Likelihood Score Research Articles

The role of transvaginal ultrasound in the third-trimester evaluation of patients at high risk of placenta accreta spectrum at birth

Transvaginal ultrasound imaging has become an essential tool in the prenatal evaluation of the lower uterine segment and anatomy of the cervix, but there are only limited data on the role of transvaginal ultrasound in the management of patients at high risk of placenta accreta spectrum at birth. This study aimed to evaluate the role of transvaginal sonography in the third trimester of pregnancy in predicting outcomes in patients with a high probability of placenta accreta spectrum at birth. This was a retrospective analysis of prospectively collected data of patients presenting with a singleton pregnancy and a history of at least 1 previous cesarean delivery and patients diagnosed prenatally with an anterior low-lying placenta or placenta previa delivered electively after 32 weeks of gestation. All patients had a least 1 detailed ultrasound examination, including transabdominal and transvaginal scans, within 2 weeks before delivery. Of note, 2 experienced operators, blinded to the clinical data, were asked to make a judgment on the likelihood of placenta accreta spectrum as a binary, low or high-probability of placenta accreta spectrum, and to predict the main surgical outcome (conservative vs peripartum hysterectomy). The diagnosis of accreta placentation was confirmed when one or more placental cotyledons could not be digitally separated from the uterine wall at delivery or during the gross examination of the hysterectomy or partial myometrial resection specimens. A total of 111 patients were included in the study. Abnormal placental tissue attachment was found in 76 patients (68.5%) at birth, and histologic examination confirmed superficial villous attachment (creta) and deep villous attachment (increta) in 11 and 65 cases, respectively. Of note, 72 patients (64.9%) had a peripartum hysterectomy, including 13 cases with no evidence of placenta accreta spectrum at birth because of failure to reconstruct the lower uterine segment and/or excessive bleeding. There was a significant difference in the distribution of placental location (X2=12.66; P=.002) between transabdominal and transvaginal ultrasound examinations, but both ultrasound techniques had similar likelihood scores in identifying accreta placentation that was confirmed at birth. On transabdominal scan, only a high lacuna score was significantly associated (P=.02) with an increased chance of hysterectomy, whereas on transvaginal scan, significant associations were found between the need for hysterectomy and the thickness of the distal part of the lower uterine segment (P=.003), changes in the cervix structure (P=.01), cervix increased vascularity (P=.001), and the presence of placental lacunae (P=.005). The odds ratio for peripartum hysterectomy were 5.01 (95% confidence interval, 1.25-20.1) for a very thin (<1-mm) distal lower uterine segment and 5.62 (95% confidence interval, 1.41-22.5) for a lacuna score of 3+. Transvaginal ultrasound examination contributes to both prenatal management and the prediction of surgical outcomes in patients with a history of previous cesarean delivery with and without ultrasound signs suggestive of placenta accreta spectrum. Transvaginal ultrasound examination of the lower uterine segment and cervix should be included in clinical protocols for the preoperative evaluation of patients at risk of complex cesarean delivery.

On the pitfalls of Gaussian likelihood scoring for causal discovery

Abstract We consider likelihood score-based methods for causal discovery in structural causal models. In particular, we focus on Gaussian scoring and analyze the effect of model misspecification in terms of non-Gaussian error distribution. We present a surprising negative result for Gaussian likelihood scoring in combination with nonparametric regression methods.

"10,000 Available" or "10% Remaining": The Impact of Scarcity Framing on Ticket Availability Perceptions in the Secondary Ticket Market.

The aim of this study was to investigate the effect of numeracy framing and demand on participants' perceived ticket availability and likelihood of finding a lower-priced deal in the secondary ticket market for National Football League (NFL) games. A total of 640 participants were recruited via Qualtrics where participants were solicited electronically via 10 date-specific email blasts prior to a New York Giants Sunday Night Football home game. Participants were randomly assigned to one of five treatment conditions (control, percentage frame × low demand, percentage frame × high demand, frequency frame × low demand, frequency frame × high demand) to complete an online survey. Multivariate analysis of variance (MANOVA) was performed to discern overall differences in the mean likelihood scores of the dependent variable between groups. The results showed that participants presented with the "percentage" frame perceived tickets as less available than those presented with the "frequency scarcity" frame, and the effect was greater for high-demand games. Additionally, game demand moderated the effect of scarcity framing on participants' perceived ticket availability and expected lower rate. Several manipulation checks were applied to ensure the study's validity. The findings of this study have practical implications for ticket marketers in the sport industry to effectively frame scarcity information and facilitate transactions for online buyers and sellers.

Detection of anomalies of a non-deterministic software-defined networking control

Software Defined Networking (SDN) is a network architecture within the control is centralized through a software-based controller. Being a single point of attack makes the controller the preferred target in the SDN architecture. Multi-controller architecture has been introduced to reinforce the control plane. However it requires a communication interface between the controllers which is a security threat. In this objective, a dual controller architecture is introduced and it consists of one nominal controller in charge of the data plane computation plus a second one in charge of the detection of anomalies in the decisions taken by the main controller. In the case of non-determinist algorithm, the detection logic aims at determining a likelihood score of the decisions taken by the controller. A multi-criterion detection approach is proposed by considering both the performance of the decisions and the structure of the decisions taken by the controller. Such computations are probabilistic and attention has been paid to machine learning algorithms to determine this likelihood. More precisely, three formalisms are compared: Probabilistic Finite Automaton, Hidden Markov Model and Recurrent Neural Network. The impact of the control variance in the detection accuracy depending on the formalism used is discussed on a case study.

Corrigendum to: Efficacy and acceptability of 'nudges' aimed at promoting pre-exposure prophylaxis (PrEP) use: a survey of overseas born men who have sex with men.

This study explores the potential for behavioural economics techniques called 'nudges' to encourage the use of HIV pre-exposure prophylaxis (PrEP) by overseas-born men who have sex with men (MSM) in Australia. We investigated the preferences of overseas-born MSM for different nudges and the effect of nudges on reported likelihood of seeking information about PrEP. We conducted an online survey of overseas-born MSM, in which they were asked: (1) how likely they and a relevant friend would be to click on PrEP advertisements that used behavioural economics strategies; and (2) what they most and least liked about each ad. We conducted ordered logistic regression of reported likelihood scores against participant age and sexual orientation, use of a model in an advertisement, use of statistics about PrEP, reference to the World Health Organization (WHO), rewards for seeking further information, and use of a call-to-action. Participants (n =324) reported higher likelihoods of clicking on advertisements with images of people, statistics about PrEP, rewards for seeking further information, and calls-to-action. They reported lower likelihoods of clicking on advertisements referencing the WHO. They had negative emotional responses to sexualised humour, gambling metaphors, and the slogan 'Live Fearlessly'. Overseas-born MSM prefer public health messages that feature representative messengers and statistics about PrEP. These preferences are consistent with previous data on descriptive norms (i.e. statistics about the number of peers doing the desired behaviour) and gain-framed information (i.e. focusing on what can be gained from an intervention).

Homogeneity Test of the First-Order Agreement Coefficient in a Stratified Design

Gwet's first-order agreement coefficient (AC1) is widely used to assess the agreement between raters. This paper proposes several asymptotic statistics for a homogeneity test of stratified AC1 in large sample sizes. These statistics may have unsatisfactory performance, especially for small samples and a high value of AC1. Furthermore, we propose three exact methods for small pieces. A likelihood ratio statistic is recommended in large sample sizes based on the numerical results. The exact E approaches under likelihood ratio and score statistics are more robust in the case of small sample scenarios. Moreover, the exact E method is effective to a high value of AC1. We apply two real examples to illustrate the proposed methods.

Modeling Under-Dispersed Count Data by the Generalized Poisson Distribution via Two New MM Algorithms

Under-dispersed count data often appear in clinical trials, medical studies, demography, actuarial science, ecology, biology, industry and engineering. Although the generalized Poisson (GP) distribution possesses the twin properties of under- and over-dispersion, in the past 50 years, many authors only treat the GP distribution as an alternative to the negative binomial distribution for modeling over-dispersed count data. To our best knowledge, the issues of calculating maximum likelihood estimates (MLEs) of parameters in GP model without covariates and with covariates for the case of under-dispersion were not solved up to now. In this paper, we first develop a new minimization–maximization (MM) algorithm to calculate the MLEs of parameters in the GP distribution with under-dispersion, and then we develop another new MM algorithm to compute the MLEs of the vector of regression coefficients for the GP mean regression model for the case of under-dispersion. Three hypothesis tests (i.e., the likelihood ratio, Wald and score tests) are provided. Some simulations are conducted. The Bangladesh demographic and health surveys dataset is analyzed to illustrate the proposed methods and comparisons with the existing Conway–Maxwell–Poisson regression model are also presented.

Filter Likelihood as an Observation-Based Verification Metric in Ensemble Forecasting

In numerical weather prediction (NWP), ensemble forecasting aims to quantify the flow-dependent forecast uncertainty. The focus here is on observation-based verification of the reliability of ensemble forecasting systems. In particular, at short forecast lead times, forecast errors tend to be relatively small compared to observation errors and hence it is very important that the verification metric also accounts for observational uncertainties. This paper studies the so-called <em>filter likelihood score</em> which is deep-rooted in Bayesian estimation theory and fits naturally to the filtering setup of NWP. The filter likelihood score considers observation errors, ensemble mean skill, and ensemble spread in one metric. Importantly, it can be made multivariate and effortlessly expanded to simultaneous verification against all observation types through the observation operators contained in the parental data assimilation scheme. Here observations from the global radiosonde network and satellites (AMSU-A channel 5) are included in the verification of OpenIFS-based ensemble forecasts using different types of initial state perturbations. Our results show that the filter likelihood score is sensitive to the ensemble prediction system quality and compares consistently with other verification metrics such as the relationships between ensemble spread and ensemble mean forecast error, and Dawid-Sebastiani score. Our conclusion is that the filter likelihood score provides a very well-behaving verification metric, that can be made truly multivariate by including covariances, for ensemble prediction systems with a strong foundation in estimation theory.

Performance of clear cell likelihood scores in characterizing solid renal masses at multiparametric MRI: an external validation study.

The purpose of this study is to evaluate the accuracy and interobserver agreement of ccLS in diagnosing clear cell renal cell carcinoma (ccRCC). This retrospective single-center study evaluated consecutive patients with solid renal masses who underwent mpMRI followed by percutaneous biopsy and/or surgical excision between January 2010 and December 2020. Predominantly (> 75%) cystic masses, masses with macroscopic fat and infiltrative masses were excluded. Two abdominal radiologists independently scored each renal mass according to the proposed ccLS algorithm. The diagnostic performance of ccLS categories for ccRCC was calculated using logistic regression modeling. Diagnostic accuracy for predicting ccRCC was calculated using 2 × 2 contingency tables. Interobserver agreement for ccLS was evaluated with Cohen's k statistic. A total of 79 patients (mean age, 63years ± 12 [SD], 50 men) with 81 renal masses were evaluated. The mean size was 36mm ± 28 (range 10-160). Of the renal masses included, 44% (36/81) were ccRCC. The area under the receiver operating characteristic curve was 0.87 (95% CI 0.79-0.95). Using ccLS ≥ 4 to diagnose ccRCC, the sensitivity, specificity, and positive predictive value were 93% (95% CI 79, 99), 63% (95% CI 48, 77), and 67% (95% CI 58, 75), respectively. The negative predictive value of ccLS ≤ 2 was 93% (95% CI 64, 99). The proportion of ccRCC by ccLS category 1 to 5 were 10%, 0%, 10%, 57%, and 84%, respectively. Interobserver agreement was moderate (k = 0.47). In this study, clear cell likelihood score had moderate interobserver agreement and resulted in 96% negative predictive value in excluding ccRCC.

Preparing for Group B Streptococcus vaccine. Attitudes of pregnant women in two countries

ABSTRACT Group B Streptococcus (GBS) vaccines, designed to be given to pregnant women, are in clinical trials. There is an opportunity to conduct preparatory research now to understand the drivers of and barriers to GBS vaccine acceptance. This will enable targeted interventions so that delays in vaccine uptake might be avoided. A multicenter, mixed-methodology, cross-sectional study evaluated the acceptability of a hypothetical GBS vaccine among pregnant women in two countries with differing health systems. Pregnant women in Philadelphia, US, and Dublin, Ireland, completed an electronic survey and a Discrete Choice Experiment. Five hundred and two women were included in the final analysis. Fifty-three percent of US and 30% of Irish participants reported both awareness and understanding of GBS. The median likelihood score for vaccine receipt (measured on a 10-point scale) was 9 (US: 9 (IQR 7–10), IRL: 9 (IQR 6–10)). Among the US participants, identifying as Black or African American was associated with a lower likelihood of vaccine receipt. Possession of a college degree was associated with increased likelihood of vaccine receipt. Perceived infant benefit was the most important driver of GBS vaccine acceptance. Safety concerns about a novel vaccine was the most prominent barrier identified. Good GBS vaccine uptake is achievable through strong messaging that highlights vaccine safety and the potential infant benefits. Preparation for vaccine implementation should include efforts to increase awareness among pregnant women about GBS infection and a continued focus on improving acceptability of currently recommended maternal vaccines, particularly in population subgroups with low uptake of maternal immunizations.

Abstract A010: Deciphering molecular and cellular differences between black and white colorectal cancers through in-depth transcriptomic analyses

Abstract Background Compared to all other racial groups, black Americans are most likely to be diagnosed with colorectal cancer (CRC) and are most likely to die from the disease. These racial disparities have been largely attributed to social and economic conditions that impact the prevalence of co-morbidities and access to quality health care. However, the molecular and cellular differences in CRC tumors between black and white subjects remain largely uncharacterized. Methods We analyzed paired tumor and adjacent normal RNA sequencing reads collected from 15 black and 18 white subjects at Downstate Medical Center and Stony Brook University Hospital (DSSBU). 273 additional RNA sequencing and clinicopathologic data from The Cancer Genome Atlas (TCGA) were used as a validation cohort. In the TCGA dataset, 54 black subjects were propensity score matched to 108 white subjects by age, sex, tumor location and tumor stage. Differential expression analysis was performed with EdgeR and Wilcoxon rank-sum test, which was recently recommended for human population samples. Gene Ontology (GO) enrichment analysis of DEGs (absolute log 2-fold change &amp;gt; 1, adjust p value &amp;lt; 0.05) was performed using enrichR. Gene set enrichment analysis (GSEA) of hallmark gene sets was applied to the gene list ranked by fold changes. The TPM (transcript per million) data was used to estimate relative cell abundance via xCell. Antitumor immune response likelihood scores were generated via the Estimate Systems Immune Response (EaSIeR) R package. Results EdgeR analysis identified 15 and 80 DEGs in the DSSBU and TCGA datasets, respectively. Two DEGs, CXCL10 and RPPH1 have decreased expression while EEF1A1P13 has higher expression in black samples from both datasets. CXCL10 remained a DEG under the Wilcoxon test in the TCGA dataset and in propensity score matched samples. GO enrichment analysis identified “Chemokine-mediated signaling pathway” as a significantly different biological process between tumor groups with a FDR of 0.043 and 0.0004 in TCGA and DSSBU respectively. GSEA computed a negative enrichment score for IL6 JAK STAT3 signaling, Interferon alpha response, Interferon gamma response, Allograft rejection and Myc targets V1 in the black group from both datasets. xCell analysis of the DSSBU data revealed significant enrichment for endothelial cells, adipocytes and hematopoietic stem cells in black tumor samples, while white samples were significantly enriched for basophils, T helper 2 cells and common lymphoid progenitor cells. No significant differences in cell type enrichment were identified in the TCGA dataset. Predicted immunotherapy response scores were not significantly different between races in both cohorts. Conclusion Our results show differences in genes and pathways associated with inflammation and immune cell recruitment which may impact anti-tumor activity. Our future work will include experimental validation of interesting genes accounting for the differences. Citation Format: Dimitri F. Joseph, Ellen Li, Bin Chen, Joseph F. LaComb, Alexandra Guillaume. Deciphering molecular and cellular differences between black and white colorectal cancers through in-depth transcriptomic analyses [abstract]. In: Proceedings of the 15th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2022 Sep 16-19; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2022;31(1 Suppl):Abstract nr A010.

Efficacy and acceptability of 'nudges' aimed at promoting pre-exposure prophylaxis (PrEP) use: a survey of overseas born men who have sex with men.

This study explores the potential for behavioural economics techniques called 'nudges' to encourage the use of HIV pre-exposure prophylaxis (PrEP) by overseas-born men who have sex with men (MSM) in Australia. We investigated the preferences of overseas-born MSM for different nudges and the effect of nudges on reported likelihood of seeking information about PrEP. We conducted an online survey of overseas-born MSM, in which they were asked: (1) how likely they and a relevant friend would be to click on PrEP advertisements that used behavioural economics strategies; and (2) what they most and least liked about each ad. We conducted ordered logistic regression of reported likelihood scores against participant age and sexual orientation, use of a model in an advertisement, use of statistics about PrEP, reference to the World Health Organization (WHO), rewards for seeking further information, and use of a call-to-action. Participants (n =324) reported higher likelihoods of clicking on advertisements with images of people, statistics about PrEP, rewards for seeking further information, and calls-to-action. They reported lower likelihoods of clicking on advertisements referencing the WHO. They had negative emotional responses to sexualised humour, gambling metaphors, and the slogan 'Live Fearlessly'. Overseas-born MSM prefer public health messages that feature representative messengers and statistics about PrEP. These preferences are consistent with previous data on descriptive norms (i.e. statistics about the number of peers doing the desired behaviour) and gain-framed information (i.e. focusing on what can be gained from an intervention).

Diagnostic signature for heart failure with preserved ejection fraction (HFpEF): a machine learning approach using multi-modality electronic health record data

BackgroundHeart failure with preserved ejection fraction (HFpEF) is thought to be highly prevalent yet remains underdiagnosed. Evidence-based treatments are available that increase quality of life and decrease hospitalization. We sought to develop a data-driven diagnostic model to predict from electronic health records (EHR) the likelihood of HFpEF among patients with unexplained dyspnea and preserved left ventricular EF.Methods and resultsThe derivation cohort comprised patients with dyspnea and echocardiography results. Structured and unstructured data were extracted using an automated informatics pipeline. Patients were retrospectively diagnosed as HFpEF (cases), non-HF (control cohort I), or HF with reduced EF (HFrEF; control cohort II). The ability of clinical parameters and investigations to discriminate cases from controls was evaluated by extreme gradient boosting. A likelihood scoring system was developed and validated in a separate test cohort. The derivation cohort included 1585 consecutive patients: 133 cases of HFpEF (9%), 194 non-HF cases (Control cohort I) and 1258 HFrEF cases (Control cohort II). Two HFpEF diagnostic signatures were derived, comprising symptoms, diagnoses and investigation results. A final prediction model was generated based on the averaged likelihood scores from these two models. In a validation cohort consisting of 269 consecutive patients [with 66 HFpEF cases (24.5%)], the diagnostic power of detecting HFpEF had an AUROC of 90% (P < 0.001) and average precision of 74%.ConclusionThis diagnostic signature enables discrimination of HFpEF from non-cardiac dyspnea or HFrEF from EHR and can assist in the diagnostic evaluation in patients with unexplained dyspnea. This approach will enable identification of HFpEF patients who may then benefit from new evidence-based therapies.

Value of clear cell likelihood score in differentiation between renal oncocytoma and clear cell renal cell carcinoma

Objective: To evaluate the value of clear cell likelihood score (ccLS) in identifying renal oncocytoma (RO) and clear cell renal cell carcinoma (ccRCC). Methods: Retrospective data of pathologically confirmed 43 RO patients [24 men and 19 women, aged 22-77 (54±14) years] between February 2008 and September 2021 and 43 ccRCC patients [30 men and 13 women, aged 29-78 (56±12) years] between May and July 2021 were consecutively included in the department of radiology, Chinese PLA General Hospital. Two radiologists used ccLS to assess each case independently, and disagreements were resolved by consensus. The ability of ccLS to identify RO and ccRCC was examined by the receiver operating characteristic (ROC) curve which identified the best optimal diagnostic cut-off values, sensitivity, specificity, accuracy, positive predictive value, and negative predictive value. Results: The mean tumor diameter was 3.8 cm in RO patients and 3.7 cm in ccRCC patients. Central scar and segmental enhancement inversion (SEI) were more frequently observed in the RO group compared to the ccRCC group [53.5% (23∶43) versus 11.6% (5∶43) and 41.9% (18∶43) versus 7.0% (3∶43), respectively], with statistical differences (P<0.001). The ccLS scores in the RO group ranged from 1 to 4, while 79.0% of the cases were 3. The ccLS scores in the ccRCC group ranged from 2 to 5, while 72% of the cases were 4. The scores of the two groups were statistically different (P<0.001). The ccLS showed the best performance when the threshold was 4 according to the ROC curve. The sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of ccLS for distinguishing RO from ccRCC were 83.7%, 90.7%, 87.2%, 90.0%, and 84.8%, respectively, and the area under the ROC curve value was 0.879. Conclusion: The ccLS has credible sensitivity and specificity in differentiating renal oncocytoma from clear cell carcinoma, which may be helpful for the preoperative diagnosis.

Evaluation of Fatigue Risk, Subjective Fatigue, and Work Motivation Level of Intercity Bus Driver

In Indonesia, the bus is one of the affordable transportation for intercity/province routes. However, there are accident risks faced by bus drivers related to fatigue caused by inadequate rest and sleep. The objective of this study is to assess the bus driver fatigue risks level, accompanied by work motivation and subjective fatigue. Using the convenience sampling method, 300 intercity bus drivers from 12 bus companies (male, age between 25-68 years old (44.06 ± 9.83)) agree to involve in the study. Fatigue risk level was estimated by fatigue likelihood score (FLS) which indicated by driving tasks hours and the rest time opportunity. Work motivation and subjective fatigue were measured using 20 questions. The FLS result has a range between 13-39 (scale 0-40) or the risks are categorized into medium to very high. The major finding in this study, while a fatigue risks were relatively high, the subjective fatigue level were low to medium, and work motivations were medium to somewhat high. Statistical test showed that work motivation has a significant correlation with fatigue risks (p &lt; 0.05; r = -0.139), but not with a subjective fatigue level.

83P The Immune-related adverse event (IRAE) Likelihood Score (ILS) identifies “pure” IRAEs strongly associated with outcome in a phase I-II trial population

Immune-related adverse events (IRAE) pose a significant diagnostic and therapeutic challenge in patients treated with immuno-oncology (IO) drugs. IRAEs have been found to correlate with better outcome, but studies are conflicting on the magnitude and significance of this correlation. Estimating the true incidence of IRAEs is particularly difficult in the early phase 1/2 trial setting, with factors contributing to both over- and under-estimation. A key issue is the lack of IRAE diagnostic criteria, necessary to discriminate “pure” IRAEs from other treatment-related adverse events not sustained by an autoimmune process.

Age Differences in Heritability of a Latent Dementia Index Score in Men and Women

AbstractBackgroundHeritability estimates of dementia are ∼50‐60% of the total variance across older adulthood. Earlier age of onset, however, might be associated with greater genetic influence whereas older age of onset might be associated with greater environmental influences. The categorical nature of dementia diagnosis outcomes has made testing the relative importance of genetic and environmental influences on dementia difficult due to low statistical power. To overcome this methodological limitation, we tested linear and quadratic effects of age on genetic and environmental variance underlying a continuous measure of dementia likelihood in men and women.MethodThe analytic sample was 5,670 (53% female) pairs of twins from 10 studies in the Interplay of Genes and Environment Across Multiple Studies (IGEMS) consortium. Dementia likelihood is quantified continuously using a latent dementia index measure that is estimated from twins’ available cognitive, memory‐specific, and functional ability measures. Linear and quadratic age differences in genetic and environmental influences on dementia likelihood were tested by modeling univariate correlation between twins’ dementia scores as a linear function of zygosity and age in Mplus. Separate models were estimated for male and female twin groups.ResultsAge moderated genetic and environmental influences on dementia likelihood scores. Both genetic and nonshared environmental variance increased as a function of age, whereas common environmental variance declined. Sex significantly moderated the effect of zygosity on the twin correlations (monozygotic male: .57, monozygotic female: .51, dizygotic male: .19, dizygotic female: .25). Notably, MZ twin correlations were consistent with previously reported heritability estimates for dementia. The pattern of age moderation was the same in men as women, although the linear effect of age on both genetic and common environmental influences was greater in men than women (see Figure 1).ConclusionAge linearly moderated genetic and environmental variance underlying dementia likelihood similarly for both men and women, despite quantitative genetic and environmental sex differences in dementia likelihood. Genetic variance accounted for similar proportions of total variance at younger ages, whereas genetic variance accounted for greater variance in men than in women at older ages.

The Integrated 31-Gene Expression Profile Test (i31-GEP) for Cutaneous Melanoma Outperforms the CP-GEP at Identifying Patients who can Forego Sentinel Lymph Node Biopsy when Applying NCCN Guidelines

Introduction: Eighty-eight percent of patients who receive a sentinel lymph node biopsy (SLNB) receive a negative result. Current guidelines suggest avoiding SLNB if the risk of positivity is &lt;5%. A low threshold for performing SLNB indicates worry about missing positive nodes but results in many unnecessary biopsies. The integrated 31-gene expression profile (i31-GEP) test combines the 31-GEP with Breslow thickness, age, ulceration, and mitotic rate using a validated neural network algorithm to identify patients with &lt;5% risk by reporting a precise, individualized likelihood score. Using logistic regression, a separate GEP (8 genes plus Breslow thickness and age; CP-GEP) also aims to identify patients who can forego SLNB by binning patients into low- and high-risk categories. This study compared the i31-GEP for SLNB with the CP-GEP to identify patients who can safely forego SLNB. Methods: This study evaluated the genes, gene pathways, and relative contribution of the two gene signatures relative to clinical and pathologic factors and further analyzed patients with T1b-T2 tumors from previously published studies in U.S. cohorts for the i31-GEP (n=763) and CP-GEP (n=153). A ratio of true-to-false-negative i31-GEP for SLNB and CP-GEP test results was compared to the guideline-established ratio of 19 negatives to one missed positive (19:1) if foregoing SLNB using a 5% risk threshold. Results: The i31-GEP had a 30:1 true-to-false-negative ratio, compared to 15:1 using CP-GEP. In T1b tumors, the i31-GEP ratio increased to 35:1 compared to 14:1 using CP-GEP. Limitations: Retrospective design, which could lead to bias. Patients included in both analyses were primarily from institutional surgical centers, and referral bias may not make the result generalizable. Conclusion: Only the i31-GEP provided benefit over guideline-established care for identifying patients with low risk of SLN positivity.

Neural Network Diagnoses the Antiphospholipid Syndrome without Interrupting Anticoagulant Therapy

Background The antiphospholipid syndrome (APS) is characterized by the presence of antiphospholipid antibodies (aPL) and is associated with an increased risk of thrombosis and pregnancy morbidity. The Laboratory diagnosis of APS is difficult because most patients already use anticoagulants due to a thrombotic event, resulting in interference with the clotting assays used for lupus anticoagulant (LAC) testing. Nevertheless, the aPL profile defines patient management, making aPL testing warranted during anticoagulant therapy. Sometimes, cessation of anticoagulant treatment is necessary to perform the lupus anticoagulant (LAC) test, which is unfavorable. The thrombin generation (TG) test is a functional and global hemostatic test. Clinical studies have shown that increased TG is associated with thrombosis, also in thrombotic APS patients. We have previously shown that the artificial intelligence method neural networks can be used to diagnose untreated APS patients from a group of healthy, auto-immune disease, and thrombosis controls with high accuracy. At the moment of aPL testing, many patients are already on anticoagulants, hampering the detection of LAC. Therefore, we developed a neural net (NN) for diagnoses APS in a cohort of anticoagulated subjects, based on the TG test. Methods Antiphospholipid antibodies positivity was diagnosed according to the Sapporo-Sydney guidelines. TG was measured using the calibrated automated thrombogram assay (CAT) on platelet poor plasma with PPP reagent (intermediate concentration of tissue factor) in 48 APS patients and 64 control non-APS patients, both treated with vitamin K antagonists (VKAs) TG was explored in more depth by quantifying parameters of prothrombin conversion and thrombin inactivation, using thrombin dynamics (TD) analysis, and assessing the sensitivity of the activated protein C pathway through the addition of thrombomodulin (TM). Five NNs were developed and the most accurate NN was selected for clinical validation based on its sensitivity and specificity. Input parameters for the NN were TG parameters lag time, peak, endogenous thrombin potential, time-to-peak, and velocity index, and TD parameters total prothrombin conversion, maximum prothrombin conversion rate, thrombin-antithrombin complex formation, and thrombin-α2-macroglobulin complex formation, and the inhibition of TG and TD by TM. The top performing NN was clinically validated using the TG and TD data acquired in the validation cohort, consisting of 33 APS patients and 62 controls, both on VKA therapy, 49 auto-immune disease controls, 38 thrombosis controls, 92 patients visiting the hospital for other indications and 37 healthy subjects. All control subjects were negative for aPL. Results TG was slightly increased in APS patients compared to controls. VKA therapy abolished the typical lag time prolongation seen in non-anticoagulated APS patients due to the LAC effect. The rate of prothrombin conversion was significantly increased in anticoagulated APS patients compared to controls, indicating a more prothrombotic phenotype. The selected NN classified APS patients with a sensitivity of 92% and a specificity of 95% in the developmental cohort consisting of APS patients and controls on VKA antagonists (ROCAUC = 0.9805; 0.9542-1.000; p<0.0001). On average, controls and APS patients were assigned an APS likelihood score of 12% and 89%, respectively. The NN was clinically validated in a separate validation cohort of 311 individuals, including 33 anticoagulated APS patients, 62 anticoagulated controls, 38 thrombosis patients, 49 auto-immune disease patients, 92 hospital controls, and 37 healthy controls. The sensitivity of the NN for the detection of APS was 85%. The overall specificity of the NN was 93%, and 100%, 96%, 96%, 92% and 76% respectively in healthy controls, auto-immune disease controls, hospital controls, anticoagulated controls, and thrombosis controls. Conclusions We developed a NN that accurately classifies APS under anticoagulant treatment. Therefore, this NN could be an alternative for diagnosing APS patients already treated with VKAs. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal

Development of a core set of gait features and their potential underlying impairments to assist gait data interpretation in children with cerebral palsy.

The interpretation of clinical gait data in children with cerebral palsy (CP) is time-consuming, requires extensive expertise and often lacks transparency. Here we aimed to develop a set of look-up tables to support this process, linking typical gait features as present in CP to their potential underlying impairments. We developed an initial core set of gait features and their potential underlying impairments based on biomechanical reasoning, literature and clinical experience. This core set was further specified through a Delphi process in a multidisciplinary group of experts in gait analysis of children with CP and evaluated on 20 patient cases. The likelihood of the listed gait feature-impairment relationships was scored by the expert panel on a five-point scale. The final core set included 120 relevant gait feature-impairment relations including likelihood scores. This set was presented in the form of look-up tables in both directions, i.e., sorted by gait features with potential underlying impairment, and sorted by impairments with potential related gait features. The average likelihood score for the relations was 3.5 ± 0.6 (range 2.1-4.6). The developed set of look-up tables linking gait features and impairments, can assist gait analysts and clinicians in standardized biomechanical reasoning, to support treatment decision-making for gait impairments in children with CP.