essential factors such as utilizing multiple excitation lasers and appropriate CCD detectors, establishing a comprehensive reference library of colourants and additives, and employing advanced techniques like time-gated Raman spectroscopy or Surface-Enhanced Raman Spectroscopy (SERS) should be considered.

The lateral approach for total knee arthroplasty in valgus osteoarthritis provides similar clinical and radiological results compared with the medial approach.

In total knee arthroplasty (TKA), valgus deformities require special approaches to achieve precise mechanical axis correction and joint stability. This study aimed to determine whether the lateral approach without tibial tubercle osteotomy allows mechanical axis correction in valgus knee osteoarthritis (OA), which is at least as effective as the medial approach, while also assessing whether it provides clinically comparable results and complications. A total of 204 patients undergoing TKA for valgus knee OA were included in this retrospective comparative study, with 131 patients in the lateral approach group and 73 patients in the medial approach group. The surgical goal was to achieve stability with a neutral axis based on mechanical alignment. Detailed pre- and postoperative clinical and radiographic data were obtained, including measurements of the hip-knee-ankle angle (HKAA) and patella tilt angle. Functional status was determined based on the Oxford Knee Score (OKS). Age, body mass index, sex and preoperative HKAA were adjusted between both groups using propensity score matching (n = 45 per group). In the propensity score-matched cohort, the postoperative HKAA did not show differences when comparing the lateral and the medial approach at a mean follow-up of 19.7 months (p = 0.385). The achieved correction of the patella tilt angle was higher in the lateral approach group in the overall (p < 0.001) and matched cohort (p < 0.001). No differences were detected regarding the postoperative OKS and complications. However, wound healing disorders showed a tendency to be more prevalent in patients who underwent the lateral approach in the overall cohort (6.1%, n = 8 vs. 0.0%, n = 0, p = 0.053) but not in the propensity score-matched cohort (p = 0.494). This study demonstrated that for pronounced valgus OA, the lateral approach achieves favourable results in HKAA correction, patellar tilt and complications comparable to the medial approach. However, the risk of wound healing disorders appears higher with severe valgus deformity. Level III therapeutic study.

Atrial pace mapping with automatic intracardiac pattern matching for non-pulmonary vein trigger ablation: accuracy and feasibility

Abstract Atrial pace mapping (APM) is an alternative method to localize non-pulmonary vein (non-PV) triggers of atrial fibrillation(AF) that are otherwise not eligible for activation mapping due to their instability. The technique applies pattern matching of p-waves and intracardiac signals of atrial trigger and pacing from the site of the suspected origin. A new automated algorithm for intracardiac (IC) pattern matching was recently introduced into the 3D mapping system using the unipolar signal of the reference catheter. The study aims to assess the accuracy and feasibility of APM using IC-pattern matching for localizing non-PV triggers. Methods We assessed the precision of APM using a single chamber and double chamber reference in 20 patients who underwent index pulmonary vein isolation. APM was performed at the anterior and posterior wall of the left atrium (LApost, Laant) and the posterior and lateral wall of the right atrium (Rapost, Ralat). The first pacing point was declared as the origin. The IC-pattern matching algorithm was applied to calculate the similarity of the IC-pattern of the origin to the paced IC-pattern across various distances. The distribution of IC-pattern similarities was analyzed in the Python environment. The feasibility of APM was tested in patients who underwent REDO ablation of AF if all PVs were isolated and no low-voltage substrate was detected. The induction protocol for non-PV triggers included programmed and burst atrial pacing, high-dose isoproterenol, and adenosine infusion. If the IC pattern of the trigger was reproducible, the IC pattern matching was conducted. Ablation was performed at the site of the best IC-pattern matching. Results Atrial PM with IC-pattern matching and a single referent catheter showed lower precision in the RA lateral region (the median[Q1; Q3 ]area of 5% best matching score 1.48 [0.76; 2.01], 0.62[0.15; 1.06], 0.6[0.38; 1.19] and 5.68[2.89; 11.1] cm2 for LApost, LAant, RApost, RAlat accordingly, p = 0.0316). IC-pattern matching with double referent catheter showed high precision at all four anatomical regions (the median [Q1; Q3] area of 5% best matching score 0.126 [0.0879; 0.356], 0.486 [0.205; 0.717], 0.269 [0.198; 0.719], and 0.257 [0.0740; 0.572] cm2 for LApost, LAant, RApost, RAlat accordingly, Fig. 1). Seven out of 37 patients referred to REDO ablation had all PV isolated and absence of low-voltage substrate. In 3 of them, sustained tachycardias (AVNRT, focal atrial tachycardia and typical atrial flutter) were induced, mapped, and ablated. Non-sustained FAT from the RA posterior wall and PAC in the RA septum were induced under the high-dose isoproterenol infusion in two patients. Atrial pace mapping with double chamber referent and successful ablation of these non-PV triggers were performed (Fig.2). Conclusion The atrial pace mapping of non-PV triggers employing the automatic intracardiac pace mapping algorithm is feasible with high accuracy if a double-chamber referent is used.Accuracy of atrial pace mappingNon-PV trigger pace mapping

Long-term outcomes of percutaneous coronary intervention versus coronary artery bypass grafting in women with severe coronary artery disease

Abstract Background The optimal revascularization strategy for severe coronary artery disease (CAD) in women remains controversial. Despite differences between women and men in the development of CAD, there remains a paucity of comparative real-world observational evidence comparing percutaneous coronary intervention (PCI) to coronary artery bypass grafting (CABG) in women. Purpose We evaluated short and long-term outcomes of PCI and CABG in women with severe CAD. Methods We conducted a propensity score-matched retrospective cohort study using linked clinical and administrative databases in Ontario, Canada to obtain records of women with left main, 3-vessel disease, or 2-vessel disease with proximal left anterior descending artery involvement undergoing PCI or CABG from April 1st, 2012 to December 31st, 2021. Patients who were &amp;gt;80 years of age and patients who underwent urgent or emergent procedures were excluded for the primary analysis. The primary outcome was a composite of myocardial infarction, stroke, repeat revascularization, and death (MACCE). Secondary outcomes included the individual components of MACCE and a composite of readmission for MI, heart failure, and stroke. Following multiple imputation and propensity score-matching, the Cox proportional hazards model and Fine-Gray sub-distribution hazard model with robust variance estimators accounting for death as a competing risk were used to evaluate long-term outcomes. Results In the primary cohort, 2469 patients underwent PCI and 3,721 patients underwent CABG. After propensity score matching, including by coronary disease territory, 2,033 well-balanced pairs were identified, with standardized mean differences &amp;lt;0.10 for all covariates. In-hospital death and stroke were significantly lower with PCI than with CABG (0.7% vs 1.5%, p=0.036) and (&amp;lt;0.2% vs 1.4%, p&amp;lt;0.001), respectively, and there was no difference in perioperative MI between PCI and CABG (1.1% vs 0.6%, p=0.121). At a median (IQR) of 5.1 (2.9-7.5) years, major adverse cardiac and cerebrovascular events (MACCE) was higher with PCI compared with CABG (37.7% vs 23.3%, HR 1.79, [95% CI: 1.66-1.91], p&amp;lt;0.001) (Figure 1). All-cause mortality was higher with PCI compared with CABG (17.8% vs 13.4%, HR 1.33, [95% CI: 1.21-1.45], p&amp;lt;0.001) (Figure 2). The composite readmission outcome (MI, heart failure, stroke) was higher with PCI compared with CABG (16.2% vs 11.2%, HR 1.42, [95% CI: 1.37-1.48], p&amp;lt;0.001). Conclusion In women with severe CAD, CABG appears to be associated with a long-term reduction in MACCE and mortality compared with PCI. Our findings provide evidence to suggest women with severe CAD may derive greater long-term benefit from CABG compared to PCI and suggest that the treatment recommendation should shift towards CABG in appropriately selected women.Figure 1.MACCE free survivalFigure 2.Overall survival

Comparing methods of measuring interest fit: A large prediction study with career choice satisfaction

AbstractVocational interest inventories are widely used in both research and practice to help match people to well‐fitting work environments. However, because there are many different methods to operationalize interest fit, a debate remains regarding the best ways to do so. To empirically inform this debate, our study compared the predictive power of four widely used interest fit indices (i.e., matching scale scores, profile deviance scores, profile correlations, and polynomial regression scores) for predicting career choice satisfaction. Using a large and diverse U.S. sample (N = 257,320), results indicated that among the three single‐term interest fit measures, profile correlations (R2 = .04) explained more variance in career choice satisfaction than matching scale scores (R2 = .02) and profile deviance scores (R2 = .00). By comparison, the full 30‐term polynomial regression model explained the most variance in career choice satisfaction (R2 = .09); in this case, however, the nonlinear terms that capture fit effects only accounted for about 22% (R2 = .02) of the total variance explained by the model. Overall, these results indicate that researchers and practitioners should be cautious of the greater criterion‐related validity of polynomial regression models as fit information may not be a substantial contributor to their predictive capacities. In addition, our findings support the use of profile correlations as a predictive, single‐term measure of interest fit.

Exploring the medicinal potential of Dendrobium: Uncovering the spatial distribution of flavonoids and alkaloids in 15 species of Dendrobium using MALDI-MSI

Dendrobium, the second largest genus in the Orchidaceae family, has been regarded as a valuable medicinal plant throughout Chinese history. Secondary metabolites present in the stem of Dendrobium, including alkaloids and flavonoids, confer significant pharmacological effects such as anti-inflammatory, antibacterial, immunomodulatory, antitumor, and antioxidant properties. To our knowledge, this study is the first to apply matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI-MSI) technology to comprehensively identify the spatial distribution of secondary metabolites, particularly flavonoids and alkaloids, in the mature stems of 15 common Dendrobium species. By analyzing the matching scores and relative concentrations of compounds, 20 flavonoids and 19 alkaloids with high identification credibility were identified, along with 4 sesquiterpenoid compounds. Their spatial distributions and compositions were analyzed, and the results revealed that despite the similar structural basis of the stems in the 15 Dendrobium species, significant differences existed in the distribution of secondary metabolites among different species. Flavonoids and alkaloids with different medicinal effects exhibited certain specific distribution patterns; for instance, flavonoids with antioxidant effects were mostly concentrated in the outer circle of thin-walled cells of Dendrobium stems, whereas those with antitumor effects were found in higher concentrations in the central areas of some Dendrobium species than in the periphery. Alkaloids with antitumor properties typically exhibited a higher concentration in peripheral thin-walled cells than in central regions. Furthermore, statistical analysis was conducted on the content of secondary metabolites with different effects among the 15 Dendrobium species to identify species with greater medicinal potential in terms of antioxidant and antitumor effects. The findings of this study not only enhance our understanding of the distribution patterns of flavonoids and alkaloids within Dendrobium stems but also offer valuable insights for the medicinal development and quality assessment of Dendrobium.

Annotating full-scan MS data using tandem MS libraries.

Full-scan mass spectrometry (MS) data from both liquid chromatography (LC) and MS imaging capture multiple ion forms, including their in-source fragments. Here we leverage such fragments to structurally annotate full-scan data from LC-MS or MS imaging by matching against peak intensity scaled tandem MS spectral libraries using precursor-tolerant reverse match scoring. Applied to inflammatory bowel disease and imaging datasets, we show the approach facilitates re-analyses of data in public repositories.

Stereotactic body radiotherapy versus lenvatinib for hepatocellular carcinoma with portal vein tumor thrombosis: a propensity matching score analysis

Background and objectivesThe purpose of this study was to investigate the survival benefit of Stereotactic Body Radiotherapy (SBRT) versus lenvatinib as first-line therapy in the treatment of hepatocellular carcinoma (HCC) with portal vein tumor thrombosis (PVTT).Materials and methods147 HCC patients with PVTT were included in this retrospective study, 70 were treated with SBRT and 77 of were treated with lenvatinib. Propensity score matching (PSM) analysis was employed to balance the differences in baseline characteristics between the two groups. Overall survival (OS), progression-free survival (PFS) and objective response rate (ORR) were compared between the two groups. In addition, the safety of patients in both groups was also evaluated.ResultsAfter PSM, 38 patients were matched in each of the two groups. The median OS was 14.5 (95% CI: 10.1–18.9) and 11.1 (95% CI: 9.3–12.9) months in the SBRT and lenvatinib groups, respectively (P = 0.014). The median PFS was 6.8 (95% CI: 5.1–8.5) and 5.0 (95% CI: 3.0–7.0) months, respectively (P = 0.010). The 1-, 2-years OS rates in the two groups were 65.8% vs. 39.5% and 31.6% vs. 10.5%, respectively. The 6-, 12-months PFS rates in the two groups were 57.9% vs. 44.7% and 28.9% vs. 10.5%, respectively. In addition, the SBRT group had a better ORR than the lenvatinib group (52.6% vs. 23.7%, P = 0.009). Patients with good response to SBRT had better survival. Cox proportional hazard model showed that SBRT was an important prognostic factor for OS and PFS. The incidence of hypertension (34.2% vs. 0%) was higher in the LEN group, however, both treatment modalities were well tolerated in the two groups of patients.ConclusionIn HCC patients with PVTT, SBRT had a better survival benefit than Lenvatinib treatment as first-line therapy.

LLM-enhanced Composed Image Retrieval: An Intent Uncertainty-aware Linguistic-Visual Dual Channel Matching Model

Composed image retrieval (CoIR) involves a multi-modal query of the reference image and modification text describing the desired changes, allowing users to express image retrieval intents flexibly and effectively. The key of CoIR lies in how to properly reason the search intent from the multi-modal query. Existing work either aligns the composite embedding of the multi-modal query and the target image embedding in the visual domain through late-fusion or converts all images into text descriptions and leverage large language models (LLM) for text semantic reasoning. However, this single-modality reasoning approach fails to comprehensively and interpretably capture the users’ ambiguous and uncertain intents in the multi-modal queries, incurring the inconsistency between retrieved results and ground truth. Besides, the expensive manually-annotated datasets limit the further performance improvement of CoIR. To this end, this paper proposes an LLM-enhanced Intent Uncertainty-aware Linguistic-Visual Dual Channel Matching Model (IUDC), which combines the strengths of multi-modal late-fusion and LLMs for composed image retrieval. We first construct an LLM-based triplet augmentation strategy to generate more synthetic training triplets. Based on this, the core of IUDC consists of two matching channels: the semantic matching channel is responsible for intent reasoning on the aspect-level attributes extracted by an LLM, and the visual matching channel accounts for the fine-grained visual matching between multi-modal fusion embedding and target images. Considering the intent uncertainty presented in the multi-modal queries, we introduce Probability Distribution Encoder (PDE) to project the intents as probabilistic distributions in the two matching channels. Consequently, a mutually enhanced module is designed to share knowledge between the visual and semantic representations for better representation learning. Finally, the matching scores of two channels are added to retrieve the target image. Extensive experiments conducted on two real datasets demonstrate the effectiveness and superiority of our model. Notably, with the help of the proposed LLM-based triplet augmentation strategy, our model achieves a new record of state-of-the-art performance among all datasets.

Cutting-edge applications of oleic acid-modified CdSiO3: Ce3+ phosphors: Artificial intelligence enhanced latent fingerprint detection, anti-counterfeiting and optical thermometry

A series of (1–11 mol%) Ce³⁺ doped CdSiO₃ phosphors are synthesized via a solution combustion method, and the sample with the optimal doping concentration underwent surface modification with oleic acid (OA). X-ray diffraction (XRD) analysis confirmed the formation of high-purity, well-crystallized CdSiO3:Ce3+ phosphors. Upon 347 nm UV excitation, the phosphor exhibited intense blue emission at 400 nm, with an optimized Ce³⁺ concentration of 5 mol%, attributed to dipole-dipole interactions. The OA-modified phosphor demonstrated significantly enhanced properties, achieving a color purity (CP) of 97.2% and Internal quantum efficiency (IQE) of 82.5%. The phosphor also exhibited excellent thermal stability, retaining 93.0% of its emission intensity at 423 K and a thermal quenching temperature exceeding 483 K, with a high activation energy (Eₐ) of 0.384 eV. In practical applications, the OA modified phosphor showed exceptional performance in latent fingerprints (LFPs) detection and anti-counterfeiting (AC) across different substrates, offering high resolution, contrast, and minimal background interference. MATLAB based analysis yielded a matching score of 94.52%, surpassing conventional benchmarks, underscoring the phosphor's potential for advanced fingerprint identification. These results demonstrate that the CdSiO₃:5Ce³⁺ phosphor is a promising candidate for applications in white light-emitting diodes (w-LEDs), optical thermometry, forensic analysis and AC technologies.

Lung ultrasound assessment of atelectasis following different anesthesia induction techniques in pediatric patients: a propensity score-matched, observational study

IntroductionAtelectasis is a well-documented complication in pediatric patients undergoing general anesthesia. Its incidence varies significantly based on surgical procedures and anesthesia techniques. Inhalation induction, commonly used to avoid the discomfort of venipuncture, is suspected to cause higher rates of respiratory complications, including atelectasis, compared to intravenous induction. This study aimed to evaluate the impact of inhalation versus intravenous anesthesia induction on atelectasis formation in pediatric patients, as assessed by lung ultrasound (LUS).MethodsThis propensity score-matched observational study was conducted at a tertiary pediatric hospital in Milan, Italy. Inclusion criteria were children ≤ 18 years undergoing elective surgery with general anesthesia. Patients were divided into inhalation and intravenous induction groups. LUS was performed before and after anesthesia induction to assess lung aeration. The primary endpoint was the global LUS score post-induction, with secondary endpoints including the incidence and distribution of atelectasis.ResultsOf the 326 patients included, 65% underwent inhalation induction and 35% intravenous induction. The global LUS score was significantly higher in the inhalation group (12.0 vs. 4.0, p < 0.001). After propensity score matching (for age, presence of upper respiratory tract infection, duration of induction, and PEEP levels at induction), average treatment effect (ATE) of mask induction was 5.89 (95% CI, 3.21–8.58; p < 0.001) point on LUS global score and a coefficient of 0.35 (OR 1.41) for atelectasis.DiscussionInhalation induction is associated with a higher incidence of atelectasis in pediatric patients also when we adjusted for clinically relevant covariates.Trial registrationClinicalTrials.gov identifier: NCT06069414.

Long-Term Outcomes of Pneumatic Balloon Dilation versus Peroral Endoscopic Myotomy in Type I and II Achalasia Patients: A Propensity Score-Matched Analysis

Peroral esophageal myotomy (POEM) shows higher efficacy than pneumatic balloon dilation (PBD) for type I and II achalasia over 2 years. However, long-term durability beyond 5 years remains under investigated. This study aimed to assess the long-term efficacy of PBD versus POEM. This retrospective comparative study examined treatment-naïve achalasia patients who underwent either PBD or POEM. Cohort balance was achieved through propensity score (PS) matching based on age, sex, Eckardt's score, and Charlson's comorbidity index. The primary outcome was the time-to-second intervention. Secondary outcomes included treatment failure, Eckardt's score, number of intervention sessions, weight gain, and adverse events. Of the 127 patients, 67 underwent PBD and 60 underwent POEM, with an average follow-up of 7 years (IQR 5-10 years). After PS matching, 100 patients with 1:1 matching were included in the analysis. Baseline characteristics, including clinical, endoscopic, and manometric attributes, were comparable between groups. The POEM group had a significantly longer time-to-second intervention than the PBD group (hazard ratio 0.139, 95% CI 0.048-0.405) and underwent more endoscopic sessions (p< 0.001 by linear marginal model analysis) within the first 2 years. Beyond 2 years, subsequent interventions were comparable between groups. There were no differences in the Eckardt score or weight gain. Hospital stays were shorter in the PBD group (2 vs. 4 days; p < 0.001). POEM patients had more adverse events, including perforation. POEM demonstrated greater long-term durability than PBD, but PBD was associated with shorter hospital stays and fewer adverse events.

Impact of Protonation Sites on Collision-Induced Dissociation-MS/MS Using CIDMD Quantum Chemistry Modeling.

Protonation is the most frequent adduct found in positive electrospray ionization collision-induced mass spectra (CID-MS/MS). In a parallel report Lee, J. J. Chem. Inf. Model. 2024, 10.1021/acs.jcim.4c00760, we developed a quantum chemistry framework to predict mass spectra by collision-induced dissociation molecular dynamics (CIDMD). As different protonation sites affect fragmentation pathways of a given molecule, the accuracy of predicting tandem mass spectra by CIDMD ultimately depends on the choice of its protomers. To investigate the impact of molecular protonation sites on MS/MS spectra, we compared CIDMD-predicted spectra to all available experimental MS/MS spectra by similarity matching. We probed 10 molecules with a total of 43 protomers, the largest study to date, including organic acids (sorbic acid, citramalic acid, itaconic acid, mesaconic acid, citraconic acid, and taurine) as well as aromatic amines including uracil, aniline, bufotenine, and psilocin. We demonstrated how different protomers can converge different fragmentation pathways to the same fragment ions but also may explain the presence of different fragment ions in experimental MS/MS spectra. For the first time, we used in silico MS/MS predictions to test the impact of solvents on proton affinities, comparing the gas phase and a mixture of acetonitrile/water (1:1). We also extended applications of in silico MS/MS predictions to investigate the impact of protonation sites on the energy barriers of isomerization between protomers via proton transfer. Despite our initial hypothesis that the thermodynamically most stable protomer should give the best match to the experiment, we found only weak inverse relationships between the calculated proton affinities and corresponding entropy similarities of experimental and CIDMD-predicted MS/MS spectra. CIDMD-predicted mechanistic details of fragmentation reaction pathways revealed a clear preference for specific protomer forms for several molecules. Overall, however, proton affinity was not a good predictor corresponding to the predicted CIDMD spectra. For example, for uracil, only one protomer predicted all experimental MS/MS fragment ions, but this protomer had neither the highest proton affinity nor the best MS/MS match score. Instead of proton affinity, the transfer of protons during the electrospray process from the initial protonation site (i.e., mobile proton model) better explains the differences between the thermodynamic rationale and experimental data. Protomers that undergo fragmentation with lower energy barriers have greater contributions to experimental MS/MS spectra than their thermodynamic Boltzmann populations would suggest. Hence, in silico predictions still need to calculate MS/MS spectra for multiple protomers, as the extent of distributions cannot be readily predicted.