Final Tree Research Articles

Separable Paravector Orientation Tensors for Enhancing Retinal Vessels.

Robust detection of retinal vessels remains an unsolved research problem, particularly in handling the intrinsic real-world challenges of highly imbalanced contrast between thick vessels and thin ones, inhomogeneous background regions, uneven illumination, and complex geometries of crossing/bifurcations. This paper presents a new separable paravector orientation tensor that addresses these difficulties by characterizing the enhancement of retinal vessels to be dependent on a nonlinear scale representation, invariant to changes in contrast and lighting, responsive for symmetric patterns, and fitted with elliptical cross-sections. The proposed method is built on projecting vessels as a 3D paravector valued function rotated in an alpha quarter domain, providing geometrical, structural, symmetric, and energetic features. We introduce an innovative symmetrical inhibitory scheme that incorporates paravector features for producing a set of directional contrast-independent elongated-like patterns reconstructing vessel tree in orientation tensors. By fitting constraint elliptical volumes via eigensystem analysis, the final vessel tree is produced with a strong and uniform response preserving various vessel features. The validation of proposed method on clinically relevant retinal images with high-quality results, shows its excellent performance compared to the state-of-the-art benchmarks and the second human observers.

Evolutionary Trend in Passiflora Species Revealed by RAPD-PCR, Gene Specific ITS and rbcl Markers

Background: The present study represents a preliminary analysis of genetic diversity among Passiflora species using amplified genotypic data of specific ITS and rbcL sequences and non-specific RAPD-PCR markers for investigation of the molecular phylogeny.
 Methods: The PCR-RAPD uses ten primers for polymorphic DNA, which are compiled on NTSYS software to construct dendrogram. The gene specific ITS and rbcL primers are used for specific amplification from genomic. The amplified ITS and rbcL markers assembled using Maximum Parsimony (MP) and Maximum likelihood (ML) methods. The BLAST, CLUSTAL W, and MEGA 6.0 have been used to conclude final genetic relation tree.
 Results: The PCR-RAPD primers translate 133 random amplified polymorphic DNA. NTSYS dendrogram placed P. vitifolia from Ramdaspeth and Shankar Nagar Nagpur, India in same clade (similarity coefficient 0.609) confirming same origin Nagpur India. Moreover, P. foetida from England is not coming in same clade with Indian P. foetida showing geographically intra-specific variation. In addition, the change in a constructed tree was observed with respect to change in phylogeny methods MS/ML. The ITS MP consensus tree is supported by strong 100 bootstrap value, clusters P. vitifolia (HNI and RNI) and P. foetida (UAI and UGI) in equivalent clade. However, no single species have been recovered using rbcL in MP and ML method. Thus, it is inference that rbcL have tendency to differentiate Passiflora species not allowing clustering around same species in same clade and the ITS region having parsimony informative sites that provide valid resolution and identification at inter-intra species level.
 Conclusions: The evaluation of properties of RAPD indicates 100% PCR success, sometimes with low rate of amplification. The ITS region found to be best for identification at inter-intra species, On the contrary, rbcL region is good to distinguished inter species, making it best as local DNA barcode for marking a Passiflora species in phylogenetic community.

Phylotranscriptomics and evolution of key genes for terpene biosynthesis in Pinaceae.

Pinaceae is the largest family of conifers, dominating forest ecosystems and serving as the backbone of northern, temperate and mountain forests. The terpenoid metabolism of conifers is responsive to pests, diseases, and environmental stress. Determining the phylogeny and evolution of terpene synthase genes in Pinaceae may shed light on early adaptive evolution. We used different inference methods and datasets to reconstruct the Pinaceae phylogeny based on our assembled transcriptomes. We identified the final species tree of Pinaceae by comparing and summarizing different phylogenetic trees. The genes encoding terpene synthase (TPS) and cytochrome P450 proteins in Pinaceae showed a trend of expansion compared with those in Cycas. Gene family analysis revealed that the number of TPS genes decreased while the number of P450 genes increased in loblolly pine. Expression profiles showed that TPSs and P450s were mainly expressed in leaf buds and needles, which may be the result of long-term evolution to protect these two vulnerable tissues. Our research provides insights into the phylogeny and evolution of terpene synthase genes in Pinaceae and offers some useful references for the investigation of terpenoids in conifers.

Deep reinforcement learning for cerebral anterior vessel tree extraction from 3D CTA images.

Extracting the cerebral anterior vessel tree of patients with an intracranial large vessel occlusion(LVO) is relevant to investigate potential biomarkers that can contribute to treatment decision making. The purpose of our work is to develop a method that can achieve this from routinely acquired computed tomography angiography(CTA)and computed tomography perfusion(CTP)images. To this end, we regard the anterior vessel tree as a set of bifurcations and connected centerlines. The method consists of a proximal policy optimization(PPO) based deep reinforcement learning (DRL) approach for tracking centerlines, a convolutional neural network based bifurcation detector, and a breadth-first vessel tree construction approach taking the tracking and bifurcation detection results as input. We experimentally determine the added values of various components of the tracker. Both DRL vessel tracking and CNN bifurcation detection were assessed in a cross validation experiment using 115 subjects. The anterior vessel tree formation was evaluated on an independent test set of 25 subjects, and compared to interobserver variation on a small subset of images. The DRL tracking result achieves a median overlapping rate until the first error (1.8mm off the reference standard) of 100, [46, 100]% on 8032 vessels over 115 subjects. The bifurcation detector reaches an average recall and precision of 76% and 87% respectively during the vessel tree formation process. The final vessel tree formation achieves a median recall of 68% and precision of 70%, which is in line with the interobserver agreement.

Time and type of pruning affect tree growth and yield in high-density olive orchards

The effect of different strategies and the time of pruning on olive vegetative-reproductive behavior was studied in a 5-year-old high-density orchard (1,250 trees ha−1, cv. ‘Ascolana tenera’). The treatments were: a) winter lateral pruning and topping, b) summer lateral pruning and winter topping, c) winter lateral pruning and summer topping and d) summer lateral pruning and topping. Selective spring pruning maintaining a conical shape of the canopy was used as the Control. Topping significantly maintained tree height recorded at the end of the 3-year trial in comparison to the Control (3.84 ± 0.45 m), in particular summer treatments maintained the final tree height at 3.0 ± 0.26 m. Summer topping and lateral pruning significantly reduced canopy volume compared to winter pruning and the Control. There was less vegetative re-sprouting in the control and, on the average, less for summer than for winter pruning in the same year of pruning. There were no significant differences between the treatments regarding the seasonal vegetative growth of 1-year-old mixed shoots. Fruit yield per tree was significantly higher with time-split pruning (part in winter and part in summer) with respect to conical-shape pruning of the canopy. Results suggest that the proposed management of pruning with winter lateral pruning and summer topping contributes to good vegetative-reproductive balance of young trees with high yield and a more compact shape.

Comparative biogeography of North American turtle faunas: Neogene regionalization

North America harbors substantial species diversity in nonmarineturtles (includes tortoises and terrapins), much of which arose in the Neogene Period (Miocene and Pliocene epochs) within Kinosternon, Emydinae, Trachemys, Pseudemys, Graptemys, and Gopherus. This diversity is distributed among 16 biogeographical provinces, but natural, hierarchical relations among provinces are unresolved. We used three-item analysis to identify such relations among provinces for these clades, following a recent, relatively complete phylogenetic reconstruction. The final three-item analysis identified 53,353 taxon‐area cladograms — free of paralogy and multiple areas on a single branch. The final intersection tree has a retention index of 73.6% and a completeness index of 75.4%, both indicating moderately strong congruence in patterns among turtle clades. All six turtle clades support some nodes on the intersection tree, which is divided into eastern and western forks. The crown group on the eastern fork includes four provinces almost entirely east of the Mississippi River drainage, whereas the western fork is split into wet tropics and aridlands sub-forks. The east-west transition zone spans the Mississippi River drainage and Great Plains. Our summary of divergence estimates and geological history suggests that although phylogenetic synchrony existed between select taxa, there was general asynchrony because provincial turtle faunas developed over an extended period (Neogene). Temperature-sensitive distributions of various taxa responded to climatic cycles by expanding during periods of warming, but contracting during periods of cooling. Grassland expansion and geomorphic change in the Neogene created provincial boundaries that, although sometimes crossed under favorable climates, more commonly acted as dispersal barriers. Despite asynchrony in faunal assembly, existence of shared patterns reveals natural relations among provinces, producing a regionalization for North American turtles, useful for understanding their evolution as well as the biogeography of North America.

Adjusting paid jobs to informal caregiving: a qualitative study in the Netherlands

ABSTRACT Increasingly many individuals are expected to combine informal caregiving with paid jobs. We assume that individuals may aim for reconciliation of both by engaging in job crafting, which concerns making adjustments to tasks, relations and cognitions regarding their jobs. This research note presents a qualitative study that aims to explore the experiences of individuals who combine informal caregiving with paid work, with respect to how they craft their jobs because of caregiving. Informal caregivers (19 individuals) were interviewed. After reading the interview transcripts, a preliminary coding tree was entered into NVivo 12. Iterations of examining its fit with the data, adapting and re-examining resulted in the final coding tree. Informal caregivers reported: seeking recognition and understanding by workplace stakeholders (relational crafting), adjusting work schedules, environments and tasks on a day-to-day basis (which includes task and cognitive crafting), preventing a work overload (this covers task and relational crafting) and focusing on how paid work and informal care can reinforce each other (this includes cognitive crafting). The results suggest that individuals use job crafting to combine caregiving and work. As such, job crafting by informal caregivers needs attention of researchers to investigate its preconditions and outcomes.

3D Graph-Based Individual-Tree Isolation (Treeiso) from Terrestrial Laser Scanning Point Clouds

Using terrestrial laser scanning (TLS) technology, forests can be digitized at the centimeter-level to enable fine-scale forest management. However, there are technical barriers to converting point clouds into individual-tree features or objects aligned with forest inventory standards due to noise, redundancy, and geometric complexity. A practical model treeiso based on the cut-pursuit graph algorithm was proposed to isolate individual-tree points from plot-level TLS scans. The treeiso followed the local-to-global segmentation scheme, which grouped points into small clusters, large segments, and final trees in a hierarchical manner. Seven tree attributes were investigated to understand the underlying determinants of isolation accuracy. Sensitivity analysis based on the PAWN index was performed using 10,000 parameter combinations to understand the treeiso’s parameter importance and model robustness. With sixteen reference TLS plot scans from various species, an average of 86% of all trees were detected. The mean intersection-over-union (mIoU) between isolated trees and reference trees was 0.82, which increased to 0.92 within the detected trees. Sensitivity analysis showed that only three parameters were needed for treeiso optimization, and it was robust against parameter variations. This new treeiso method is operationally simple and addresses the growing need for practical 3D tree segmentation tools.

Evaluation of Concurrent Chemoradiotherapy for Survival Outcomes in Patients With Synchronous Oligometastatic Esophageal Squamous Cell Carcinoma

The optimal treatment for and potential benefit populations of synchronous oligometastatic esophageal squamous cell carcinoma (SOESCC) remain unclear. To evaluate outcomes of concurrent chemoradiotherapy (CCRT) and to construct decision tree models for predicting the risk of progression and mortality in patients with SOESCC. This prognostic study included 532 patients with SOESCC who were treated at 2 cancer centers in China from January 2012 to December 2018 and consisted of a development cohort (n = 381) and a validation cohort (n = 151). Data were analyzed from March 2019 to December 2021. All patients received chemotherapy alone or CCRT. The primary end points of the study were progression-free survival (PFS) and overall survival (OS), and the secondary end points were locoregional control and treatment-related toxic effects. Propensity score matching was performed to control potential confounding factors. Cox regression was used to screen important explanatory variables. Decision trees for optimally partitioning patients were established using recursive partitioning analysis and were then subjected to internal and independent external validation. Among the 532 patients (median [range] age, 63 [32-82] years; 367 men [69.0%]), 292 patients received chemotherapy alone and 240 patients underwent CCRT. With a median (IQR) follow-up time of 37.0 (21.6-55.8) months, CCRT was associated with improved objective response rate (139 of 240 [57.9%] vs 123 of 292 [42.1%]; P < .001), median (IQR) PFS (9.7 [8.5-10.9] months vs 7.6 [6.6-8.6] months; P < .001), and median (IQR) OS (18.5 [16.1-20.9] months vs 15.2 [13.6-16.8] months; P < .001) compared with chemotherapy alone. Propensity score matching analysis verified the results. Cox multivariate analysis indicated that treatment modality (CCRT vs chemotherapy alone) was an independent prognostic factor related to PFS (hazard ratio, 0.69; 95% CI, 0.57-0.83; P < .001) and OS (hazard ratio, 0.75; 95% CI, 0.61-0.93; P = .008). The final decision trees divided patients with SOESCC into low-, intermediate-, and high-risk groups in both the internal and external validations, and the corresponding cumulative risk function curves had significant differences (all P < .001). Time-dependent maximum areas under receiver operating curves of decision trees for progression risk at 3 years and mortality risk at 5 years were 0.820 (95% CI, 0.693-0.948) and 0.894 (95% CI, 0.822-0.966), respectively. Calibration curves also demonstrated that the decision trees had favorable performance of risk stratification. In this study, CCRT vs chemotherapy alone as a first-line treatment for patients with SOESCC had superior survival. Patients with low risk had promising long-term survival based on the current treatment modality. The predictive information of the decision tree could provide accurate decision-making for the management of patients with SOESCC.

Untying the Gordian knot of plastid phylogenomic conflict: A case from ferns.

Phylogenomic studies based on plastid genome have resolved recalcitrant relationships among various plants, yet the phylogeny of Dennstaedtiaceae at the level of family and genera remains unresolved due to conflicting plastid genes, limited molecular data and incomplete taxon sampling of previous studies. The present study generated 30 new plastid genomes of Dennstaedtiaceae (9 genera, 29 species), which were combined with 42 publicly available plastid genomes (including 24 families, 27 genera, 42 species) to explore the evolution of Dennstaedtiaceae. In order to minimize the impact of systematic errors on the resolution of phylogenetic inference, we applied six strategies to generate 30 datasets based on CDS, intergenic spacers, and whole plastome, and two tree inference methods (maximum-likelihood, ML; and multispecies coalescent, MSC) to comprehensively analyze the plastome-scale data. Besides, the phylogenetic signal among all loci was quantified for controversial nodes using ML framework, and different topologies hypotheses among all datasets were tested. The species trees based on different datasets and methods revealed obvious conflicts at the base of the polypody ferns. The topology of the "CDS-codon-align-rm3" (CDS with the removal of the third codon) matrix was selected as the primary reference or summary tree. The final phylogenetic tree supported Dennstaedtiaceae as the sister group to eupolypods, and Dennstaedtioideae was divided into four clades with full support. This robust reconstructed phylogenetic backbone establishes a framework for future studies on Dennstaedtiaceae classification, evolution and diversification. The present study suggests considering plastid phylogenomic conflict when using plastid genomes. From our results, reducing saturated genes or sites can effectively mitigate tree conflicts for distantly related taxa. Moreover, phylogenetic trees based on amino acid sequences can be used as a comparison to verify the confidence of nucleotide-based trees.

Establishing a glutamine metabolism-based model for predicting the prognosis of low-grade glioma.

Background: The natural history of patients with low-grade glioma (LGG) varies widely, but most patients eventually deteriorate, leading to poor prognostic outcomes. We aim to develop biological models that can accurately predict the outcome of LGG prognosis. Methods: Prognostic genes for glutamine metabolism were searched by univariate Cox regression, and molecular typing was constructed. Functional enrichment analysis was done to evaluate potential prognostic-related pathways by analyzing differential genes in different subtypes. Enrichment scores of specific gene sets in different subtypes were measured by gene set enrichment analysis. Different immune infiltration levels among subtypes were calculated using algorithms such as CIBERSORT and ESTIMATE. Gene expression levels of prognostic-related gene signatures of glutamine metabolism phenotypes were used to construct a RiskScore model. Receiver operating characteristic curve, decision curve and calibration curve analyses were used to evaluate the reliability and validity of the risk model. The decision tree model was used to determine the best predictor variable ultimately. Results: We found that C1 had the worst prognosis and the highest level of immune infiltration, among which the highest macrophage infiltration can be found in the M2 stage. Moreover, most of the pathways associated with tumor development, such as MYC_TARGETS_V1 and EPITHELIAL_MESENCHYMAL_TRANSITION, were significantly enriched in C1. The wild-type IDH and MGMT hypermethylation were the most abundant in C1. A five-gene risk model related to glutamine metabolism phenotype was established with good performance in both training and validation datasets. The final decision tree demonstrated the RiskScore model as the most significant predictor of prognostic outcomes in individuals with LGG. Conclusion: The RiskScore model related to glutamine metabolism can be an exceedingly accurate predictor for LGG patients, providing valuable suggestions for personalized treatment.

Abstract 15607: Right Ventricular Diastolic Function: A Machine Learning-Based Echocardiographic Classification Scheme

Background: Right ventricular (RV) diastolic dysfunction is associated with heart failure and adverse cardiac events. Echocardiography guidelines include a limited assessment of RV diastology based on a scheme borrowed from the left ventricle, which is suboptimal. Hypothesis: An RV-specific classification scheme could be developed using echocardiographic parameters and machine learning to better reflect right-sided physiology and diagnose diastolic dysfunction. Methods: A total of 540 consecutive outpatients undergoing clinically-indicated echocardiograms at the Jewish General Hospital between March and December 2019 were prospectively recruited. Measurements relevant to RV diastology were acquired in triplicate by trained cardiac sonographers and verified by level III echocardiography staff and fellows. A sequential approach of unsupervised and supervised machine learning was used to develop a de novo classification model for RV diastology. Results: First, to prevent the use of redundant echocardiographic parameters within any given model, a feature-level clustering algorithm classified the 35 candidate parameters into mutually exclusive groups (for example, a group containing various size parameters). Second, a patient-level clustering algorithm analyzed all possible combinations of parameters in order to classify patients into three groups: normal, intermediate, and abnormal diastolic function. Third, a decision tree algorithm analyzed the top performing combinations in order to rank the importance of the echocardiographic parameters in predicting the diastology groups. Finally, after selecting the most predictive parameters - Ep, Ep/Ap, and IVC sniff - the final decision tree algorithm was generated. The prevalence of heart failure increased stepwise from 14% in the normal group, to 19% in the intermediate group, and 32% in the abnormal group. Conclusions: Our echocardiographic classification scheme for RV diastology is uniquely tailored to right-sided physiology and identifies patients with increasing rates of heart failure. Further study is warranted to validate this scheme in an independent population and examine its prognostic value for adverse cardiac events.

Prognosis and Benefit Factors of Definitive Concurrent Chemoradiotherapy for Patients with Oligometastatic Esophageal Squamous Cell Carcinoma

<h3>Purpose/Objective(s)</h3> To clarify the efficacy of definitive concurrent chemoradiotherapy (dCRT) for patients with synchronous oligometastatic esophageal squamous cell carcinoma (SOEC), and to construct the decision tree models for predicting the risk of progression and death. <h3>Materials/Methods</h3> A total of 532 patients with SOEC from two centers between January 2012 and December 2018 were enrolled in this study, including 292 patients who received chemotherapy alone (CT alone) and 240 patients who underwent dCRT, and each patient had ≤5 metastatic lesions and metastatic organs ≤3. Propensity score matching (PSM) was performed to control potential confounding factors. The primary endpoints were progression-free survival (PFS) and overall survival (OS), and the secondary endpoints were locoregional control and treatment-related toxicities. Cox regression was used to screen important explanatory variables. The decision trees that optimally partitioned patients with different PFS and OS rates were then built by recursive partitioning analysis (RPA). <h3>Results</h3> The median age of the entire cohort was 63 years (range, 32–82). With a median follow-up time of 37 months (IQR: 21.6-55.8), ORR was 42.1% for the CT alone group and 57.9% for the dCRT group (<i>P</i>=0.000). Median PFS of the CT alone group and dCRT group was 7.6 and 9.7 months, respectively (<i>P</i>=0.000), and median OS was 15.2 and 18.5 months, respectively (<i>P</i>=0.000). PSM analysis verified the same trend (all <i>P</i><0.05). Cox multivariate analysis indicated that treatment modality was an independent prognostic factor associated with PFS (<i>P</i>=0.000) and OS (<i>P</i>=0.008). The minimum complexity parameters (CP) corresponding to a standard deviation range of the minimum cross-validation error were 0.014 and 0.011 for PFS and OS, respectively. The final decision trees after pruning divided patients with SOEC into low, intermediate and high-risk groups by RPA, and the corresponding cumulative risk function curves were drawn by fitting the survival function of patients with different risk levels (all <i>P</i><0.01). Time-dependent receiver operating characteristic (ROC) curves and maximum area under the curves (AUC) of decision trees for progression and death risk were 0.807(95% CI: 0.719-0.894) and 0.911(95% CI: 0.850-0.973), respectively. Calibration curve also demonstrated that the decision trees have favorable hierarchical value. Other major treatment-related toxicities of grade III or higher included radiation pneumonitis (6.7%) and radiation esophagitis (7.1%) for the dCRT group. No treatment-related deaths were observed throughout the study period. <h3>Conclusion</h3> Compared with CT alone, dCRT as a first-line treatment for patients with SOEC has superior survival and amenable toxicity. With the continuous consolidation chemotherapy, OS also achieved significant improvement. The predictive information of our decision tree should be integrated into future clinical trials to provide effective and accurate decision-making for the management of patients with OEC.

S9 Development of a Clinical Screening Tool for Exocrine Pancreatic Insufficiency in Patients With Chronic Pancreatitis

Introduction: No accurate diagnostic tests exist for exocrine pancreatic insufficiency (EPI). As a result, EPI is misdiagnosed, including in chronic pancreatitis (CP), and there is an unmet need for clinical tools to help clinicians evaluate patients with CP at risk for EPI. The aim of this study was to generate a simple clinical tool to predict EPI in patients with CP. Methods: 49 variables from medical records of CP patients with or without EPI were entered into a Classification and Regression Tree (CART) model using the unimputed full analysis set and 4 other models (Logistic Regression with Least Absolute Shrinkage and Selection Operator regularization, Support Vector Machine, Random Forest, and Gradient Boosting Machine) using imputed full analysis set. EPI misclassification rate (mRate) served as primary metric to train the 5 prediction models. Each model’s generalizability was assessed using 5-fold cross-validation and the model that was best suited for clinician use based on mRate and other characteristics was selected. Results: Records of 274 patients with CP from 6 pancreatitis centers across the US were included. Among the 160 (58%)/114 (42%) patients with EPI/without EPI, 48%/55% were female, mean age was 55/52 years, 76%/69% were white, 29%/52% were never smokers, and 39%/53% were never alcohol users. All 5 prediction models demonstrated similar EPI mRates. The CART model was selected. The final CART decision tree includes 10 variables (in order of importance): smoking, pancreatic atrophy, steatorrhea, alcohol use, hemoglobin, serum vitamin D, pancreatic duct contour change, serum albumin, pancreatic parenchymal calcifications, and use of opioid analgesics. The mRate without and with 5-fold cross-validation of the CART was 0.153 (training error) and 0.314 (prediction error), and the area under the receiver operating characteristic curve was 0.889 and 0.682, respectively (Figure 1). Sensitivity/specificity without/with 5-fold cross validation was 0.888/0.789 and 0.794/0.535, respectively. Conclusion: A parsimonious set of 10 variables associated with EPI in CP was identified. The CART decision tree demonstrated good prediction performance for EPI and was selected as the most practical screening tool to aid clinicians in the assessment of EPI in patients with CP. Further external validation of the EPI screening tool is planned.Figure 1.: AUC-ROC of the CART model trained on the full analysis set and after 5-fold cross-validation. AUC, area under the curve; CART, classification and regression tree; CV, cross-validation; ROC, receiver operating characteristic.

Analyze the Influencing Factors and Predictability of Metabolic Syndrome in Schizophrenic Patients Treated with Olanzapine through Decision Tree Model

Abstract: Olanzapine is a typical antipsychotic that has demonstrated efficacy for the treatment of schizophrenia, but the patients treated with olanzapine usually appear it’s specific adverse events such as metabolism. It shows some factors that may affect metabolism such as age, BMI, high-dose antipsychotics, etc. Up to now the main predictors for metabolism in a patient with schizophrenia have not been comprehensively evaluated. Subjects and research methods: In this prospective cohort study, a total of 202 inpatients with schizophrenia at Vietnamese National Psychiatric Hospital No1 were included. The univariate regression and decision tree model were applied to find out the statistically significant factors. Results: The factors influencing metabolism were: baseline waist, baseline triglyceride, baseline HDL, age, baseline BMI, baseline metabolism, cholesterol ≥6.2 history, and schizophrenia duration. The final decision tree model included 3 important nodes: baseline waist &lt; 89 cm, baseline triglyceride &lt;3.1 mmol/l, age &lt;36. The predictive accuracy and other parameters were good to be able to apply for predictive purposes: accuracy 0.88, precision 0.90, recall 0.69, f1-score 0.78. Conclusion: The final model was good to predict no metabolism (0 code). In contrast, it is necessary to verify the metabolism status and have appropriate routine monitoring.&#x0D; Keywords: Metabolism, schizophrenia, olanzapine, predictive model, decision tree model.

Speed up discharge planning at the acute stroke unit: A development and external validation study for the early prediction of discharge home.

BackgroundTo reduce healthcare costs, it has become increasingly important to shorten the length of stay in acute stroke units. The goal of this study was to develop and externally validate a decision tree model applicable < 48 h poststroke for discharge home from an acute stroke unit with a short length of stay, and to assess the inappropriate home discharge rate.MethodsA prospective study including two samples of stroke patients admitted to an acute stroke unit. The outcome was discharge home (yes/no). A classification and regression tree analysis was performed in Sample 1. The model's performance was tested in Sample 2.ResultsIn total, 953 patients were included. The final decision tree included the patients' activities of daily living (ADL) performance <48 h poststroke, including motor function, cognition, and communication, and had an area under the curve (AUC) of 0.84 (95% confidence interval 0.76, 0.91). External validation resulted in an AUC of 0.74 (95% confidence interval 0.72, 0.77). None of the patients discharged home were re-admitted < 2 months after discharge to a hospital or admitted to a rehabilitation center for symptoms that had needed inpatient neurorehabilitation.ConclusionsThe developed decision tree shows acceptable external validity in predicting discharge home in a heterogeneous sample of stroke patients, only based on the patient's actual ADL performance <48 h poststroke. Importantly, discharge was safe, i.e., no re-hospitalization was registered. The tree's application to speed up discharge planning should now be further evaluated.

Mortality and Growth in a Sessile Oak [Quercus petraea (Matt.) Liebl.]-Dominated Young Stand Managed through Silvicultural Operations of Different Types and Intensities

A small-scale R&D project, including a block with four plots (P1-4) of 200 m2, was established in 2001 in a 15-year-old sessile oak-dominated stand, regenerated naturally through the application of group shelterwood cutting. In each plot, “potential” final crop trees were selected, based on vigour-quality-distribution criteria, and painted. Silvicultural interventions (cleaning-respacing and thinning), of different types and intensities were performed in P1-3 (P4 was kept as control) as well as P5 (500 m2), established in 2009, in 2001, 2004, and 2009. The mortality intensity between 2001 and 2019 was the highest in P4 and the lowest in P1, with the minimum stand density. Sessile oak showed the highest mortality, followed by Hungarian oak and Turkey oak. The fastest diameter growers were the “potential” final crop trees, their quadratic mean diameter (QMD) reaching values close to 20 cm at 35 (30-40) years in the plots with the lowest stand density. In all plots, trees have reached heights corresponding to the QMD of ca. 15 m, which are typical to a sessile oak stand of high productivity (production class II). The best solution for managing sessile oak young and medium-aged stands seems to be a “dynamic”, crop tree silviculture, with the most valuable individuals selected as “potential” final crop trees at the end of thicket stage. These trees should be favoured by subsequent heavy intensity thinning from above, in order to produce timber with as uniform as possible radial increments of 2-3 mm, as requested by veneer and high-quality saw log buyers.

Analysis of the genealogy process in forensic genetic genealogy.

The genealogy process is typically the most time-consuming part of-and a limiting factor in the success of-forensic genetic genealogy, which is a new approach to solving violent crimes and identifying human remains. We formulate a stochastic dynamic program that-given the list of matches and their genetic distances to the unknown target-chooses the best decision at each point in time: which match to investigate (i.e., find its ancestors and look for most recent common ancestors between the match and the target), which set of potential most recent common ancestors to descend from (i.e., find its descendants, with the goal of identifying a marriage between the maternal and paternal sides of the target's family tree), or whether to terminate the investigation. The objective is to maximize the probability of finding the target minus a cost associated with the expected size of the final family tree. We estimate the parameters of our model using data from 17 cases (eight solved, nine unsolved) from the DNA Doe Project. We assess the Proposed Strategy using simulated versions of the 17 DNA Doe Project cases, and compare it to a Benchmark Strategy that ranks matches by their genetic distance to the target and only descends from known common ancestors between a pair of matches. The Proposed Strategy solves cases ≈10-fold faster than the Benchmark Strategy, and does so by aggressively descending from a set of potential most recent common ancestors between the target and a match even when this set has a low probability of containing the correct most recent common ancestor. Our analysis provides a mathematical foundation for improving the genealogy process in forensic genetic genealogy.

An application of CART algorithms for detection of an association between VDR polymorphisms and reduced bone density in individuals with type 2 diabetes: a population-based cross-sectional study

Introduction: An important part of preventing major common diseases is identifying genetic factors that contribute to their occurrence. For the first time in our knowledge, we investigated the association between five polymorphisms of vitamin D receptor (VDR) gene (ApaI, BsmI, FokI, EcoRV, and TaqI) and low bone density/osteopenia/osteoporosis in individuals with type 2 diabetes using classification and regression tree (CART) algorithms. Methods: Data from 158 participants with T2D were used to develop the CART analysis. The binary output variable was "bone state" with low or normal values. Age and BMI (continuous variables), vitamin D deficiency (yes/no), and gender (binary variables), as well as the studied polymorphism of the VDR gene (categorical variables) all played a role in the explanatory model. A 5-fold cross-validation process was used for model validation. Results: Participants were divided into three groups: men, women, and both sexes. In all groups, age was the major factor predicting the low state in the final obtained tree model. The second most significant predictor in each model was BMI in both sexes (accuracy:75.30% ± 2.80%, AUC: 0.740 ± 0.064), EcoRV polymorphism in women (accuracy: 80.79% ± 6.58%, AUC:0.785 ± 0.063), and TaqI polymorphism in men (accuracy: 76.36% ± 3.05%, AUC:0.706 ± 0.125). Conclusion: Model validation of the final tree models demonstrated that the use of CART algorithms could be an acceptable technique for risk factors of osteoporosis among individuals with T2D. Our recommendation is to conduct more population-based studies. We hope this study will serve as a basis for future research.

Performance-Driven X-Architecture Routing Algorithm for Artificial Intelligence Chip Design in Smart Manufacturing

The new 7-nm Artificial Intelligence (AI) chip is an important milestone recently announced by the IBM research team, with a very important optimization goal of performance. This chip technology can be extended to various business scenarios in the Internet of Things. As the basic model for very large scale integration routing, the Steiner minimal tree can be used in various practical problems, such as wirelength optimization and timing closure. Further considering the X-architecture and the routing resources within obstacles, an effective performance-driven X-architecture routing algorithm for AI chip design in smart manufacturing is proposed to improve the delay performance of the chip. First, a special particle swarm optimization algorithm is presented to solve the discrete length-restricted X-architecture Steiner minimum tree problem in combination with genetic operations, and a particle encoding scheme is presented to encode each particle into an initial routing tree. Second, two lookup tables based on pins and obstacles are established to provide a fast information query for the whole algorithm flow. Third, a strategy of candidate point selection is designed to make the particles satisfy the constraints. Finally, a refinement strategy is implemented to further improve the quality of the final routing tree. Compared with other state-of-the-art algorithms, the proposed algorithm achieves a better total wirelength, which is an important index of performance, thus better satisfying the demand for delay performance of AI chip design in smart manufacturing.