Stratification Of Data Research Articles

EXPLORING FACTORS ASSOCIATED WITH SEROLOGICAL RESPONSE TO SARS-COV-2 VACCINES IN INDIVIDUALS WITH INFLAMMATORY BOWEL DISEASE: INTERACTIVE VISUALIZATION METHODS FOR MULTIDIMENSIONAL DATA

Abstract BACKGROUND SARS-CoV-2 vaccine response may be reduced in seniors (age 65+) with inflammatory bowel disease (IBD) and those on anti-TNF therapy. A challenge of the COVID-19 era has been conveying rapid-evolving health information, especially for groups with higher risk of poor outcomes of COVID-19 such as seniors and immunocompromised individuals. AIM To develop sharable, interactive 3D models of the relationship of age and days from vaccine dose on serological responses to SARS-CoV-2 vaccines in people with IBD who are either on or not on anti-TNF therapy. METHODS A prospective cohort study of individuals with IBD (n=528) who received 1–4 doses of a COVID-19 vaccine were assessed for serological responses 1–8 weeks and 8–16 weeks following each dose using the SARS-CoV-2 IgG II Quant assay. The cohort was stratified into those on and not on anti-TNF therapy at vaccination. We created 3D models with the plotly R package. Data were plotted as days from vaccine dose on the x-axis, age at the time of serology on the y-axis, and spike-protein antibody levels on the z-axis. Linear regressions were run with Anti-Spike IgG levels (AU/mL) as the outcome, and days from vaccine and age at serology as continuous exposures. Results were visualized as regression planes in 3D models. To share our visualizations, we built an open-access, online, interactive dashboard (https://covid19-sero.shinyapps.io/dashboard/). RESULTS The dashboard allows knowledge users to make selections based on desired data stratifications, and then zoom, pan, and rotate the plot to inspect the data or download an image of the plot they generate. Post-2nd dose samples comprised the largest dose subset in our study (n=658). Our model for post-2nd dose data shows that antibody levels significantly decrease by 20.9 (AU/mL) per day from a dose (p<0.001) and by 116.6 (AU/mL) per year increase in age (p=0.002) (Figure 1). As age increases, antibody levels significantly decrease by 107.5 (AU/mL) for those on anti-TNFs (p=0.026) and by 124.8 (AU/mL) for those not on anti-TNFs (p=0.002) (Figure 2). Those not on anti-TNFs experienced a decrease of 31.1 (AU/mL) (p<0.001) per day after 2nd vaccine dose; we did not find a significant decrease over time for those on anti-TNFs (p=0.927) (Figure 2). DISCUSSION In people with IBD, advancing age is associated with lower antibody responses to SARS-CoV-2 vaccines; however, antibody response does not significantly decrease over time for those on anti-TNF therapy. Integrating visualizations of quantitative data with participant demographics and medical therapy in open-access, interactive web applications promotes rapid release of research findings in a format that is accessible by a maximum number of knowledge users, which is of continued importance for those with a high risk of complications resulting from COVID-19. Figure 1 A scatterplot with post-2nd dose data for all possible data stratifications, downloaded from our online dashboard. The plot depicts decreasing trends in Anti-Spike IgG levels (AU/mL) as days from dose and participant age increase. Figure 2 Scatterplots with post-2nd dose data comparing Anti-Spike IgG levels (AU/mL) for participants either on or not on anti-TNF therapy. Plots depict lower antibody levels for individuals with IBD on anti-TNF therapy. Panel A shows no change in antibody levels for those on anti-TNF therapy as days from vaccine dose increase and Panel B shows that antibody levels in those not on anti-TNF therapy decrease as both days from vaccine dose and age increase.

Rotavirus in Water Environments: A Systematic Review and Meta-Analysis

Background: Despite the adoption of rotavirus vaccines, sporadic outbreaks of the virus have been reported in many parts of the world. These outbreaks are facilitated by several factors including the ease of transmission of rotavirus through water environments. This systematic review aimed to determine the global prevalence of rotavirus in water environments. Methodology: Comprehensive Boolean searches were conducted in PubMed, SCOPUS, and Web of Science. A total of 75 eligible studies were included in the study, from which data was extracted for both systematic review and meta-analysis. Extracted prevalence data was grouped according to six water categories: drinking water, untreated sewage, treated sewage, surface water, groundwater, and others. A single-group prevalence meta-analysis was conducted in RStudio version 4.3.3 subjecting the data to the random-effects model. Results: The included studies were conducted in 32 countries that span 5 continents: Africa, Asia, Europe, North America, and South America. The pooled prevalence of rotavirus in water environments was 40.86%. Among the individual water environments, untreated sewage had the highest prevalence (68.27%), followed by treated sewage (53.07%), surface water (33.40%), groundwater (25.64%) and drinking water (9.46%). Continental stratification of the prevalence data was as follows: Africa (51.75%), Asia (32.48%), Europe (55.90%), North America (41.80%), and South America (28.51%). Conclusion: There is a high prevalence of rotavirus in water environments, especially in untreated sewage, and in Europe. Further research is needed to find more efficient methods that can effectively eliminate rotavirus to insignificant levels in water environments.

Short-term ambient ozone exposure increases the risk of hospitalization with depression: a multi-city time-stratified case–crossover study

Background Depression, the most common mental illness worldwide, has been studied and air pollution has been found to increase the risk of depression hospitalization, but research results on ozone (O3) remain limited. In this context, we investigated the relationship between short-term O3 exposure and depression-related hospital admissions (HAs). Methods The 10,459 records of HAs for depression from medical institutions across in 9 cities, China, were collected between 1 January 2017, and 31 December 2018. Air pollutants and meteorological data was obtained from provincial ecological environment monitoring stations in the study area. Conditional Poisson regression was employed to estimate the association between O3 and hospitalizations for depression, with data stratification by sex, age, weather, and economic level. Results Short-term O3 exposure was positively associated with the number of depression-related hospitalizations (Relative risk: 1.04 [95% CI: 1.02, 1.05]). O3 had a significant effect on the risk of depression-related hospitalizations on warm days (P = 0.021, Relative risk: 1.05 [1.03, 1.08]). The high gross domestic product group was more likely to be affected by O3 exposure-associated depression-related hospitalizations (P = 0.005, Relative risk: 1.03 [1.01, 1.05]). Conclusions Short-term changes to O3 exposure may increase the risk of depression related hospitalizations, especially on warm days.

No Improvement in the Age Adjusted Mortality of Older Adults with Acute Lymphoblastic Leukemia Despite the Availability of Blinatumumab, Inotuzumab and Novel TKIs : A 1999-2020 Analysis in the United States

Introduction: Acute Lymphoblastic Leukemia (ALL) has a bimodal distribution and while it is the most common malignancy in children, it can also present in adults 55 years of age and older. Adverse disease biology and comorbidities that preclude delivering curative regimens are factors that underlie the markedly disparate outcomes seen between childhood and older adult ALL. Advancements in therapy have improved prognosis in children, however in adults' treatment outcomes are worse. Our study aims to compare mortality trends in older adults compared to children within the United States with ALL. Methods: Using the Center for Disease Control and Prevention Wide-Ranging Online Data for Epidemiologic Research database, we reviewed death certificate data for ALL patients from 1999 to 2020, age 55 and older in the United States. ICD code C91.0 (Acute lymphoblastic leukemia - Malignant neoplasms) was used to identify patients with ALL. Age-adjusted mortality rates (AAMRs) were calculated per 1,000,000 people and stratified patients by sex, race/ethnicity, and census region (Northeast, Midwest, South, and West). A comparison group between ages 0 and 15 was added with an identical stratification of sub-group data. Results: In the United States from 1999 to 2020 there were 17,238 deaths in ALL patients 55 and older. For this population there was not an improvement in mortality over the last twenty years (10.8, AAMR 1999) and (10.6, AAMR 2020). Further subgroup analysis did not reveal a significant difference over the years by reported sex, race, and US census region. Men and hispanic patients had the highest AAMR consistently, 12 and 13.6 in 2020. There was a statistically significant improvement in ages 0-15 for overall AAMR, 3.5 in 1999 to 2.2 in 2020 (Table 1). Conclusion: There has not been a significant improvement in AAMR for adults 55+ with ALL despite improvements seen in the pediatric patient population. Over the last twenty years AAMR in the older adult population has unfortunately been stagnant. This is despite the approval of several new therapies for adult patients with ALL including blinatumumab, inotuzumab, and the increased recognition of the importance of targeted tyrosine kinase inhibitor-based therapy for patients with Ph+ ALL, which is certainly much more prevalent in the older adult ALL population. The lack of AAMR despite the numerous seemingly therapeutic advances made in the adult population underscores the need to specifically study new therapeutics, their outcomes, utilization, and toxicities, in the sub-population of older adults with ALL in order to understand the lack of collective benefit seen in this unique patient population. Continued monitoring of the AAMR trends in this patient population will soon begin to reflect the availability of CART therapy.

The BioRef Infrastructure, a Framework for Real-Time, Federated, Privacy-Preserving, and Personalized Reference Intervals: Design, Development, and Application.

Reference intervals (RIs) for patient test results are in standard use across many medical disciplines, allowing physicians to identify measurements indicating potentially pathological states with relative ease. The process of inferring cohort-specific RIs is, however, often ignored because of the high costs and cumbersome efforts associated with it. Sophisticated analysis tools are required to automatically infer relevant and locally specific RIs directly from routine laboratory data. These tools would effectively connect clinical laboratory databases to physicians and provide personalized target ranges for the respective cohort population. This study aims to describe the BioRef infrastructure, a multicentric governance and IT framework for the estimation and assessment of patient group-specific RIs from routine clinical laboratory data using an innovative decentralized data-sharing approach and a sophisticated, clinically oriented graphical user interface for data analysis. A common governance agreement and interoperability standards have been established, allowing the harmonization of multidimensional laboratory measurements from multiple clinical databases into a unified "big data" resource. International coding systems, such as the International Classification of Diseases, Tenth Revision (ICD-10); unique identifiers for medical devices from the Global Unique Device Identification Database; type identifiers from the Global Medical Device Nomenclature; and a universal transfer logic, such as the Resource Description Framework (RDF), are used to align the routine laboratory data of each data provider for use within the BioRef framework. With a decentralized data-sharing approach, the BioRef data can be evaluated by end users from each cohort site following a strict "no copy, no move" principle, that is, only data aggregates for the intercohort analysis of target ranges are exchanged. The TI4Health distributed and secure analytics system was used to implement the proposed federated and privacy-preserving approach and comply with the limitations applied to sensitive patient data. Under the BioRef interoperability consensus, clinical partners enable the computation of RIs via the TI4Health graphical user interface for query without exposing the underlying raw data. The interface was developed for use by physicians and clinical laboratory specialists and allows intuitive and interactive data stratification by patient factors (age, sex, and personal medical history) as well as laboratory analysis determinants (device, analyzer, and test kit identifier). This consolidated effort enables the creation of extremely detailed and patient group-specific queries, allowing the generation of individualized, covariate-adjusted RIs on the fly. With the BioRef-TI4Health infrastructure, a framework for clinical physicians and researchers to define precise RIs immediately in a convenient, privacy-preserving, and reproducible manner has been implemented, promoting a vital part of practicing precision medicine while streamlining compliance and avoiding transfers of raw patient data. This new approach can provide a crucial update on RIs and improve patient care for personalized medicine.

A Quasi Experimental Study of Helicobacter Pylori Eradication in Afghanistan: Comparison of Treatment Preferences for Helicobacter Pylori Eradications, Sequential Therapy Versus Hybrid Therapy

Colonization by Helicobacter pylori (H. pylori) of the gastric epithelial lining epitomizes a pervasive and clinically consequential infectious etiology. Therapeutic strategies aimed at bacterial eradication inherently hinge on a confluence of determinants-regional antimicrobial resistance spectra, localized clinical algorithms, and pharmaceutical attainability. Best practices in the medical milieu underscore sagacious antibiotic stewardship, predicated on a patient-centric antibiotic historiography prior to therapeutic commencement.
 The overarching objective of this study-executed in a quasi-experimental comparative framework at Sheikh Zayed University Teaching Hospital, Khost, Afghanistan-was to discerningly evaluate and contrast two paradigms of treatment: Sequential Therapy and Hybrid Therapy. The subject pool encapsulated a demographically diverse set of individuals, aged 16 to 80, manifesting symptoms of peptic ulcer diseases or gastritis. Those ineligible for inclusion included follow-up patients, immunocompromised subjects, and individuals undergoing oncological interventions.
 Employing the Statistical Package for the Social Sciences (SPSS) Version 25 for data stratification, the Hybrid Therapy arm was subjected to an initial seven-day treatment window with amoxicillin (1g, BID) and omeprazole (20mg, BID), succeeded by a seven-day regimen involving clarithromycin (500mg, BID) and metronidazole (400mg, TID). Conversely, the Sequential Therapy cohort received a bifurcated, five-day course of omeprazole and amoxicillin, followed by an additional five-day regimen of omeprazole, metronidazole, and clarithromycin. Outcome variables, inclusive of treatment efficaciousness, regimen completion, and patient adherence, were rigorously evaluated via Chi-Square statistical testing.
 Encompassing 145 patients, the study observed a gender dispersion of 156 males (41.1%) to 224 females (58.9%). Per-protocol analysis unveiled a statistically significant eradication rate-86.8% vs 83.0% (P=0.021)-tilted in favor of Hybrid Therapy. Both Modified Intention-to-Treat and Intention-to-Treat analyses corroborated this proclivity, registering significant eradication rates of 92.1% versus 67% (P=0.031) for Hybrid Therapy.
 Conclusively, Hybrid Therapy demonstrated a superior safety and efficacy profile vis-à-vis Sequential Therapy in extirpating H. pylori, intimating its plausible preferability for the clinical management of associated peptic ulcer diseases and gastritis.

Achieving health equity: A patient safety imperative

Achieving health equity requires achieving zero preventable harm for all people. Health inequity must be inextricably linked to safety if all people are to be free from harm. There is no safety without equity and no equity without safety. Health inequity is an unsafe condition. Equity-related near misses, adverse events and sentinel events must undergo a comprehensive systematic analysis. Each root cause should have a strong corrective action(s). Measurable actions might include the collection and stratification of race, ethnicity and language data, or addressing stereotype bias, implicit bias, structural competency and institutional and structural racism. Leadership committed to creating a culture of equity is required. Measures of success should be linked to payment and restoring trust.

CSI: Contrastive data Stratification for Interaction prediction and its application to compound-protein interaction prediction.

Accurately predicting the likelihood of interaction between two objects (compound-protein sequence, user-item, author-paper, etc.) is a fundamental problem in Computer Science. Current deep-learning models rely on learning accurate representations of the interacting objects. Importantly, relationships between the interacting objects, or features of the interaction, offer an opportunity to partition the data to create multi-views of the interacting objects. The resulting congruent and non-congruent views can then be exploited via contrastive learning techniques to learn enhanced representations of the objects. We present a novel method, Contrastive Stratification for Interaction Prediction (CSI), to stratify (partition) a dataset in a manner that can be exploited via Contrastive Multiview Coding to learn embeddings that maximize the mutual information across congruent data views. CSI assigns a key and multiple views to each data point, where data partitions under a particular key form congruent views of the data. We showcase the effectiveness of CSI by applying it to the compound-protein sequence interaction prediction problem, a pressing problem whose solution promises to expedite drug delivery (drug-protein interaction prediction), metabolic engineering, and synthetic biology (compound-enzyme interaction prediction) applications. Comparing CSI with a baseline model that does not utilize data stratification and contrastive learning, and show gains in average precision ranging from 13.7% to 39% using compounds and sequences as keys across multiple drug-target and enzymatic datasets, and gains ranging from 16.9% to 63% using reaction features as keys across enzymatic datasets. Code and dataset available at https://github.com/HassounLab/CSI.

Predictive Risk Factors for Childbirth-Associated Breast Infections in the United States: A 10-Year Perspective.

Infectious mastitis is a common condition that affects up to 33% of lactating women. Several risk factors have been suggested to be strongly associated with breast abscess, nipple infection, and non-purulent mastitis associated with childbirth. In this retrospective cohort study, we gathered data from the National Inpatient Sample (NIS) between 2005 and 2014 and utilized data stratification and backward linear regression to analyze the predictive factors associated with patients hospitalized with breast infection after childbirth, with special consideration of risk factors affecting hospital length of stay (LOS). In the ten-year period, 4614 women were hospitalized with a primary diagnosis of breast abscess, nipple infection, or non-purulent mastitis associated with childbirth. Mean (SD) age was 26.75 (6) years. The highest frequency distribution of cases was observed in patients aged 22-30 years (49.82%). Mean (SD) LOS was 2.83 (1.95) days. Mean (SD) LOS in patients with procedure was 3.53 (2.47) days, which was significantly longer than that in those with no procedure (2.39 (1.36) days, p < 0.001). Primary diagnosis of breast abscess and occurrence of a hospital procedure were most significantly associated with prolonged LOS. Factors such as age, socioeconomic position, severity of functional loss, as well as comorbidities were also contributing risk factors to the development of breast infection and increased hospital LOS. Further studies should examine these findings, as they relate to breastfeeding practices and concentrate on establishing best practices for risk reduction and prevention of childbirth-associated breast and nipple infections and hospitalizations.

Correlation between hypothyroidism and type 2 diabetes: A cross-sectional study

Diabetes is the most common chronic endocrine disease. Numerous studies suggest that hypothyroidism is associated with hypertension, high cholesterol, and abnormal homocysteine level and patients with hypothyroidism have a higher risk of metabolic syndrome. This prospective study was conducted at Lahore General Hospital to estimate the frequency of hypothyroidism in patients with T2DM. Objectives: To determine the frequency of hypothyroidism in patients with type 2 diabetes mellitus. Study design: This was a cross-sectional study. Setting: Department of Medicine, Lahore General Hospital, Lahore. Duration of study: Study duration was six months from 06-12-2019 to 05-05-2020. Subjects: Hundred patients through Non-probability consecutive sampling. Methods: Patients fulfilling inclusion and exclusion criteria were included. Detailed history and examination and Free Serum Thyroxine (T4), Thyroid Stimulating Hormone (TSH) levels were measured in all patients using standard immunoassay in the hospital laboratory. Data was analysed using SPSS 21. Results: Among 100 patients, 64% patients were male and 58% patients were between ages of 41-60 years with mean age of 46.6 + 4.7 years. Regarding blood sugar levels, 62% patients had &lt; 200 mg/dl whereas remaining patients had blood sugar levels of &gt; 200 mg/dl. Results of study showed that 12% of patients had hypothyroidism. Further stratification of data revealed no significant association between different age groups, gender, duration of diabetes, high BMI and presence of hypothyroidism i.e. p value was 0.9801, 0.9081, 0.8825 and 0.9601 respectively. Conclusion: Frequency of hypothyroidism in patients with type 2 diabetes mellitus is high.

Calibration for iron redox state and oxygen fugacity in silicate glasses from x-ray absorption spectroscopy

Prior work to quantify glass Fe-redox ratios in situ using x-ray absorption spectroscopy (XAS) has relied heavily on the pre-edge region. The pre-edge-based methods do not take advantage of other information in the spectra above the pre-edge, up to and past the main edge. This study updates the previous calibration (Dyar et al., 2016a) with additional experimental glasses spanning a wide range of geologically relevant compositions (SiO2 weight % = 38.5–78.0), equilibrated over a range of oxygen fugacity (fO2) values from reducing (IW-2) to oxidizing (air), for which Fe XAS analysis has been collected for the whole spectral region (pre-edge, main edge, EXAFS). The resulting dataset was used to evaluate multiple multivariate statistical models for their ability to accurately predict log fO2 or %Fe3+. Accuracies are estimated using cross-validated root mean square error, for which purpose unseen test data are held out from the regressions. Datasets covering small compositional ranges were tested and compared with datasets covering broad compositional ranges. These ‘stratified’ sets improved predictive accuracy in some cases; however, the extra complications associated with data stratification did not justify the small improvement in accuracy. Overall, a simple partial least squares (PLS) model based on all data collected over the entire XAS region was found to result in a universal calibration for determination of log fO2 and %Fe3+ in glasses.

Measurement Strategies for The Joint Commission Health Care Disparities Standard: Defining Hospital-Based Requirements—Part 1

Measurement Strategies for The Joint Commission Health Care Disparities Standard: Defining Hospital-Based Requirements—Part 1

Advantages of selective ischemia and indocyanine green fluorescence navigation in laparoscopic partial nephrectomy: preliminary results

An important stage of the nephron-sparing surgery for the treatment of kidney tumors is the temporary creation of thermal ischemia, which can lead to a further decrease in kidney function, therefore this stage needs to be optimized. The aim of our work was to investigate the benefits of using segmental ischemia using intraoperative indocyanine green (ICG) navigation for kidney resection in patients with kidney tumors. The prospective study included 73 patients diagnosed with a kidney tumor who underwent partial nephrectomy (PN). Patients were divided into three groups, according to the use of the thermal ischemia method: total ischemia by clamping the main renal artery (n=33), segmental ischemia by clamping the segmental artery with the use of ICG navigation (n=26), and without it (n=16). Demographic, physiological, and perioperative data were analyzed for all patients. Before and on the fourth day after surgery, the glomerular filtration rate was evaluated according to the level of blood creatinine – eGFR (calculated glomerular filtration rate (eGFR). Multiple linear regression was used to model the effect of variables on change in eGFR. The data testify to the significant index difference in eGFR level change on the fourth postoperative day in the groups of total and segmental ischemia in favor of the latter (p=0.001). A decrease in eGRF was associated with the length of hospital stay (p=0.021). The data stratification in tumor size showed that it is the most important factor defining eGFR level change during total or segmental ischemia application. Renal tumor localization influences eGFR level change during the postoperative period for a maximum diameter of &lt;40 mm tumors. However, for &gt;40 mm diameter tumors, the statistically significant index of eGFR deviation was not found. At the same time, in the patient group with &gt;40 mm tumors, there was a notable difference in eGFR deviation among the groups of total and segmental ischemia (about 15 mL/min/1.73 m2). In the early postoperative period (up to 4 days) after partial nephrectomy using selective warm ischemia, less pronounced eGFR deviations are observed compared to total warm ischemia. Tumor size and its localization are the risk factors for eGFR level decline in the early postoperative period. We did not find any statistically significant differences in intra- and early postoperative indexes for groups II and III patients where segmental ischemia was applied either with indocyanine green fluorescence imaging or without it.

Factors Influencing the Spatial Distribution of Regulating Agro-Ecosystem Services in Agriculture Soils: A Case Study of Slovakia

Agro-ecosystem services assessment and mapping are one of the main requirements for implementing the concept of ecosystem services into institutional decision-making within the European Union and worldwide. The aim of this study was to identify the most important regional drivers of the natural potential of regulating ecosystem services with agricultural soil in Slovakia, resulting in an original set of macro-scale spatial stratification criteria for agricultural land, and to estimate typical middle values of its potential in newly defined regions. The non-monetary assessment was based on natural environment parameters and land use factors. For the assessment of individual indicators of agro-ecosystem services, we utilized an expert scoring system. We evaluated ecosystem services on the basis of biophysical indicators that determine the corresponding soil functions and are part of the existing databases available in the Slovak Republic. The new methodological combination enabled us to provide unique mapping and assessment of agro-ecosystem services within Slovakia. Regional climate, land cover, and soil slope were identified as key factors impacting agro-ecosystem services potential within the country, which can be used as data stratification levels for further analyses. Linking the value of individual regulating agro-ecosystem services potential with geographical distribution can help to optimize its potential depending on the needs of the inhabitants living in different regions by introducing appropriate measures and can contribute to effective agricultural policymaking.

The Outcome Of The Inverted Internal Limiting Membrane Flap Technique For The Repair Of Large Idiopathic Macular Holes.

Full-thickness macular hole is defined as an anatomical defect in the fovea that spans from the internal limiting membrane to the retinal pigment epithelium, assessed by spectral domain optical coherence tomography. The Objectives of the study are to determine the anatomical and visual outcome in patients undergoing pars plana vitrectomy along with inverted internal limiting flap closure in large idiopathic full-thickness macular holes (>400 μm). A prospective interventional study was conducted at a tertiary teaching eye hospital in Karachi, where patients of either gender and having macular holes greater than >400 μm were recruited. The study was conducted From January 9 to July 8, 2022, and all patients underwent pre-operative fundus examination and pars plana vitrectomy with inverted ILM flap closure. Data was entered and analyzed using SPSS 23. Follow-ups were conducted at 1 and 3 months. A total of 94 patients were enrolled with a mean age of 49.17±13.8 years. The mean duration of symptoms was 3.1±1.4 months. The mean pre-operative macular hole diameter was 854.31±08.36 μm and Stage 3 and 4 MH was present in 36.2% and 63.8% of patients, respectively. Anatomical closure was achieved in 93.6% of eyes (n=88/94). Pre-operative mean BCVA was LogMAR 0.90±0.24, which improved to LogMAR mean 0.70±0.27 at the final follow-up. As of the last follow-up, 92.6% of patients showed improved visual outcomes, with a mean three-line improvement in Snellen lines. After data stratification, no statistically significant result was obtained. The use of the inverted ILM flap technique resulted in improved anatomical and visual outcomes, in cases of large idiopathic macular holes.

On Information Granulation via Data Filtering for Granular Computing-Based Pattern Recognition: A Graph Embedding Case Study

Granular Computing is a powerful information processing paradigm, particularly useful for the synthesis of pattern recognition systems in structured domains (e.g., graphs or sequences). According to this paradigm, granules of information play the pivotal role of describing the underlying (possibly complex) process, starting from the available data. Under a pattern recognition viewpoint, granules of information can be exploited for the synthesis of semantically sound embedding spaces, where common supervised or unsupervised problems can be solved via standard machine learning algorithms. In this companion paper, we follow our previous paper (Martino et al. in Algorithms 15(5):148, 2022) in the context of comparing different strategies for the automatic synthesis of information granules in the context of graph classification. These strategies mainly differ on the specific topology adopted for subgraphs considered as candidate information granules and the possibility of using or neglecting the ground-truth class labels in the granulation process and, conversely, to our previous work, we employ a filtering-based approach for the synthesis of information granules instead of a clustering-based one. Computational results on 6 open-access data sets corroborate the robustness of our filtering-based approach with respect to data stratification, if compared to a clustering-based granulation stage.

Similarity and variability of blocked weather-regime dynamics in the Atlantic–European region

Abstract. Weather regimes govern an important part of the sub-seasonal variability of the mid-latitude circulation. Due to their role in weather extremes and atmospheric predictability, regimes that feature a blocking anticyclone are of particular interest. This study investigates the dynamics of these “blocked” regimes in the North Atlantic–European region from a year-round perspective. For a comprehensive diagnostic, wave activity concepts and a piecewise potential vorticity (PV) tendency framework are combined. The latter essentially quantifies the well-established PV perspective of mid-latitude dynamics. The four blocked regimes (namely Atlantic ridge, European blocking, Scandinavian blocking, and Greenland blocking) during the 1979–2021 period of ERA5 reanalysis are considered. Wave activity characteristics exhibit distinct differences between blocked regimes. After regime onset, Greenland blocking is associated with a suppression of wave activity flux, whereas Atlantic ridge and European blocking are associated with a northward deflection of the flux without a clear net change. During onset, the envelope of Rossby wave activity retracts upstream for Greenland blocking, whereas the envelope extends downstream for Atlantic ridge and European blocking. Scandinavian blocking exhibits intermediate wave activity characteristics. From the perspective of piecewise PV tendencies projected onto the respective regime pattern, the dynamics that govern regime onset exhibit a large degree of similarity: linear Rossby wave dynamics and nonlinear eddy PV fluxes dominate and are of approximately equal relative importance, whereas baroclinic coupling and divergent amplification make minor contributions. Most strikingly, all blocked regimes exhibit very similar (intra-regime) variability: a retrograde and an upstream pathway to regime onset. The retrograde pathway is dominated by nonlinear PV eddy fluxes, whereas the upstream pathway is dominated by linear Rossby wave dynamics. Importantly, there is a large degree of cancellation between the two pathways for some of the mechanisms before regime onset. The physical meaning of a regime-mean perspective before onset can thus be severely limited. Implications of our results for understanding predictability of blocked regimes are discussed. Further discussed are the limitations of projected tendencies in capturing the importance of moist-baroclinic growth, which tends to occur in regions where the amplitude of the regime pattern, and thus the projection onto it, is small. Finally, it is stressed that this study investigates the variability of the governing dynamics without prior empirical stratification of data by season or by type of regime transition. It is demonstrated, however, that our dynamics-centered approach does not merely reflect variability that is associated with these factors. The main modes of dynamical variability revealed herein and the large similarity of the blocked regimes in exhibiting this variability are thus significant results.

Evaluation of Structured, Semi-Structured, and Free-Text Electronic Health Record Data to Classify Hepatitis C Virus (HCV) Infection

Evaluation of the United States Centers for Disease Control and Prevention (CDC)-defined HCV-related risk factors are not consistently performed as part of routine care, rendering risk-based testing susceptible to clinician bias and missed diagnoses. This work uses natural language processing (NLP) and machine learning to identify patients who are at high risk for HCV infection. Models were developed and validated to predict patients with newly identified HCV infection (detectable RNA or reported HCV diagnosis). We evaluated models with three types of variables: structured (structured-based model), semi-structured and free-text notes (text-based model), and all variables (full-set model). We applied each model to three stratifications of data: patients with no history of HCV prior to 2020, patients with a history of HCV prior to 2020, and all patients. We used XGBoost and ten-fold C-statistic cross-validation to evaluate the generalizability of the models. There were 3564 unique patients, 487 with HCV infection. The average C-statistics on the structured-based, text-based, and full-set models for all the patients were 0.777 (95% CI: 0.744–0.810), 0.677 (95% CI: 0.631–0.723), and 0.774 (95% CI: 0.735–0.813), respectively. The full-set model performed slightly better than the structured-based model and similar to text-based models for patients with no history of HCV prior to 2020; average C-statistics of 0.780, 0.774, and 0.759, respectively. NLP was able to identify six more risk factors inconsistently coded in structured elements: incarceration, needlestick, substance use or abuse, sexually transmitted infections, piercings, and tattoos. The availability of model options (structured-based or text-based models) with a similar performance can provide deployment flexibility in situations where data is limited.

Using hierarchical text classification to investigate the utility of machine learning in automating online analyses of wildlife exploitation

Expanding digital data sources, including social media, online news articles and blogs, provide an opportunity to understand better the context and intensity of human-nature interactions, such as wildlife exploitation. However, online searches encompassing large taxonomic groups can generate vast datasets, which can be overwhelming to filter for relevant content without the use of automated tools. The variety of machine learning models available to researchers, and the need for manually labelled training data with an even balance of labels, can make applying these tools challenging. Here, we implement and evaluate a hierarchical text classification pipeline which brings together three binary classification tasks with increasingly specific relevancy criteria. Crucially, the hierarchical approach facilitates the filtering and structuring of a large dataset, of which relevant sources make up a small proportion. Using this pipeline, we also investigate how the accuracy with which text classifiers identify relevant and irrelevant texts is influenced by the use of different models, training datasets, and the classification task. To evaluate our methods, we collected data from Facebook, Twitter, Google and Bing search engines, with the aim of identifying sources documenting the hunting and persecution of bats (Chiroptera). Overall, the ‘state-of-the-art’ transformer-based models were able to identify relevant texts with an average accuracy of 90%, with some classifiers achieving accuracy of >95%. Whilst this demonstrates that application of more advanced models can lead to improved accuracy, comparable performance was achieved by simpler models when applied to longer documents and less ambiguous classification tasks. Hence, the benefits from using more computationally expensive models are dependent on the classification context. We also found that stratification of training data, according to the presence of key search terms, improved classification accuracy for less frequent topics within datasets, and therefore improves the applicability of classifiers to future data collection. Overall, whilst our findings reinforce the usefulness of automated tools for facilitating online analyses in conservation and ecology, they also highlight that the effectiveness and appropriateness of such tools is determined by the nature and volume of data collected, the complexity of the classification task, and the computational resources available to researchers.

Obesity Associated With Increased Mortality and Hospital Length of Stay in Trauma Laparotomy Patients.

Although obesity in trauma patients is accepted as a risk factor for postoperative complications, recent literature offers conflicting evidence regarding the effect of body mass index (BMI) on mortality in trauma patients undergoing laparotomy. To address this question, we examined the patient population of a Level 1 Trauma Center during a 3-year period to compare mortality rates and other outcomes between BMI groups undergoing laparotomy. Through retrospective chart review of electronic medical records, with subsequent stratification of data based on BMI, we found that mortality, injury severity score, and hospital length of stay all increase significantly with each incremental increase in BMI class. From these data, we concluded that higher BMI class leads to greater morbidity and mortality in trauma patients undergoing laparotomy at this institution.