Statistical Significance Tests Research Articles

A novel approach for the replacement of accuracy and precision measurements with the visual instrument agreement scale (VIAS) in evaluating the performance of four intraoral scanners and one shade measurement device

ObjectivesThe terms 'accuracy' and 'precision' are tightly defined in color science but are often used ambiguously in dental research. This study introduces the visual instrument agreement scale (VIAS), a new method for determining visual-instrumental agreement in dental colorimetry by comparing visually perceived and measured color differences. Materials and MethodsIn-vivo tooth color measurements were taken from 16 participants using four intraoral scanners (Primescan, Medit i700, Carestream CS3700, Trios 3) and one spectrophotometer (Vita Easyshade V). Visual shade assessment was also performed by one expert observer using the 3D Master shade guide. Statistical significance testing was conducted using the STRESS index to calculate VIAS, and the F-statistic was used to evaluate device performance. ResultsCarestream CS3700 achieved the highest visual-instrumental agreement with a VIAS score of 82%, performing significantly better than the other devices. Primescan, Medit i700, and Trios 3 showed scores of 76%, 75%, and 72%, respectively, with no significant differences between them. Vita Easyshade V scored 57%, performing significantly worse than the other devices. ConclusionsThe overall performance of the intraoral scanners was strong, with Carestream CS3700 approaching excellent performance. The VIAS method offers a practical, color science-based framework for evaluating visual-instrumental agreement and can be easily replicated using the freely available toolbox. Clinical SignificanceIntraoral scanners performed surprisingly better than a spectrophotometer specifically designed for tooth color measurement and which is often regarded as the gold standard. Additionally, VIAS offers a new, scientifically grounded approach for testing visual-instrumental agreement in dental colorimetry.

Scalable test of statistical significance for protein-DNA binding changes with insertion and deletion of bases in the genome

Scalable test of statistical significance for protein-DNA binding changes with insertion and deletion of bases in the genome

Assessing CMIP Models’ Ability to Detect Observed Surface Warming Signals Related to Climate Change

Abstract Assessing if CMIP models can detect observed climate change signals is crucial for evaluating their realism and strengthening confidence in future projections. These signals can only be compared once distinguishable from the background climate “noise.” Built for impact studies but not detection purpose, the time of emergence (ToE) estimates when this occurs using a fixed signal-to-noise ratio. To more accurately assess models’ ability to reproduce observed climate change signals, we introduce the time of detection (ToD), which employs a statistical significance test of the nonlinear trend, accounting for the time series length. ToD and ToE are computed for sea surface temperature (SST) and relative SST (RSST) across four observational and 69 CMIP5/6 historical and shared socioeconomic pathways 5–8.5 (SSP5–8.5) scenario datasets. ToD precedes ToE and is less sensitive to the chosen threshold. While already detectable in most of the tropics, SST warming in the equatorial Pacific is detectable in ∼90% of models but only one out of four observational datasets, due to a stronger modeled warming. Because RSST signals are weaker than SST signals, they do not emerge anywhere but are detectable in two regions. The enhanced warming in the western Indian Ocean, detected in ∼45% of models, is not robust in observations when excluding pre-1960 dubious data. Conversely, the subdued southeast Pacific warming detected in observations and ∼80% of models is robust and consistent with a poleward extension of the Southern Hemisphere Hadley cell. The ToD’s ability to detect subtle signals allows for assessing the realism of the CMIP warming pattern in this region. Significance Statement Recent studies emphasize the importance of the climate change warming pattern and its apparent discrepancy between climate models and observations in the equatorial Pacific. We argue that our newly proposed time of detection (ToD) is better suited than the well-known time of emergence (ToE) for investigating where modeled and observed warming patterns can be confidently compared. In agreement with previous studies, our analysis indicates model–observations disagreement over the equatorial Pacific. Using ToD instead of ToE allows for detecting a robust subdued warming (i.e., weaker than the tropical average) in the southeast Pacific and therefore confirms the realism of the CMIP warming pattern in this region.

DIFFERENTIATED IN VITRO LYSOZYME ACTIVITY IN AVES AND MAMMALIA IN RESPONSE TO SEABUCKTHORN (HIPPOPHAE RHAMNOIDES) STIMULATION

Lysozyme, an intrinsic component of the immune system, is a naturally occurring enzyme with antimicrobial activity (Khorshidian et al., 2022), by hydrolyzing the muramyl dipeptide in the bacteria cell wall. Considered to be an endogenous antibiotic, it differs by species (Ferraboschi et al., 2021). Some of the medicinal plants were cited to inhibit the anti-lysozyme activity and biofilm formation by bacteria (Bukharin et al., 2003). The aim of this research was to evaluate the differences in the in vitro activity of lysozyme between the classes Aves and Mammalia when treated with a protein-carotenoid extract of Hypopphae rhamnoides compared to well-known immune modulating preparations (selenium or selenium and copper compounds). The investigations were carried out on serum samples from: a) commercial broiler chickens aged 34 days (n = 19) and b) 5-month-old Supercunirom breed male rabbits (n = 19). The agar gel radial immune diffusion method and the Micrococcus lysodeicticus test strain were used to define the in vitro lysozyme activity. The sera were mixed with serial dilutions (1:2, 1:4, etc.) of the tested compounds. The groups were compared by Student’s t test for statistical significance of the results. He increase in activity (%) versus control were calculated. Sea buckthorn extract significantly (t= 7.22, p < 0.001) decreased the in vitro activity of lysozyme at both dilutions used (1:2, 1:4). The concentration of serum lysozyme was higher in rabbits than in chickens and its lytic activity was enhanced by selenium and copper combinations in chickens (183.69 ± 37.91%) and less in rabbits (128.45 ± 84.10%) in a dose dependent manner. At lower dilutions (3:4), the lysozyme activity remained below that of the control treated with saline. The protein-carotenoid extract of sea buckthorn acted inhibiting on lysozyme activity, proving the need for tailored extraction and treatment protocols depending on the bacteria and host species.

Drug risks associated with sarcopenia: a real-world and GWAS study

IntroductionDrug-induced sarcopenia has not received adequate attention. Meanwhile, there is growing recognition of the importance of effective pharmacovigilance in evaluating the benefits and risks of medications.AimsThe primary aim of this study is to investigate the potential association between drug use and sarcopenia through an analysis of adverse event reports from the Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS) and to evaluate the genetic factors contributing to drug-induced sarcopenia using summary-data-based Mendelian randomization (SMR).MethodsWe obtained reports of adverse drug reactions from FAERS. Primary outcomes included sarcopenia and potential sarcopenia. We calculated the Proportional reporting ratio (PRR) to assess the risk of specific adverse events associated with various drugs, applying chi-square tests for statistical significance. Additionally, we used SMR based on Genome-wide association study (GWAS) to evaluate the potential associations between drug target genes of some significant medications and sarcopenia outcomes. The outcome data for sarcopenia included metrics as hand grip strength and appendicular lean mass (ALM).ResultsA total of 55 drugs were identified as inducing potential sarcopenia, and 3 drugs were identified as inducing sarcopenia. The top 5 drugs causing a potential risk of sarcopenia were levofloxacin (PRR = 9.96, χ2 = 1057), pregabalin (PRR = 7.20, χ2 = 1023), atorvastatin (PRR = 4.68, χ2 = 903), duloxetine (PRR = 4.76, χ2 = 527) and venlafaxine (PRR = 5.56, χ2 = 504), and the 3 drugs that had been proved to induced sarcopenia included metformin (PRR = 7.41, χ2 = 58), aspirin (PRR = 5.93, χ2 = 35), and acetaminophen (PRR = 4.73, χ2 = 25). We identified electron-transfer flavoprotein dehydrogenase (ETFDH) and protein Kinase AMP-Activated Non-Catalytic Subunit Beta 1 (PRKAB1) as the primary drug target genes for metformin, while Prostaglandin-endoperoxide Synthase 1 (PTGS1) and Prostaglandin-endoperoxide Synthase 2 (PTGS2) were considered the primary action target genes for aspirin and acetaminophen according to DrugBank database. SMR showed that the expression abundance of ETFDH was negatively correlated with right hand grip strength (blood: OR = 1.01, p-value = 1.27e-02; muscle: OR = 1.01, p-value = 1.42e-02) and negatively correlated with appendicular lean mass (blood: OR = 1.03, p-value = 7.73e-08; muscle: OR = 1.03, p-value = 1.67e-07).ConclusionsWe find that metformin, aspirin, and acetaminophen are specifically noted for their potential to induce sarcopenia based on the analyses conducted. We perform signal mining for drug-associated sarcopenia events based on real-world data and provides certain guidance for the safe use of medications to prevent sarcopenia.

Assessment of Formaldehyde’s Impact on Indoor Environments and Human Health via the Integration of Satellite Tropospheric Total Columns and Outdoor Ground Sensors

Formaldehyde (HCHO) is harmful to human health and an adequate assessment of its concentrations, both in outdoor and indoor environments, is necessary in the context of sustainable policies designed to mitigate health risks. In this research, ground indoor and outdoor HCHO measurements are integrated with the analysis of tropospheric total columns obtained by satellite surveys to assess the concentrations of HCHO in a number of environments, exploiting the proximity of a World Meteorological Organization—Global Atmosphere Watch (WMO/GAW) observation site in Calabria, Southern Italy to a National Institute for Insurance against Accidents at Work (INAIL) department in the municipality of Lamezia Terme. The meteorological parameters used by the WMO station are also used to provide additional data and test new correlations. Using statistical significance tests, this study demonstrates the presence of a correlation between indoor and outdoor HCHO concentrations, thus showing that an exchange between indoor and outdoor formaldehyde does occur. Rooms located in the local INAIL building where indoor measurements took place also demonstrate degrees of susceptibility to HCHO exposure, which are correlated with the orientation of prevailing wind corridors in the area. The new findings constitute an unprecedented characterization of HCHO hazards in Calabria and provide regulators with new tools with which to mitigate formaldehyde-related risks.

Balancing accuracy and Interpretability: An R package assessing complex relationships beyond the Cox model and applications to clinical prediction

BackgroundAccurate and interpretable models are essential for clinical decision-making, where predictions can directly impact patient care. Machine learning (ML) survival methods can handle complex multidimensional data and achieve high accuracy but require post-hoc explanations. Traditional models such as the Cox Proportional Hazards Model (Cox-PH) are less flexible, but fast, stable, and intrinsically transparent. Moreover, ML does not always outperform Cox-PH in clinical settings, warranting a diligent model validation. We aimed to develop a set of R functions to help explore the limits of Cox-PH compared to the tree-based and deep learning survival models for clinical prediction modelling, employing ensemble learning and nested cross-validation. MethodsWe developed a set of R functions, publicly available as the package “survcompare”. It supports Cox-PH and Cox-Lasso, and Survival Random Forest (SRF) and DeepHit are the ML alternatives, along with the ensemble methods integrating Cox-PH with SRF or DeepHit designed to isolate the marginal value of ML. The package performs a repeated nested cross-validation and tests for statistical significance of the ML’s superiority using the survival-specific performance metrics, the concordance index, time-dependent AUC-ROC and calibration slope.To get practical insights, we applied this methodology to clinical and simulated datasets with varying complexities and sizes. ResultsIn simulated data with non-linearities or interactions, ML models outperformed Cox-PH at sample sizes ≥ 500. ML superiority was also observed in imaging and high-dimensional clinical data. However, for tabular clinical data, the performance gains of ML were minimal; in some cases, regularised Cox-Lasso recovered much of the ML’s performance advantage with significantly faster computations. Ensemble methods combining Cox-PH and ML predictions were instrumental in quantifying Cox-PH’s limits and improving ML calibration. Traditional models like Cox-PH or Cox-Lasso should not be overlooked while developing clinical predictive models from tabular data or data of limited size. ConclusionOur package offers researchers a framework and practical tool for evaluating the accuracy-interpretability trade-off, helping make informed decisions about model selection.

Detecting Differential Item Functioning Using Response Time

This study investigated uniform differential item functioning (DIF) detection in response times. We proposed a regression analysis approach with both the working speed and the group membership as independent variables, and logarithm transformed response times as the dependent variable. Effect size measures such as Δ[Formula: see text] and percentage change in regression coefficients in conjunction with the statistical significance tests were used to flag DIF items. A simulation study was conducted to assess the performance of three DIF detection criteria: (a) significance test, (b) significance test with Δ[Formula: see text], and (c) significance test with the percentage change in regression coefficients. The simulation study considered factors such as sample sizes, proportion of the focal group in relation to total sample size, number of DIF items, and the amount of DIF. The results showed that the significance test alone was too strict; using the percentage change in regression coefficients as an effect size measure reduced the flagging rate when the sample size was large, but the effect was inconsistent across different conditions; using Δ R2 with significance test reduced the flagging rate and was fairly consistent. The PISA 2018 data were used to illustrate the performance of the proposed method in a real dataset. Furthermore, we provide guidelines for conducting DIF studies with response time.

Generalization of neural network models for complex network dynamics

Differential equations are a ubiquitous tool to study dynamics, ranging from physical systems to complex systems, where a large number of agents interact through a graph. Data-driven approximations of differential equations present a promising alternative to traditional methods for uncovering a model of dynamical systems, especially in complex systems that lack explicit first principles. A recently employed machine learning tool for studying dynamics is neural networks, which can be used for solution finding or discovery of differential equations. However, deploying deep learning models in unfamiliar settings-such as predicting dynamics in unobserved state space regions or on novel graphs-can lead to spurious results. Focusing on complex systems whose dynamics are described with a system of first-order differential equations coupled through a graph, we study generalization of neural network predictions in settings where statistical properties of test data and training data are different. We find that neural networks can accurately predict dynamics beyond the immediate training setting within the domain of the training data. To identify when a model is unable to generalize to novel settings, we propose a statistical significance test.

Examining Long-Term Influences of Frailty on Outcomes for Adults Undergoing Left Ventricular Assist Device Therapy.

Frailty is common in adults with end-stage heart failure receiving a left ventricular assist device (LVAD). Short-term studies show frailty reversal post-LVAD. Little is known about long-term frailty and how frailty relates to key LVAD outcomes, including depression, quality of life (QoL), and cognition beyond 6 months. The purpose of this study was to examine physical frailty, depression, QoL, and cognition from pre-LVAD implantation to 3, 6, and 12 months post-LVAD implantation. Clinical data and measures of physical frailty, depression, QoL, and cognition pre- and post-LVAD were extracted from an existing institutional database. Descriptive statistics and tests for statistical significance were used to describe and compare changes over time. Frailty trajectories were identified to describe physical frailty from pre-LVAD to 12 months post-LVAD. The sample (n = 46) was predominantly male (76%), with a mean age of 64.7 ± 11 years, and over half (n = 25) were physically frail. Physical frailty was reduced by nearly half at 12 months. Frailty, depression, and QoL significantly improved at all time points (P < .001). Improvements in cognition did not reach statistical significance. Recipients of LVAD without improvement in frailty exhibited worse depression, QoL, and cognition scores at 12 months compared with those not frail pre-LVAD or no longer frail post-LVAD implantation. Some recipients of LVADs experienced physical frailty reversal with benefits sustained long-term. Those without improvement recorded worse long-term patient-reported outcomes. Identifying factors that predict physical frailty response to LVAD implantation is a key area of future research.

Mammographic density and breast cancer risk among Black American women.

High mammographic density is a well-established risk factor for breast cancer; however, data from Black women are limited. It is largely unknown how mammographic density is associated with breast cancer subtypes among Black women. We examined the association between percent mammographic density (PMD) and breast cancer risk among participants in the Black Women's Health Study. Digital screening mammograms were available for 363 cases and 5541 non-cases. Cumulus software was used to assess PMD. We used inverse probability of sampling weights and Cox proportional hazards models, adjusted for age and body mass index, to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) overall and by age at mammography and estrogen receptor (ER) status of the breast tumors. Multivariable models included additional breast cancer risk factors. Tests of statistical significance were 2-sided. In simple models, women in the highest quartile of PMD had 53% increased odds of breast cancer compared to those in the lowest quartile (HR 1.53; 95% CI: 1.11, 2.11). HRs were 1.37 (95% CI: 0.83, 2.24) among women <55 years of age and 1.68 (95% CI: 1.10, 2.56) among women aged ≥55 years. HRs were 1.49 (95% CI: 1.02, 2.16) for ER+ cancer and 1.45 (95% CI: 0.73, 2.87) for ER- cancer. Associations were largely unchanged in multivariable models. In this study of U.S. Black women, higher PMD was associated with ER+ and ER- breast cancer risk. Findings from this study reinforce the importance of breast density as a risk factor for breast cancer in Black women.

Sequence-to-Sequence Models and Their Evaluation for Spoken Language Normalization of Slovenian

Sequence-to-sequence models have been applied to many challenging problems, including those in text and speech technologies. Normalization is one of them. It refers to transforming non-standard language forms into their standard counterparts. Non-standard language forms come from different written and spoken sources. This paper deals with one such source, namely speech from the less-resourced highly inflected Slovenian language. The paper explores speech corpora recently collected in public and private environments. We analyze the efficiencies of three sequence-to-sequence models for automatic normalization from literal transcriptions to standard forms. Experiments were performed using words, subwords, and characters as basic units for normalization. In the article, we demonstrate that the superiority of the approach is linked to the choice of the basic modeling unit. Statistical models prefer words, while neural network-based models prefer characters. The experimental results show that the best results are obtained with neural architectures based on characters. Long short-term memory and transformer architectures gave comparable results. We also present a novel analysis tool, which we use for in-depth error analysis of results obtained by character-based models. This analysis showed that systems with similar overall results can differ in the performance for different types of errors. Errors obtained with the transformer architecture are easier to correct in the post-editing process. This is an important insight, as creating speech corpora is a time-consuming and costly process. The analysis tool also incorporates two statistical significance tests: approximate randomization and bootstrap resampling. Both statistical tests confirm the improved results of neural network-based models compared to statistical ones.

Critical analysis of Parkinson’s disease detection using EEG sub-bands and gated recurrent unit

Parkinson’s disease, the second most common neurodegenerative disorder in the world. A progressive disease, which can worsen over time and lead to complications like mild cognitive impairments and dementia. The accurate diagnosis of Parkinson’s Disease (PD) remains a critical challenge in medical engineering. This study explores the potential of brain wave patterns for PD detection in healthy and unhealthy patients. The Power Spectral Density (PSD) and proposed PD detection model based on the Gated Recurrent Unit (GRU) is used to analyze brain activities. It was found that the detection of PD in patients was improved by classifying the RAW EEG in conjunction with the sub-bands using PSD as a feature and GRU as a classifier. The performance matrices including accuracy, precision, recall, and F1-score fall within a range of 90% to 98% for alpha, beta, and gamma sub-bands, while the area under the curve in the case of receiver operating characteristics curve achieved the maximum value of 1.00. To assess the differences between the groups with Parkinson’s disease and the healthy group, a statistical significance test was performed. The power spectral density of the two groups differed statistically significantly, according to the results, indicating that they could be useful as biomarkers for the identification of Parkinson’s disease. The results are compared and validated with the standard performance measures.

Leveraging Machine Learning for Sophisticated Rental Value Predictions: A Case Study from Munich, Germany

There has been very limited research conducted to predict rental prices in the German real estate market using an AI-based approach. From a general perspective, conventional approaches struggle to handle large amounts of data and fail to consider the numerous elements that affect rental prices. The absence of sophisticated, data-driven analytical tools further complicates this situation, impeding stakeholders, such as tenants, landlords, real estate agents, and the government, from obtaining the accurate insights necessary for making well-informed decisions in this area. This paper applies novel machine learning (ML) approaches, including ensemble techniques, neural networks, linear regression (LR), and tree-based algorithms, specifically designed for forecasting rental prices in Munich. To ensure accuracy and reliability, the performance of these models is evaluated using the R2 score and root mean squared error (RMSE). The study provides two feature sets for model comparison, selected by particle swarm optimisation (PSO) and CatBoost. These two feature selection methods identify significant variables based on different mechanisms, such as seeking the optimal solution with an objective function and converting categorical features into target statistics (TSs) to address high-dimensional issues. These methods are ideal for this German dataset, which contains 49 features. Testing the performance of 10 ML algorithms on two sets helps validate the robustness and efficacy of the AI-based approach utilising the PyTorch framework. The findings illustrate that ML models combined with PyTorch-based neural networks (PNNs) demonstrate high accuracy compared to standalone ML models, regardless of feature changes. The improved performance indicates that utilising the PyTorch framework for predictive tasks is advantageous, as evidenced by a statistical significance test in terms of both R2 and RMSE (p-values &lt; 0.001). The integration results display outstanding accuracy, averaging 90% across both feature sets. Particularly, the XGB model, which exhibited the lowest performance among all models in both sets, significantly improved from 0.8903 to 0.9097 in set 1 and from 0.8717 to 0.9022 in set 2 after being combined with the PNN. These results showcase the efficacy of using the PyTorch framework, enhancing the precision and reliability of the ML models in predicting the dynamic real estate market. Given that this study applies two feature sets and demonstrates consistent performance across sets with varying characteristics, the methodology may be applied to other locations. By offering accurate projections, it aids investors, renters, property managers, and regulators in facilitating better decision-making in the real estate sector.

Rice Leaf Diseases Classification Using Deep Learning Algorithms for Smartphone Application Development: An Empirical Study

Rice is the most widely consumed grain across the world. The rice plants often suffer from diseases. Early detection of such diseases and adopting remedial measures can help the farmers to avoid major losses and can produce best-quality crops in large quantities. However, the conventional rice leaf disease detection techniques are often not accurate, time-consuming and sometimes require laboratory testing. In this context, automatic rice leaf disease detection techniques are presented based on the various deep learning classifiers (namely MobileNetV2, ResNet50, VGG16 and Le-Net5) and an Android application is also developed in order to instantly determine the possible rice diseases from the uploaded rice leaf images captured by the smartphone. The developed models are tested using the publicly available benchmark rice leaf dataset containing three types of rice leaf diseases, namely bacterial leaf blight, leaf smut and brown spot. Experimental results show that MobileNetV2 model performed better compared to other models in terms of classification accuracy, recall and [Formula: see text]-score. The results of statistical significance test also confirmed the superiority of the MobileNetV2 model over other compared deep learning models.

Prevalence of Internet Use among Adolescents of 15 to 18 Years of Age in A Village under RHTC of A Medical College: Cross-Sectional Study

Background: The environment in which today’s adolescents are growing has changed drastically and affected their access to internet as well. The study was conducted with the aim of estimating the prevalence of internet use and pattern of internet use among adolescents of a village area. Methods: About 114 adolescents of a village in Western Maharashtra belonging to age group 15-18 years participated in the study. They were selected based on the inclusion criteria of age group and willingness to participate in the study via universal sampling method and were interviewed with the help of semi-structured questionnaire, which included questions regarding socio-demography, internet use and its interrelationship with other variables. Chi-square test was used as a test of statistical significance to study the association between the predictor variable and the outcome with (p&lt;0.05) as a level of significance. Results: The prevalence of internet use was 99.12%. Where 74% of adolescents used internet for an average of 1-2 hours in a day and 26% used internet for an about 2-4 hours in a day. Taking p &lt; 0.05 as a level of significance we found gender was significantly associated with duration of internet use (p = 0.01) and Purpose of internet use (p = 0.004). Conclusion: Almost all adolescents have access to internet and are more inclined towards its use for non-educational purpose which can be detrimental for their educational progress and long-term career goals. Hence, monitoring of internet use by adolescents has to be done by guardians/teachers. Further studies should be conducted on cause-effect relationship to study the either way of association.

Emergency physicians perspectives of state continuing medical education requirements for medical licensure.

This study aims to better understand the perspectives of emergency medicine physicians' on the role that state-mandated, topic-specific continuing medical education (CME) plays in addressing knowledge gaps, its relevance to current emergency practice, its reported burden and costs of CME activities to emergency physicians, and its perceived improvement in patient care. A cross-sectional survey was designed by the Coalition of Board-Certified Emergency Physicians (COBCEP) and distributed in February 2023 to all American Board of Emergency Medicine (ABEM)-certified physicians. Statistical tests of significance (Pearson's chi-square and Fisher's exact test) assessed the cost and time spent on CME as well as the perceived value placed on CME by ABEM-certified physicians to improve patient care. Data were summarized using descriptive statistics. There were 5562 (13.0%) responses from the 43656 physicians who received the survey-5506 responses were included for analysis. Over half of the physicians (53.0%) had more than 15 years of post-residency practice experience. Most physicians (57.3%) spent less than $5,000 per year on obtaining CME. Most physicians practicing in states with state-mandated, topic-specific CME requirements believed that participation in ABEM continuing certification could be used to reduce the need for state-mandated, topic-specific CME requirements (83.6%) and state-mandated, topic-specific requirements were believed to be unlikely to improve patient care (70.8%). Although well-intended, state CME requirements may lack relevancy and can, at times, place an undue burden on emergency physicians. Tailoring CME requirements to increase relevance to their patient populations and reduce barriers to completing CME could enhance knowledge translation and improve patient outcomes.

The role of effect size and significance test in research design and analysis

Effect size is a straightforward method for measuring the difference between two groups and plays a critical role in research design and significance testing. This paper addresses the concept of effect size, explaining what it is, how it is calculated, and how it can be interpreted. It also explores related topics, such as statistical significance, meta-analysis, and power analysis. Effect sizes can be understood in the context of confidence intervals, which provide similar information to statistical significance tests but focus more on the magnitude of the effect rather than the size of the sample.

ANALYTICAL STUDY ON FACTORS AFFECTING THE ATTITUDE AND PERSPECTIVES OF FEMALE PATIENTS ON THE USE OF ARTIFICIAL INTELLIGENCE IN THE ASSESSMENT OF SCREENING MAMMOGRAMS

Background: Breast cancer is one of the most common cancers among Indian women, with an incidence of 25.8 per 100,000 women according to the Ministry of Health and Family Welfare. Late detection is responsible for poor quality of life (QOL), and it is the leading cause of death. Studies have shown that people are positive towards AI performing assessment tasks in healthcare in general and in mammography screening. Moreover, it is challenging to implement strategies based on self-breast inspection or do mammography in rural regions or low and middle income nations for a variety of reasons. Objective: This study was conducted to assess the perception and attitude of various female patients reporting to the radiology unit of a tertiary care centre regarding the use of Artificial Intelligence in routine mammogram screening. Study Design: A cross sectional questionnaire based study was conducted female patients attending the radiology unit of a tertiary Care Centre in Central Kerala. Methods: After obtaining, informed written consent was taken from the study participants. A predesigned, pretested, validated checklist was used to collect the required data. The Knowledge, attitude and practice was ascertained through Likerts scale and scoring done accordingly. Statistical tests of significance was employed to assess the possible associations between various variables with the knowledge, attitude and perceptions regarding Artifical Intelligence based Mammogram Results: Among the 170 study subjects analysed, 68.6% had a satisfactory knowledge about Artificial intelligence. Among the 170 study subjects analysed, 86% had a good attitude about being screened through AI enabled mammogram techniques. 92% of the total study subjects registered for mammogram had done it based on self choice. The association between age and willingness to get screened was found to be statistically significant. It was also seen that there was a statistically very high significance between Knowledge regarding AI and the independent decision to get screened. There was also a statistically significant association between age and scores &gt;60% regarding knowledge and attitude regarding the vaccination among the study subjects. Conclusions: Majority of the study population was having a satisfactory knowledge regarding AI usage in healthcare.

Ranking co-change candidates suggested by FLeCCS using programmer sensitivity

Identifying co-change candidates when making changes to a particular program entity is an important software engineering task which is known as change impact analysis. A number of techniques for detecting co-change candidates currently exist. Most of these techniques work considering file level, class level, or method level granularity. FLeCCS is a recently introduced technique that suggests co-change candidates considering even fragment level granularity. FLeCCS ranks its co-change candidates considering file proximity. However, programmer sensitivity might be an important factor while ranking co-change candidates. Intuitively, it is more comfortable for a programmer to change her own code rather than changing someone else's code. Considering this fact, we investigate ranking co-change candidates suggested by FLeCCS considering programmer sensitivity.In our research, we have two main contributions: (1) proposing a composite ranking mechanism considering three things - file proximity, code similarity, and programmer sensitivity and (2) an empirical evaluation of our proposed mechanism by comparing it with file proximity ranking and similarity extent ranking mechanisms on thousands of revisions of five subject systems. We compare the ranking mechanisms in ranking the co-change candidates suggested by FLeCCS. According to our investigation incorporating four measures (Precision At K, Recall At K, Mean Average Precision and Mean Reciprocal Rank), composite ranking mechanism significantly outperforms the other two existing ranking mechanisms. Our findings are supported by statistical significance tests. We believe that our Proposed Composite Ranking mechanism is not only applicable to FLeCCS but also to other co-change suggestion techniques.