175030 research outputs found
Sort by
Development of Machine Learning Algorithms for Identifying Patients With Limited Health Literacy
Full text linkRationale: Limited health literacy (HL) leads to poor health outcomes, psychological stress, and misutilization of medical resources. Although interventions aimed at improving HL may be effective, identifying patients at risk of limited HL in the clinical workflow is challenging. With machine learning (ML) algorithms based on readily available data, healthcare professionals would be enabled to incorporate HL screening without the need for administering in-person HL screening tools. Aims and Objectives: Develop ML algorithms to identify patients at risk for limited HL in spine patients. Methods: Between December 2021 and February 2023, consecutive English-speaking patients over the age of 18 and new to an urban academic outpatient spine clinic were approached for participation in a cross-sectional survey study. HL was assessed using the Newest Vital Sign and the scores were divided into limited (0–3) and adequate (4–6) HL. Additional patient characteristics were extracted through a sociodemographic survey and electronic health records. Subsequently, feature selection was performed by random forest algorithms with recursive feature selection and five ML models (stochastic gradient boosting, random forest, Bayes point machine, elastic-net penalized logistic regression, support vector machine) were developed to predict limited HL. Results: Seven hundred and fifty-three patients were included for model development, of whom 259 (34.4%) had limited HL. Variables identified for predicting limited HL were age, Area Deprivation Index-national, Social Vulnerability Index, insurance category, Body Mass Index, race, college education, and employment status. The Elastic-Net Penalized Logistic Regression algorithm achieved the best performance with a c-statistic of 0.766, calibration slope/intercept of 1.044/−0.037, and Brier score of 0.179. Conclusion: Elastic-Net Penalized Logistic Regression had the best performance when compared with other ML algorithms with a c-statistic of 0.766, calibration slope/intercept of 1.044/−0.037, and a Brier score of 0.179. Over one-third of patients presenting to an outpatient spine center were found to have limited HL. While this algorithm is far from being used in clinical practice, ML algorithms offer a potential opportunity for identifying patients at risk for limited HL without administering in-person HL assessments. This could possibly enable screening and early intervention to mitigate the potential negative consequences of limited HL without taxing the existing clinical workflow
Integrin-activating Yersinia protein Invasin sustains long-term expansion of primary epithelial cells as 2D organoid sheets
Full text linkMatrigel®/BME®, a basement membrane-like preparation, supports long-term growth of epithelial 3D organoids from adult stem cells [T. Sato et al., Nature 459, 262–265 (2009); T. Sato et al., Gastroenterology 141, 1762–1772 (2011)]. Here, we show that interaction between Matrigel’s major component laminin-111 with epithelial α6β1-integrin is crucial for this process. The outer membrane protein Invasin of Yersinia is known to activate multiple integrin–β1 complexes, including integrin α6β1. A C-terminal integrin-binding fragment of Invasin, coated on culture plates, mediated gut epithelial cell adhesion. Addition of organoid growth factors allowed multipassage expansion in 2D. Polarization, junction formation, and generation of enterocytes, goblet cells, Paneth cells, and enteroendocrine cells were stable over time. Sustained expansion of other human, mouse, and even snake epithelia was accomplished under comparable conditions. The 2D “organoid sheet” format holds advantages over the 3D “in gel” format in terms of imaging, accessibility of basal and apical domains, and automation for high-throughput screening
Positive associations between mean ambient temperature and involuntary admissions to psychiatric facilities
Full text linkBACKGROUND: Temperature increases in the context of climate change affect numerous mental health outcomes. One such relevant outcome is involuntary admissions as these often relate to severe (life)threatening psychiatric conditions. Due to a shortage of studies into this topic, relationships between mean ambient temperature and involuntary admissions have remained largely elusive. AIMS: To examine associations between involuntary admissions to psychiatric institutions and various meteorological variables. METHODS: Involuntary admissions data from 23 psychiatric institutions in the Netherlands were linked to meteorological data from their respective weather stations. Generalized additive models were used, integrating a restricted maximum likelihood method and thin plate regression splines to preserve generalizability and minimize the risk of overfitting. We thus conducted univariable, seasonally stratified, multivariable, and lagged analyses. RESULTS: A total of 13,746 involuntary admissions were included over 21,549 days. In univariable and multivariable models, we found significant positive associations with involuntary admissions for ambient temperature and windspeed, with projected increases of up to 0.94% in involuntary admissions per degree Celsius temperature elevation. In the univariable analyses using all data, the strongest associations in terms of significance and explained variance were found for mean ambient temperature (p = 2.5 × 10-6, Variance Explained [r2] = 0.096%) and maximum ambient temperature (p = 8.65 × 10-4, r2 = 0.072%). We did not find evidence that the lagged associations explain the associations for ambient temperature better than the direct associations. CONCLUSION: Mean ambient temperature is consistently but weakly associated with involuntary psychiatric admissions. Our findings set the stage for further epidemiological and mechanistic studies into this topic, as well as for modeling studies examining future involuntary psychiatric admissions
Assessing the role of spatial aggregation schemes with varying campaign durations of mobile measurements on land use regression models for estimating nitrogen dioxide
Full text linkMobile air pollution measurements are typically aggregated by varying road segment lengths, grid cell sizes, and time intervals. How these spatiotemporal aggregation schemas affect the modeling performance of land use regression models has seldom been assessed. We used 5.7 million mobile nitrogen dioxide (NO2) measurements collected over 160 days in Amsterdam (The Netherlands) and subsampled them into five campaign durations (10–70 days). We aggregated the measurements from each campaign duration onto road segments and grid cells with five spatial scales (25–200 m). A stepwise linear regression (SLRs) and random forests (RFs) were trained for each aggregated dataset to predict NO2 concentrations. The model accuracies were validated using a 30% hold-out sample of mobile measurements and external Palmes long-term stationary measurements (n = 105). At increased spatial scales, the prediction accuracy decreased for RFs but increased for SLRs when validated against mobile measurements. Using long-term stationary measurements, prediction accuracy varied across scales without any clear pattern. Regardless of cells or road segments, the models performed similarly at small scales (i.e., 25 m and 50 m). Models based on road segments were less sensitive to spatial scales than those based on cells in mobile and long-term external validations. Longer campaign durations increased the prediction accuracies of long-term NO2 concentrations, though the gain in accuracy diminished after 50 days. In conclusion, our results suggest that road segments are preferred when the aggregation scale gets larger as this approach likely reduces scale-dependent influences. The campaign duration plays a more important role in long-term NO2 prediction than spatial scales
Salpingectomy With Delayed Oophorectomy Versus Salpingo-Oophorectomy in BRCA1/2 Carriers: Three-Year Outcomes of a Prospective Preference Trial
Full text linkObjective: To compare menopause-related quality of life (QoL) after risk-reducing salpingectomy (RRS) versus risk-reducing salpingo-oophorectomy (RRSO) until 3 years of post-surgery. Design: A prospective study (TUBA study) with treatment allocation based on patients' preference. Data were collected pre-surgery and at 3 months, 1 and 3 years of post-surgery. Setting: Multicentre prospective preference trial in thirteen hospitals in the Netherlands. Population: BRCA1/2 pathogenic variant (PV) carriers aged 25–40 (BRCA1) or 25–45 (BRCA2), who were premenopausal, without a future child wish and without current (treatment for) malignancy. Methods: Treatment allocation was based on patients' preference: either RRS from the age of 25 years with delayed oophorectomy at the maximum age of 45 (BRCA1) or 50 (BRCA2), or RRSO between the ages of 35–40 (BRCA1) or 40–45 (BRCA2). After RRSO, hormone replacement therapy (HRT) was recommended, if not contraindicated. Primarily, menopause-related QoL as measured with the Greene Climacteric Scale (GCS) was compared between the RRS and RRSO without HRT group. Secondarily, GSC-scores of the RRS group were compared with the scores of the RRSO with HRT after surgery group. A higher GSC-score reflects more climacteric symptoms. Results: Until April 2023, 410 participants had undergone RRS and 160 RRSO. The BRCA1/BRCA2 proportions were 51.4%/48.6%. The mean age at surgery (SD) was 37.9 (3.5) years. Participants 3 years after RRSO without HRT had a 4.3 (95% CI 2.1–6.5; p < 0.001) point higher increase in GCS-score from baseline compared to those after RRS, while the difference was 7.9 (95% CI 5.9–9.8) and 8.5 (95% CI 6.5–10.5) points at 3 and 12 months, respectively. Among participants with HRT after surgery, the RRSO group had a 2.4 (95% CI 0.8–3.9; p = 0.002) point higher increase in GCS-score from baseline compared to the RRS group. Conclusions: In this multicentre preference trial, menopause-related QoL was better after RRS than after RRSO, even with HRT after RRSO. Differences between arms were most pronounced until one-year post-surgery
Red blood cell pyruvate kinase properties in Townes and Berkeley sickle cell disease mouse models – Of mice and men
Full text linkPyruvate kinase (PK), a key ATP-generating enzyme in glycolysis, is a target for novel sickle cell disease (SCD) therapies. Enhancing PK activity lowers 2,3-diphosphyglycerate (2,3-DPG), increases adenosine triphosphate (ATP), and may prevent red blood cell (RBC) sickling. Townes and Berkeley SCD mouse models are commonly used for the development of novel drugs for SCD, but differ from humans in 2,3-DPG and ATP levels, which could be related to underlying differences in PK properties. This study revealed important distinctions with humans (SCD vs healthy controls), such as similar PK/hexokinase (HK) ratios between sickling and non-sickling mouse models and significantly lower PK thermostability in mice. We additionally investigated the effect of a novel RBC PK activator, compound A, on PK properties and sickling tendency in these mice in order to assess SCD mouse model suitability. Results showed that a single dose of compound A led to an increased affinity of PK for phosphoenolpyruvate, a significant increase in PK/HK ratio and a decrease of 2,3-DPG levels. Together, these results offer detailed characterization in the PK properties of two commonly used SCD mouse models, and provide insight into the mode of action of PK activator therapy in SCD mice models
Risk Factors for the Development of Neurological Deficits in Metastatic Spinal Disease: An International, Multicenter Delphi Study
Full text linkStudy Design: Delphi study Objective: The objective of this study was to identify risk factors associated with the development and/or progression of neurological deficits in patients with metastatic spinal disease. Methods: A three-round Delphi study was conducted between January-May 2023 including AO Spine members, comprising mainly neurosurgeons and orthopedic surgeons. In round 1, participants listed radiological factors, patient characteristics, tumor characteristics, previous cancer-related treatment factors and additional factors. In round 2, participants ranked the factors on importance per category and selected a top 9 from all factors. Kendall’s W coefficient of concordance was calculated as a measure of consensus. In the final round, participants provided feedback on the rankings resulting from round 2. Lastly, the highest-ranking factors were more clearly defined and operationalized by an expert panel. Results: Over two hundred physicians and researchers participated in each round. The factors listed in the first round were collapsed into 12 radiological factors, 14 patient characteristics, 6 tumor characteristics and 12 previous cancer-related treatment factors. High agreement was found in round 3 on the top-half lists in each category and the overall top 9, originating from round 2. Kendall’s W indicated strong agreement between the participants. ‘Epidural spinal cord compression’, ‘aggressive tumor behavior’ and ‘mechanical instability’ were deemed most influential for the development of neurological deficits. Conclusion: This study provides factors that may be related to the development and/or progression of neurological deficits in patients with metastatic spinal disease. This list can serve as a basis for future directions in prognostication research
Pharmacokinetic exposure and treatment outcomes of lenvatinib in patients with renal cell carcinoma and differentiated thyroid carcinoma
No full textPURPOSE: After initial approval of lenvatinib for radioiodine-refractory differentiated thyroid cancer (DTC), it has also shown promising outcomes in among others metastatic renal cell carcinoma (mRCC). Given that trial populations typically do not represent routine clinical care populations, questions arise about how applicable trial outcomes are in clinical practice. This study aims to compare the pharmacokinetics (PK), toxicity patterns, and survival data of lenvatinib in a real-world cohort with DTC and mRCC to those observed in pivotal clinical trials. MATERIALS AND METHODS: Patients were included when diagnosed with DTC or mRCC, had received current or prior treatment with lenvatinib, and had at least one available lenvatinib plasma concentration measurement. A descriptive comparison was made between the baseline characteristics, PK data, toxicity and survival data in this real-world cohort and those described in the phase III trials. RESULTS: Overall, 29 patients with mRCC and 35 patients with DTC were included. For mRCC, median time to treatment discontinuation (mTTD) was shorter than observed in the phase III trial (7.5 versus 11.0 months) with fewer dose-limiting toxicities, likely because 66% of the patients started with a reduced dose. mRCC patients were more pretreated and had a worse performance status than trial participants. This was resembled in overall lower PK exposure in mRCC patients. For DTC, mTTD was longer in our cohort (17.1 versus 13.8 months), with similar toxicity patterns and PK exposure as in the phase III trial. CONCLUSIONS: Our data suggests that patient characteristics and outcomes in routine clinical care deviate from clinical trials and show the need for alternative treatment strategies to manage tolerability to lenvatinib
Nanoparticle-in-Hydrogel Delivery System for the Sequential Release of Two Drugs
Full text linkBackground/Objectives: Glioblastoma is the most common and lethal primary brain tumor. Patients often suffer from tumor- and treatment induced vasogenic edema, with devastating neurological consequences. Intracranial edema is effectively treated with dexamethasone. However, systemic dexamethasone requires large doses to surpass the blood brain barrier in therapeutic quantities, which is associated with significant side effects. The aim of this study was to investigate a biodegradable, dextran-hydroxyethyl methacrylate (dex-HEMA) based hydrogel, containing polymeric micelles loaded with dexamethasone and liposomes encapsulating dexamethasone phosphate for localized and prolonged delivery. Methods: Poly(ethylene glycol)-b-poly(N-2-benzoyloxypropyl methacrylamide (mPEG-b-p(HPMA-Bz)) micelles were loaded with dexamethasone and characterized. The dexamethasone micelles, together with dexamethasone phosphate liposomes, were dispersed in an aqueous dex-HEMA solution followed by radical polymerization using a photoinitiator in combination with light. The kinetics and mechanisms of drug release from this hydrogel were determined. Results: The diameter of the nanoparticles was larger than the mesh size of the hydrogel, rendering them immobilized in the polymer network. The micelles immediately released free dexamethasone from the hydrogel for two weeks. The dexamethasone phosphate loaded in the liposomes was not released until the gel degraded and intact liposomes were released, starting after 15 days. The different modes of release result in a biphasic and sequential release profile of dexamethasone followed by dexamethasone phosphate liposomes. Conclusions: The results show that this hydrogel system loaded with both dexamethasone polymeric micelles and dexamethasone phosphate loaded liposomes has potential as a local delivery platform for the sequential release of dexamethasone and dexamethasone phosphate, for the intracranial treatment of glioblastoma associated edema
The Harms of Class Imbalance Corrections for Machine Learning Based Prediction Models: A Simulation Study
Full text linkIntroduction: Risk prediction models are increasingly used in healthcare to aid in clinical decision-making. In most clinical contexts, model calibration (i.e., assessing the reliability of risk estimates) is critical. Data available for model development are often not perfectly balanced with the modeled outcome (i.e., individuals with vs. without the event of interest are not equally prevalent in the data). It is common for researchers to correct for class imbalance, yet, the effect of such imbalance corrections on the calibration of machine learning models is largely unknown. Methods: We studied the effect of imbalance corrections on model calibration for a variety of machine learning algorithms. Using extensive Monte Carlo simulations we compared the out-of-sample predictive performance of models developed with an imbalance correction to those developed without a correction for class imbalance across different data-generating scenarios (varying sample size, the number of predictors, and event fraction). Our findings were illustrated in a case study using MIMIC-III data. Results: In all simulation scenarios, prediction models developed without a correction for class imbalance consistently had equal or better calibration performance than prediction models developed with a correction for class imbalance. The miscalibration introduced by correcting for class imbalance was characterized by an over-estimation of risk and was not always able to be corrected with re-calibration. Conclusion: Correcting for class imbalance is not always necessary and may even be harmful to clinical prediction models which aim to produce reliable risk estimates on an individual basis