Could machine learning improve the prediction of pelvic nodal status of prostate cancer patients? Preliminary results of a pilot study.

We tested and compared performances of Roach formula, Partin tables and of three Machine Learning (ML) based algorithms based on decision trees in identifying N+ prostate cancer (PC). 1,555 cN0 and 50 cN+ PC were analyzed. Results were also verified on an independent population of 204 operated cN0 patients, with a known pN status (187 pN0, 17 pN1 patients). ML performed better, also when tested on the surgical population, with accuracy, specificity, and sensitivity ranging between 48-86%, 35-91%, and 17-79%, respectively. ML potentially allows better prediction of the nodal status of PC, potentially allowing a better tailoring of pelvic irradiation.

Prediction of multiple infections after severe burn trauma: a prospective cohort study.

OBJECTIVE: To develop predictive models for early triage of burn patients based on hypersusceptibility to repeated infections. BACKGROUND: Infection remains a major cause of mortality and morbidity after severe trauma, demanding new strategies to combat infections. Models for infection prediction are lacking. METHODS: Secondary analysis of 459 burn patients (≥16 years old) with 20% or more total body surface area burns recruited from 6 US burn centers. We compared blood transcriptomes with a 180-hour cutoff on the injury-to-transcriptome interval of 47 patients (≤1 infection episode) to those of 66 hypersusceptible patients [multiple (≥2) infection episodes (MIE)]. We used LASSO regression to select biomarkers and multivariate logistic regression to built models, accuracy of which were assessed by area under receiver operating characteristic curve (AUROC) and cross-validation. RESULTS: Three predictive models were developed using covariates of (1) clinical characteristics; (2) expression profiles of 14 genomic probes; (3) combining (1) and (2). The genomic and clinical models were highly predictive of MIE status [AUROCGenomic = 0.946 (95% CI: 0.906-0.986); AUROCClinical = 0.864 (CI: 0.794-0.933); AUROCGenomic/AUROCClinical P = 0.044]. Combined model has an increased AUROCCombined of 0.967 (CI: 0.940-0.993) compared with the individual models (AUROCCombined/AUROCClinical P = 0.0069). Hypersusceptible patients show early alterations in immune-related signaling pathways, epigenetic modulation, and chromatin remodeling. CONCLUSIONS: Early triage of burn patients more susceptible to infections can be made using clinical characteristics and/or genomic signatures. Genomic signature suggests new insights into the pathophysiology of hypersusceptibility to infection may lead to novel potential therapeutic or prophylactic targets.

Reliability of maximal grip strength measurements and grip strength recovery following a stroke.

STUDY DESIGN: Clinical measurement. PURPOSE: The test-retest reliability of maximal grip strength measurements (MGSM) is examined in subjects for 12 weeks post-stroke together with maximal grip strength recovery and the maximal-grip and upper-extremity strength measurements' relationship with capacity and performance test scores. METHODS: A Jamar dynamometer and the Motricity Index (MI) were used for strength measurements. The Chedoke Arm and Hand Activity Inventory and ABILHAND questionnaire for evaluating capacities and performances. RESULTS: MGSM were reliable (Intraclass Correlation Coefficients = 0.97-0.99, Minimal Detectable Differences = 2.73-4.68 kg). Among the 34 participants, 47% did not have a measurable grip strength one week post-stroke but 50% of these recovered some strength within the first eight weeks. The MGSM and MI scores were correlated with scores of tests of capacity and performance (Spearman's Rank Correlation Coefficients = 0.69-0.94). CONCLUSIONS: MGSM are reliable in the first weeks after a stroke. LEVEL OF EVIDENCE: N/A.

Overcoming limitations of modelling rare species by using ensembles of small models

1. Species distribution models (SDMs) have become a standard tool in ecology and applied conservation biology. Modelling rare and threatened species is particularly important for conservation purposes. However, modelling rare species is difficult because the combination of few occurrences and many predictor variables easily leads to model overfitting. A new strategy using ensembles of small models was recently developed in an attempt to overcome this limitation of rare species modelling and has been tested successfully for only a single species so far. Here, we aim to test the approach more comprehensively on a large number of species including a transferability assessment. 2. For each species numerous small (here bivariate) models were calibrated, evaluated and averaged to an ensemble weighted by AUC scores. These 'ensembles of small models' (ESMs) were compared to standard Species Distribution Models (SDMs) using three commonly used modelling techniques (GLM, GBM, Maxent) and their ensemble prediction. We tested 107 rare and under-sampled plant species of conservation concern in Switzerland. 3. We show that ESMs performed significantly better than standard SDMs. The rarer the species, the more pronounced the effects were. ESMs were also superior to standard SDMs and their ensemble when they were independently evaluated using a transferability assessment. 4. By averaging simple small models to an ensemble, ESMs avoid overfitting without losing explanatory power through reducing the number of predictor variables. They further improve the reliability of species distribution models, especially for rare species, and thus help to overcome limitations of modelling rare species.

Reliability and validity of nonverbal thin slices in social interactions

Four studies investigated the reliability and validity of thin slices of nonverbal behavior from social interactions including (1) how well individual slices of a given behavior predict other slices in the same interaction; (2) how well a slice of a given behavior represents the entirety of that behavior within an interaction; (3) how long a slice is necessary to sufficiently represent the entirety of a behavior within an interaction; (4) which slices best capture the entirety of behavior, across different behaviors; and (5) which behaviors (of six measured behaviors) are best captured by slices. Notable findings included strong reliability and validity for thin slices of gaze and nods, and that a 1.5 min slice from the start of an interaction may adequately represent some behaviors. Results provide useful information to researchers making decisions about slice measurement of behavior.

Intracranial Aneurysms: Wall Motion Analysis for Prediction of Rupture.

Intracranial aneurysms are a common pathologic condition with a potential severe complication: rupture. Effective treatment options exist, neurosurgical clipping and endovascular techniques, but guidelines for treatment are unclear and focus mainly on patient age, aneurysm size, and localization. New criteria to define the risk of rupture are needed to refine these guidelines. One potential candidate is aneurysm wall motion, known to be associated with rupture but difficult to detect and quantify. We review what is known about the association between aneurysm wall motion and rupture, which structural changes may explain wall motion patterns, and available imaging techniques able to analyze wall motion.

Validation of and comparison between a semidistributed rainfall-runoff hydrological model (PREVAH) and a spatially distributed snow-evolution model (SnowModel) for snow cover prediction in mountain ecosystems

Snow cover is an important control in mountain environments and a shift of the snow-free period triggered by climate warming can strongly impact ecosystem dynamics. Changing snow patterns can have severe effects on alpine plant distribution and diversity. It thus becomes urgent to provide spatially explicit assessments of snow cover changes that can be incorporated into correlative or empirical species distribution models (SDMs). Here, we provide for the first time a with a lower overestimation comparison of two physically based snow distribution models (PREVAH and SnowModel) to produce snow cover maps (SCMs) at a fine spatial resolution in a mountain landscape in Austria. SCMs have been evaluated with SPOT-HRVIR images and predictions of snow water equivalent from the two models with ground measurements. Finally, SCMs of the two models have been compared under a climate warming scenario for the end of the century. The predictive performances of PREVAH and SnowModel were similar when validated with the SPOT images. However, the tendency to overestimate snow cover was slightly lower with SnowModel during the accumulation period, whereas it was lower with PREVAH during the melting period. The rate of true positives during the melting period was two times higher on average with SnowModel with a lower overestimation of snow water equivalent. Our results allow for recommending the use of SnowModel in SDMs because it better captures persisting snow patches at the end of the snow season, which is important when modelling the response of species to long-lasting snow cover and evaluating whether they might survive under climate change.

Prediction of Complications After Pancreaticoduodenectomy: Validation of a Postoperative Complication Score.

OBJECTIVES: Pancreatic surgery remains associated with important morbidity. Efforts are most commonly concentrated on decreasing postoperative morbidity, but early detection of patients at risk could be another valuable strategy. A simple prognostic score has recently been published. This study aimed to validate this score and discuss possible clinical implications. METHODS: From 2000 to 2012, 245 patients underwent a pancreaticoduodenectomy. Complications were graded according to the Dindo-Clavien Classification. The Braga score is based on American Society of Anesthesiologists score, pancreatic texture, Wirsung duct diameter, and blood loss. An overall risk score (0-15) can be calculated for each patient. Score discriminant power was calculated using a receiver operating characteristic curve. RESULTS: Major complications occurred in 31% of patients compared with 17% in Braga's data. Pancreatic texture and blood loss were independently statistically significant for increased morbidity. Areas under the curve were 0.95 and 0.99 for 4-risk categories and for individual scores, respectively. CONCLUSIONS: The Braga score discriminates well between minor and major complications. Our validation suggests that it can be used as a prognostic tool for major complications after pancreaticoduodenectomy. The clinical implications, that is, whether postoperative treatment strategies should be adapted according to the patient's individual risk, remain to be elucidated.

Association between parental history and genetic risk scores for coronary heart disease prediction: The population-based CoLaus study.

BACKGROUND AND AIMS: Parental history (PH) and genetic risk scores (GRSs) are separately associated with coronary heart disease (CHD), but evidence regarding their combined effects is lacking. We aimed to evaluate the joint associations and predictive ability of PH and GRSs for incident CHD. METHODS: Data for 4283 Caucasians were obtained from the population-based CoLaus Study, over median follow-up time of 5.6 years. CHD was defined as incident myocardial infarction, angina, percutaneous coronary revascularization or bypass grafting. Single nucleotide polymorphisms for CHD identified by genome-wide association studies were used to construct unweighted and weighted versions of three GRSs, comprising of 38, 53 and 153 SNPs respectively. RESULTS: PH was associated with higher values of all weighted GRSs. After adjustment for age, sex, smoking, diabetes, systolic blood pressure, low and high density lipoprotein cholesterol, PH was significantly associated with CHD [HR 2.61, 95% CI (1.47-4.66)] and further adjustment for GRSs did not change this estimate. Similarly, one standard deviation change of the weighted 153-SNPs GRS was significantly associated with CHD [HR 1.50, 95% CI (1.26-1.80)] and remained so, after further adjustment for PH. The weighted, 153-SNPs GRS, but not PH, modestly improved discrimination [(C-index improvement, 0.016), p = 0.048] and reclassification [(NRI improvement, 8.6%), p = 0.027] beyond cardiovascular risk factors. After including both the GRS and PH, model performance improved further [(C-index improvement, 0.022), p = 0.006]. CONCLUSION: After adjustment for cardiovascular risk factors, PH and a weighted, polygenic GRS were jointly associated with CHD and provided additive information for coronary events prediction.

Predicting recurrence after unprovoked venous thromboembolism: prospective validation of the updated Vienna Prediction Model.

The updated Vienna Prediction Model for estimating recurrence risk after an unprovoked venous thromboembolism (VTE) has been developed to identify individuals at low risk for VTE recurrence in whom anticoagulation (AC) therapy may be stopped after 3 months. We externally validated the accuracy of the model to predict recurrent VTE in a prospective multicenter cohort of 156 patients aged ≥65 years with acute symptomatic unprovoked VTE who had received 3 to 12 months of AC. Patients with a predicted 12-month risk within the lowest quartile based on the updated Vienna Prediction Model were classified as low risk. The risk of recurrent VTE did not differ between low- vs higher-risk patients at 12 months (13% vs 10%; P = .77) and 24 months (15% vs 17%; P = 1.0). The area under the receiver operating characteristic curve for predicting VTE recurrence was 0.39 (95% confidence interval [CI], 0.25-0.52) at 12 months and 0.43 (95% CI, 0.31-0.54) at 24 months. In conclusion, in elderly patients with unprovoked VTE who have stopped AC, the updated Vienna Prediction Model does not discriminate between patients who develop recurrent VTE and those who do not. This study was registered at www.clinicaltrials.gov as #NCT00973596.

A Meta-Analysis of Trabecular Bone Score in Fracture Risk Prediction and Its Relationship to FRAX.

Trabecular bone score (TBS) is a gray-level textural index of bone microarchitecture derived from lumbar spine dual-energy X-ray absorptiometry (DXA) images. TBS is a bone mineral density (BMD)-independent predictor of fracture risk. The objective of this meta-analysis was to determine whether TBS predicted fracture risk independently of FRAX probability and to examine their combined performance by adjusting the FRAX probability for TBS. We utilized individual-level data from 17,809 men and women in 14 prospective population-based cohorts. Baseline evaluation included TBS and the FRAX risk variables, and outcomes during follow-up (mean 6.7 years) comprised major osteoporotic fractures. The association between TBS, FRAX probabilities, and the risk of fracture was examined using an extension of the Poisson regression model in each cohort and for each sex and expressed as the gradient of risk (GR; hazard ratio per 1 SD change in risk variable in direction of increased risk). FRAX probabilities were adjusted for TBS using an adjustment factor derived from an independent cohort (the Manitoba Bone Density Cohort). Overall, the GR of TBS for major osteoporotic fracture was 1.44 (95% confidence interval [CI] 1.35-1.53) when adjusted for age and time since baseline and was similar in men and women (p > 0.10). When additionally adjusted for FRAX 10-year probability of major osteoporotic fracture, TBS remained a significant, independent predictor for fracture (GR = 1.32, 95% CI 1.24-1.41). The adjustment of FRAX probability for TBS resulted in a small increase in the GR (1.76, 95% CI 1.65-1.87 versus 1.70, 95% CI 1.60-1.81). A smaller change in GR for hip fracture was observed (FRAX hip fracture probability GR 2.25 vs. 2.22). TBS is a significant predictor of fracture risk independently of FRAX. The findings support the use of TBS as a potential adjustment for FRAX probability, though the impact of the adjustment remains to be determined in the context of clinical assessment guidelines. © 2015 American Society for Bone and Mineral Research.