A probabilistic knowledge graph for target identification.

Early identification of safe and efficacious disease targets is crucial to alleviating the tremendous cost of drug discovery projects. However, existing experimental methods for identifying new targets are generally labor-intensive and failure-prone. On the other hand, computational approaches, especially machine learning-based frameworks, have shown remarkable application potential in drug discovery. In this work, we propose Progeni, a novel machine learning-based framework for target identification. In addition to fully exploiting the known heterogeneous biological networks from various sources, Progeni integrates literature evidence about the relations between biological entities to construct a probabilistic knowledge graph. Graph neural networks are then employed in Progeni to learn the feature embeddings of biological entities to facilitate the identification of biologically relevant target candidates. A comprehensive evaluation of Progeni demonstrated its superior predictive power over the baseline methods on the target identification task. In addition, our extensive tests showed that Progeni exhibited high robustness to the negative effect of exposure bias, a common phenomenon in recommendation systems, and effectively identified new targets that can be strongly supported by the literature. Moreover, our wet lab experiments successfully validated the biological significance of the top target candidates predicted by Progeni for melanoma and colorectal cancer. All these results suggested that Progeni can identify biologically effective targets and thus provide a powerful and useful tool for advancing the drug discovery process.

Exposure Assessment of Pesticides in Surface Waters of Ontario, Canada Reveals Low Probability of Exceeding Acute Regulatory Thresholds.

The objectives of this study were to: (1) characterize the exposure of aquatic ecosystems in Southern Ontario, Canada to pesticides between 2002 and 2016 by constructing environmental exposure distributions (EEDs), including censored data; and (2) predict the probability of exceeding acute regulatory guidelines. Surface water samples were collected over a 15-year period by Environment and Climate Change Canada. The dataset contained 167 compounds, sampled across 114 sites, with a total of 2,213 samples. There were 67,920 total observations of which 55,058 were non-detects (81%), and 12,862 detects (19%). The most commonly detected compound was atrazine, with a maximum concentration of 18,600 ngL- 1 and ~ 4% chance of exceeding an acute guideline (1,000 ngL- 1) in rivers and streams. Using Southern Ontario as a case study, this study provides insight into the risk that pesticides pose to aquatic ecosystems and the utility of EEDs that include censored data for the purpose of risk assessment.

A New Model Including AMH Cut-off Levels to Predict Post-treatment Ovarian Function in Early Breast Cancer: A Prospective Cohort Study.

BACKGROUND: Breast cancer (BC) treatment decreases fertility capacity, but unnecessary fertility preservation procedures in women who would not be infertile after treatment would be a waste of time and resources and could cause the unwarranted exposure of cancer cells to exogenous sex hormones. It has been largely shown that post-treatment ovarian reserve is directly associated with pre-treatment anti-mullerian hormone levels (AMH0). A threshold for AMH0, or a model including AMH0 and patient characteristics that could distinguish the patients who will be infertile after treatments, still needs to be defined. Accordingly, this study was performed to specifically target this high-priority concern. METHODS: Women≤45 years old with newly diagnosed non-metastatic BC were entered in this multicenter prospective cohort study. AMH0 and two-year post-treatment AMH (AMH2) were measured, and hormonal patient features were recorded as well. Receiver operating characteristic (ROC) curve analysis, decision tree (DT), and random forest analyses were performed to find a cut-off point for AMH0 and define a model involving related features for the prediction of AMH2. RESULTS: The data from 84 patients were analyzed. ROC curve analysis revealed that AMH0>3 ng/mL (Area under the curve=0.69, 95% CI: 0.54‒0.84) was the best indicator for predicting AMH2≥0.7 (sensitivity=79%, specificity=60%). The best model detected by DT and random forest for predicting an AMH2>0.7 with a probability of 93% consisted of a combination of AMH0>3.3, menarche age<14, and age<31. CONCLUSION: This combination model can be used to withhold fertility preservation procedures in BC patients. Performing larger studies is suggested to further test this model.

Comparing surgical and endoscopic resection approaches for colorectal neuroendocrine tumors within the diameter range of 10-20mm: an inverse probability weighting analysis based on the SEER database.

Background: Currently, the primary treatment modalities for colorectal neuroendocrine tumors (CRNET) with a diameter between 10mm and 20mm are surgical resection (SR) and endoscopic resection (ER). However, it remains unclear which surgical approach yields the greatest survival benefit for patients. Methods: This study included data from patients diagnosed with CRNET with tumor diameters ranging from 10mm to 20mm between the years 2004 and 2019, obtained from the Surveillance, Epidemiology, and End Results (SEER) database. Patients were categorized into ER and SR groups based on the respective surgical approaches. Inverse probability weighting (IPTW) was employed to mitigate selection bias. Kaplan-Meier analysis and log-rank tests were utilized to estimate overall survival (OS) and cancer-specific survival (CSS). Cox regression analysis (univariate and multivariate) was performed to evaluate potential factors influencing survival. Results: A total of 292 CRNET patients were included in this study (ER group: 108 individuals, SR group: 184 individuals). Prior to IPTW adjustment, Kaplan-Meier analysis and Cox proportional hazard regression analysis demonstrated that the OS and CSS of the SR group were inferior to those of the ER group. However, after IPTW adjustment, no statistically significant differences in prognosis were observed between the two groups. Subgroup analyses revealed that patients with muscular invasion, positive lymph nodes, or distant metastasis derived greater survival benefits from SR. Significant differences in OS and CSS between the two groups were also observed across different age groups. Conclusion: For patients with mucosal-limited lesions and without local lymph node or distant metastasis, ER is the preferred surgical approach. However, for patients with muscular invasion or positive lymph nodes/distant metastasis, SR offers a better prognosis. The choice of surgical approach should be based on the specific clinical characteristics of patients within different subgroups.

Pollution levels and probability risk assessment of potential toxic elements in soil of Pb-Zn smelting areas.

Soil pollution around Pb-Zn smelters has attracted widespread attention around the world. In this study, we compiled a database of eight potentially toxic elements (PTEs) Pb, Zn, Cd, As, Cr, Ni, Cu, and Mn in the soil of Pb-Zn smelting areas by screening the published research papers from 2000 to 2023. The pollution assessment and risk screening of eight PTEs were carried out by geo-accumulation index (Igeo), potential ecological risk index (PERI) and health risk assessment model, and Monte Carlo simulation employed to further evaluate the probabilistic health risks. The results suggested that the mean values of the eight PTEs all exceeded the corresponding values in the upper crust, and more than 60% of the study sites had serious Pb and Cd pollution (Igeo > 4), with Brazil, Belgium, China, France and Slovenia having higher levels of pollution than other regions. Besides, PTEs in smelting area caused serious ecological risk (PERI = 10912.12), in which Cd was the main contributor to PREI (86.02%). The average hazard index (HI) of the eight PTEs for adults and children was 7.19 and 9.73, respectively, and the average value of total carcinogenic risk (TCR) was 4.20 × 10-3 and 8.05 × 10-4, respectively. Pb and As are the main contributors to non-carcinogenic risk, while Cu and As are the main contributors to carcinogenic risk. The probability of non-carcinogenic risk in adults and children was 84.05% and 97.57%, while carcinogenic risk was 92.56% and 79.73%, respectively. In summary, there are high ecological and health risks of PTEs in the soil of Pb-Zn smelting areas, and Pb, Cd, As and Cu are the key elements that cause contamination and risk, which need to be paid attention to and controlled. This study is expected to provide guidance for soil remediation in Pb-Zn smelting areas.

Bayesian two-stage modeling of longitudinal and time-to-event data with an integrated fractional Brownian motion covariance structure.

It is difficult to characterize complex variations of biological processes, often longitudinally measured using biomarkers that yield noisy data. While joint modeling with a longitudinal submodel for the biomarker measurements and a survival submodel for assessing the hazard of events can alleviate measurement error issues, the continuous longitudinal submodel often uses random intercepts and slopes to estimate both between- and within-patient heterogeneity in biomarker trajectories. To overcome longitudinal submodel challenges, we replace random slopes with scaled integrated fractional Brownian motion (IFBM). As a more generalized version of integrated Brownian motion, IFBM reasonably depicts noisily measured biological processes. From this longitudinal IFBM model, we derive novel target functions to monitor the risk of rapid disease progression as real-time predictive probabilities. Predicted biomarker values from the IFBM submodel are used as inputs in a Cox submodel to estimate event hazard. This two-stage approach to fit the submodels is performed via Bayesian posterior computation and inference. We use the proposed approach to predict dynamic lung disease progression and mortality in women with a rare disease called lymphangioleiomyomatosis who were followed in a national patient registry. We compare our approach to those using integrated Ornstein-Uhlenbeck or conventional random intercepts-and-slopes terms for the longitudinal submodel. In the comparative analysis, the IFBM model consistently demonstrated superior predictive performance.

Temporal Variation and Persistence of Methane Emissions from Shallow Water Oil and Gas Production in the Gulf of Mexico.

Methane emissions from the oil and gas supply chain can be intermittent, posing challenges for monitoring and mitigation efforts. This study examines shallow water facilities in the US Gulf of Mexico with repeat atmospheric observations to evaluate temporal variation in site-specific methane emissions. We combine new and previous observations to develop a longitudinal study, spanning from days to months to almost five years, evaluating the emissions behavior of sites over time. We also define and determine the chance of subsequent detection (CSD): the likelihood that an emitting site will be observed emitting again. The average emitting central hub in the Gulf has a 74% CSD at any time interval. Eight facilities contribute 50% of total emissions and are over 80% persistent with a 96% CSD above 100 kg/h and 46% persistent with a 42% CSD above 1000 kg/h, indicating that large emissions are persistent at certain sites. Forward-looking infrared (FLIR) footage shows many of these sites exhibiting cold venting. This suggests that for offshore, a low sampling frequency over large spatial coverage can capture typical site emissions behavior and identify targets for mitigation. We further demonstrate the preliminary use of space-based observations to monitor offshore emissions over time.

Development of the multivariate administrative data cystectomy model and its impact on misclassification bias.

BACKGROUND: Misclassification bias (MB) is the deviation of measured from true values due to incorrect case assignment. This study compared MB when cystectomy status was determined using administrative database codes vs. predicted cystectomy probability. METHODS: We identified every primary cystectomy-diversion type at a single hospital 2009-2019. We linked to claims data to measure true association of cystectomy with 30 patient and hospitalization factors. Associations were also measured when cystectomy status was assigned using billing codes and by cystectomy probability from multivariate logistic regression model with covariates from administrative data. MB was the difference between measured and true associations. RESULTS: 500 people underwent cystectomy (0.12% of 428 677 hospitalizations). Sensitivity and positive predictive values for cystectomy codes were 97.1% and 58.6% for incontinent diversions and 100.0% and 48.4% for continent diversions, respectively. The model accurately predicted cystectomy-incontinent diversion (c-statistic [C] 0.999, Integrated Calibration Index [ICI] 0.000) and cystectomy-continent diversion (C:1.000, ICI 0.000) probabilities. MB was significantly lower when model-based predictions was used to impute cystectomy-diversion type status using for both incontinent cystectomy (F = 12.75; p < .0001) and continent cystectomy (F = 11.25; p < .0001). CONCLUSIONS: A model using administrative data accurately returned the probability that cystectomy by diversion type occurred during a hospitalization. Using this model to impute cystectomy status minimized MB. Accuracy of administrative database research can be increased by using probabilistic imputation to determine case status instead of individual codes.

Application of inverse weighting analysis to assess the association of youth perceptions with the age of initiation of tobacco products.

Introduction: To examine if perceptions of harmfulness and addictiveness of hookah and cigarettes impact the age of initiation of hookah and cigarettes, respectively, among US youth. Youth (12-17 years old) users and never users of hookah and cigarettes during their first wave of PATH participation were analyzed by each tobacco product (TP) independently. The effect of perceptions of (i) harmfulness and (ii) addictiveness at the first wave of PATH participation on the age of initiation of ever use of hookah was estimated using interval-censoring Cox proportional hazards models. Methods: Users and never users of hookah at their first wave of PATH participation were balanced by multiplying the sampling weight and the 100 balance repeated replicate weights with the inverse probability weight (IPW). The IPW was based on the probability of being a user in their first wave of PATH participation. A Fay's factor of 0.3 was included for variance estimation. Crude hazard ratios (HR) and 95% confidence intervals (CIs) are reported. A similar process was repeated for cigarettes. Results: Compared to youth who perceived each TP as "a lot of harm", youth who reported perceived "some harm" had younger ages of initiation of these tobacco products, HR: 2.53 (95% CI: 2.87-4.34) for hookah and HR: 2.35 (95% CI: 2.10-2.62) for cigarettes. Similarly, youth who perceived each TP as "no/little harm" had an earlier age of initiation of these TPs compared to those who perceived them as "a lot of harm", with an HR: 2.23 (95% CI: 1.82, 2.71) for hookah and an HR: 1.85 (95% CI: 1.72, 1.98) for cigarettes. Compared to youth who reported each TP as "somewhat/very likely" as their perception of addictiveness, youth who reported "neither likely nor unlikely" and "very/somewhat unlikely" as their perception of addictiveness of hookah had an older age of initiation, with an HR: 0.75 (95% CI: 0.67-0.83) and an HR: 0.55 (95% CI: 0.47, 0.63) respectively. Discussion: Perceptions of the harmfulness and addictiveness of these tobacco products (TPs) should be addressed in education campaigns for youth to prevent early ages of initiation of cigarettes and hookah.

Towards a Clearer Causal Question Underlying the Association Between Cancer and Dementia.

BACKGROUND: Several observational studies have described an inverse association between cancer diagnosis and subsequent dementia risk. Multiple biologic mechanisms and potential biases have been proposed in attempts to explain this association. One proposed explanation is the opposite expression of Pin1 in cancer and dementia, and we use this explanation and potential drug target to illustrate the required assumptions and potential sources of bias for inferring an effect of Pin1 on dementia risk from analyses measuring cancer diagnosis as a proxy for Pin1 expression. METHODS: We used data from the Rotterdam Study, a population-based cohort. We estimate the association between cancer diagnosis (as a proxy for Pin1) and subsequent dementia diagnosis using two different proxy methods and with confounding and censoring for death addressed with inverse probability weights. We estimate and compare the complements of a weighted Kaplan-Meier survival estimator at 20 years of follow-up. RESULTS: Out of 3634 participants, 899 (25%) were diagnosed with cancer, of whom 53 (6%) had dementia, and 567 (63%) died. Among those without cancer, 15% (411) were diagnosed with dementia, and 667 (24%) died over follow-up. Depending on the confounding and selection bias control, and the way in which cancer was used as a time-varying proxy exposure, the risk ratio for dementia diagnosis ranged from 0.71 (95% confidence interval [CI] = 0.49, 0.95) to 1.1 (95% CI = 0.79, 1.3). CONCLUSION: Being explicit about the underlying mechanism of interest is key to maximizing what we can learn from this cancer-dementia association given available or readily collected data, and to defining, detecting, and preventing potential biases.

A Bayesian Bernoulli-Exponential joint model for binary longitudinal outcomes and informative time with applications to bladder cancer recurrence data.

BACKGROUND: A variety of methods exist for the analysis of longitudinal data, many of which are characterized with the assumption of fixed visit time points for study individuals. This, however is not always a tenable assumption. Phenomenon that alter subject visit patterns such as adverse events due to investigative treatment administered, travel or any other emergencies may result in unbalanced data and varying individual visit time points. Visit times can be considered informative, because subsequent or current subject outcomes can change or be adapted due to previous subject outcomes. METHODS: In this paper, a Bayesian Bernoulli-Exponential model for analyzing joint binary outcomes and exponentially distributed informative visit times is developed. Via statistical simulations, the influence of controlled variations in visit patterns, prior and sample size schemes on model performance is assessed. As an application example, the proposed model is applied to a Bladder Cancer Recurrence data. RESULTS AND CONCLUSIONS: Results from the simulation analysis indicated that the Bayesian Bernoulli-Exponential joint model converged in stationarity, and performed relatively better for small to medium sample size scenarios with less varying time sequences regardless of the choice of prior. In larger samples, the model performed better for less varying time sequences. This model's application to the bladder cancer data showed a statistically significant effect of prior tumor recurrence on the probability of subsequent recurrences.

Risk adaptive planning with biology-based constraints may lead to higher tumor control probability in tumors of the canine brain: A planning study.

BACKGROUND: Classical radiation protocols are guided by physical dose delivered homogeneously over the target. Protocols are chosen to keep normal tissue complication probability (NTCP) at an acceptable level. Organs at risk (OAR) adjacent to the target volume could lead to underdosage of the tumor and a decrease of tumor control probability (TCP). The intent of our study was to explore a biology-based dose escalation: by keeping NTCP for OAR constant, radiation dose was to be maximized, allowing to result in heterogeneous dose distributions. METHODS: We used computed tomography datasets of 25 dogs with brain tumors, previously treated with 10x4 Gy (40 Gy to PTV D50). We generated 3 plans for each patient: A) original treatment plan with homogeneous dose distribution, B) heterogeneous dose distribution with strict adherence to the same NTCPs as in A), and C) heterogeneous dose distribution with adherence to NTCP <5%. For plan comparison, TCPs and TCP equivalent doses (homogenous target dose which results in the same TCP) were calculated. To enable the use of the generalized equivalent uniform dose (gEUD) metric of the tumor target in plan optimization, the calculated TCP values were used to obtain the volume effect parameter a. RESULTS: As intended, NTCPs for all OARs did not differ from plan A) to B). In plan C), however, NTCPs were significantly higher for brain (mean 2.5% (SD±1.9, 95%CI: 1.7,3.3), p<0.001), optic chiasm (mean 2.0% (SD±2.2, 95%CI: 1.0,2.8), p=0.010) compared to plan A), but no significant increase was found for the brainstem. For 24 of 25 of the evaluated patients, the heterogenous plans B) and C) led to an increase in target dose and projected increase in TCP compared to the homogenous plan A). Furthermore, the distribution of the projected individual TCP values as a function of the dose was found to be in good agreement with the population TCP model. CONCLUSION: Our study is a first step towards risk-adaptive radiation dose optimization. This strategy utilizes a biologic objective function based on TCP and NTCP instead of an objective function based on physical dose constraints.

Theoretical evaluation of the impact of diverse treatment conditions by calculation of the tumor control probability (TCP) of simulated cervical cancer Hyperthermia-Radiotherapy (HT-RT) treatments in-silico.

INTRODUCTION: Hyperthermia (HT) induces various cellular biological processes, such as repair impairment and direct HT cell killing. In this context, in-silico biophysical models that translate deviations in the treatment conditions into clinical outcome variations may be used to study the extent of such processes and their influence on combined hyperthermia plus radiotherapy (HT + RT) treatments under varying conditions. METHODS: An extended linear-quadratic model calibrated for SiHa and HeLa cell lines (cervical cancer) was used to theoretically study the impact of varying HT treatment conditions on radiosensitization and direct HT cell killing effect. Simulated patients were generated to compute the Tumor Control Probability (TCP) under different HT conditions (number of HT sessions, temperature and time interval), which were randomly selected within margins based on reported patient data. RESULTS: Under the studied conditions, model-based simulations suggested a treatment improvement with a total CEM43 thermal dose of approximately 10 min. Additionally, for a given thermal dose, TCP increased with the number of HT sessions. Furthermore, in the simulations, we showed that the TCP dependence on the temperature/time interval is more correlated with the mean value than with the minimum/maximum value and that comparing the treatment outcome with the mean temperature can be an excellent strategy for studying the time interval effect. CONCLUSION: The use of thermoradiobiological models allows us to theoretically study the impact of varying thermal conditions on HT + RT treatment outcomes. This approach can be used to optimize HT treatments, design clinical trials, and interpret patient data.

Treatment Facility Case Volume and Disparities in Outcomes of Congenital Diaphragmatic Hernia Cases.

INTRODUCTION: Congenital diaphragmatic hernia (CDH) is a life-threatening, prenatally diagnosed congenital anomaly. We aim to characterize care and outcomes of infants with CDH in Texas and the impact of treating facilities volume of care. METHODS: Retrospective cohort study using a state-wide Hospital Inpatient Discharge Public Use Data File was conducted (2013-2021). Neonates and infants <1 year of age were included using CDH ICD-9/ICD-10 codes. Neonates transferred to an outside hospital were excluded to avoid double-counting. Descriptive statistics, chi-square and logistic regression analysis were performed. RESULTS: Of 1314 CDH patient encounters identified, 728 (55%) occurred at 5 higher volume centers (HVC, >75 cases), 326 (25%) at 9 mid-volume centers (MVC, 20-75 cases) and 268 (20%) at 79 low volume centers (LVC, <20 cases). HVC had lower mortality rates (18%, MVC 22% vs LVC 27%; p = 0.011) despite treating sicker patients (extreme illness severity: HVC 71%, MVC 62% vs LVC 50%; p < 0.001) with longer length-of-stay (p < 0.001). Extracorporeal membrane oxygenation was used in 136 (10%) and provided primarily at HVC. LVC treated proportionately more non-white Hispanic patients (p < 0.001) and patients from counties along the Mexican border (p < 0.001). The predicted probability of mortality in CDH patients decreases with higher treatment facility CDH case volume, with a 0.5% decrease in the odds of mortality for every additional CDH case treated (p < 0.001). CONCLUSIONS: Patients treated in HVC have significantly lower mortality despite increased severity. Our data suggest minority populations may be disproportionately treated at LVC associated with worse outcomes. TYPE OF STUDY: Retrospective Prognosis Study. LEVEL OF EVIDENCE: Level II.

Making sense of breast cancer risk estimates.

Individual probabilistic assessments on the risk of cancer, primary or secondary, will not be understood by most patients. That is the essence of our arguments in this paper. Greater understanding can be achieved by extensive, intensive, and detailed counseling. But since probability itself is a concept that easily escapes our everyday intuition-consider the famous Monte Hall paradox-then it would also be wise to advise patients and potential patients, to not put undue weight on any probabilistic assessment. Such assessments can be of value to the epidemiologist in the investigation of different potential etiologies describing cancer evolution or to the clinical trialist as a way to maximize design efficiency. But to an ordinary individual we cannot anticipate that these assessments will be correctly interpreted.

Elucidating the relationship between metabolites and breast cancer: A Mendelian randomization study.

The evidence about the causal roles of metabolites in breast cancer is lacking. This study conducted a systematic evaluation of the potential causal relationship between 1091 human blood metabolites, 309 metabolite ratios, and the likelihood of developing breast cancer and its subtype by employing a two-sample bidirectional Mendelian randomization (MR) approach Four metabolites, including tryptophan betaine (Odds Ratio [OR] = 1.07, 95%CI = 1.04-1.10, Bonferroni-corrected P = 0.007), X-21312 (OR = 0.90, 95%CI = 0.86-0.94, Bonferroni-corrected P = 0.02), 3-bromo-5-chloro-2,6-dihydroxybenzoic acid (OR = 0.94, 95%CI = 0.91-0.96, Bonferroni-corrected P = 0.03) and X-18921 (OR = 0.96, 95%CI = 0.94-0.98, Bonferroni-corrected P = 0.04) were significantly associated with overall breast cancer using inverse-variance weighted (IVW) method. Tryptophan betaine was also significantly associated with estrogen receptor (ER)-positive breast cancer (OR = 1.08, 95%CI = 1.04-1.11, Bonferroni-corrected P = 0.03). X-23680 (OR = 1.10, 95%CI = 1.05-1.15, Bonferroni-corrected P = 0.04) and glycine to phosphate ratio (OR = 1.07, 95%CI = 1.04-1.10, Bonferroni-corrected P = 0.04) were associated with ER-negative breast cancer. Reverse MR analysis showed no significant associations between breast cancer and metabolites. This MR study indicated compelling evidence of a causal association between metabolites and the risk of breast cancer and its subtypes, underscoring the potential impact of metabolic interference on breast cancer risk and indicating the drug targets for breast cancer.

Cancer detection and classification using a simplified binary state vector machine.

Cancer is an invasive and malignant growth of cells and is known to be one of the most fatal diseases. Its early detection is essential for decreasing the mortality rate and increasing the probability of survival. This study presents an efficient machine learning approach based on the state vector machine (SVM) to diagnose and classify tumors into malignant or benign cancer using the online lymphographic data. Further, two types of neural network architectures are also implemented to evaluate the performance of the proposed SVM-based approach. The optimal structures of the classifiers are obtained by varying the architecture, topology, learning rate, and kernel function and recording the results' accuracy. The classifiers are trained with the preprocessed data examples after noise removal and tested on the unknown cases to diagnose each example as positive or negative. Further, the positive cases are classified into different stages including metastases, malign lymph, and fibrosis. The results are evaluated against the feed-forward and generalized regression neural networks. It is found that the proposed SVM-based approach significantly improves the early detection and classification accuracy in comparison to the experienced physicians and the other machine learning approaches. The proposed approach is robust and can perform sub-class divisions for multipurpose tasks. Experimental results demonstrate that the two-class SVM gives the best results and can effectively be used for the classification of cancer. It has outperformed all other classifiers with an average accuracy of 94.90%.

Tests of fit for the power function lognormal distribution.

In this study, tests of fit for the power function lognormal distribution is considered. The probability plot, probability plot correlation coefficient, and goodness-of-fit tests-the Kolmogorov-Smirnov (KS), Cramér-von Mises (CvM), and Anderson-Darling (AD) tests are provided. Tables of critical values are presented by using simulation techniques, and the AD test outperforms KS and CvM tests based on power comparisons. Finally, to illustrate these test procedures, we fit this distribution to the data which represent the survival times of 121 breast cancer patients from one hospital.

[Selective vital dyes in macular surgery : Do they increase the probability of intraoperative identification of the internal limiting membrane also for an experienced surgeon?] / Selektive Vitalfarbstoffe in der Makulachirurgie : Erhöhen sie die Wahrscheinlichkeit der intraoperativen ILM-Identifizierung auch bei einem erfahrenen Operateur?

BACKGROUND: Various vital dyes exist on the market for intraoperative internal limiting membrane (ILM) identification. The aim of this study was to verify the added value of these dyes for ILM identification and in the difficulty of ILM peeling during pars plana vitrectomy (ppV) by a single surgeon highly experienced in this operation. MATERIAL AND METHODS: In this study 400 ppV surgical reports involving ILM peeling were retrospectively analyzed. Intraoperative assessment of identification or difficulty of intraoperative ILM peeling had to be documented in the surgical report. The total group consisted of 2 cohorts each with 200 surgical reports (first cohort without selective vital dyes, period 2004-2006; second cohort with vital dyes in the majority of ppVs, period 2013-2020). RESULTS: The difference between both groups in terms of intraoperative identification of ILM was statistically significant (p < 0.001); however, no statistically significant difference (p = 0.951) was found between the two groups in terms of difficulty of ILM peeling. In logistic regression analysis neither patient gender, age, eye side, lens status nor posterior vitreous limiting membrane status were significantly associated with ILM identification. CONCLUSION: The introduction of intravital dyes represents a decisive advancement in retinal surgery. In the investigated sample this benefit was evident from two precisely defined surgical cohorts of a single highly experienced surgeon. This underlines the additional benefit of using selective vital dyes to identify ILM in macular surgery for less experienced surgeons.

Pseudo-value regression trees.

This paper presents a semi-parametric modeling technique for estimating the survival function from a set of right-censored time-to-event data. Our method, named pseudo-value regression trees (PRT), is based on the pseudo-value regression framework, modeling individual-specific survival probabilities by computing pseudo-values and relating them to a set of covariates. The standard approach to pseudo-value regression is to fit a main-effects model using generalized estimating equations (GEE). PRT extend this approach by building a multivariate regression tree with pseudo-value outcome and by successively fitting a set of regularized additive models to the data in the nodes of the tree. Due to the combination of tree learning and additive modeling, PRT are able to perform variable selection and to identify relevant interactions between the covariates, thereby addressing several limitations of the standard GEE approach. In addition, PRT include time-dependent effects in the node-wise models. Interpretability of the PRT fits is ensured by controlling the tree depth. Based on the results of two simulation studies, we investigate the properties of the PRT method and compare it to several alternative modeling techniques. Furthermore, we illustrate PRT by analyzing survival in 3,652 patients enrolled for a randomized study on primary invasive breast cancer.