Development of clinical screening tool for exocrine pancreatic insufficiency in patients with definite chronic pancreatitis.

BACKGROUND/OBJECTIVES: No simple, accurate diagnostic tests exist for exocrine pancreatic insufficiency (EPI), and EPI remains underdiagnosed in chronic pancreatitis (CP). We sought to develop a digital screening tool to assist clinicians to predict EPI in patients with definite CP. METHODS: This was a retrospective case-control study of patients with definite CP with/without EPI. Overall, 49 candidate predictor variables were utilized to train a Classification and Regression Tree (CART) model to rank all predictors and select a parsimonious set of predictors for EPI status. Five-fold cross-validation was used to assess generalizability, and the full CART model was compared with 4 additional predictive models. EPI misclassification rate (mRate) served as primary endpoint metric. RESULTS: 274 patients with definite CP from 6 pancreatitis centers across the United States were included, of which 58 % had EPI based on predetermined criteria. The optimal CART decision tree included 10 variables. The mRate without/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. Sensitivity and specificity without/with 5-fold cross-validation was 0.888/0.789 and 0.794/0.535, respectively. A trained second CART without pancreas imaging variables (n = 6), yielded 8 variables. Training error/prediction error was 0.190/0.351; sensitivity was 0.869/0.650, and specificity was 0.728/0.649, each without/with 5-fold cross-validation. CONCLUSION: We developed two CART models that were integrated into one digital screening tool to assess for EPI in patients with definite CP and with two to six input variables needed for predicting EPI status.

Magnetic resonance imaging-based prediction models for differentiating intraspinal schwannomas from meningiomas: classification and regression tree and random forest analysis.

Background: Due to the variations in surgical approaches and prognosis between intraspinal schwannomas and meningiomas, it is crucial to accurately differentiate between the two prior to surgery. Currently, there is limited research exploring the implementation of machine learning (ML) methods for distinguishing between these two types of tumors. This study aimed to establish a classification and regression tree (CART) model and a random forest (RF) model for distinguishing schwannomas from meningiomas. Methods: We retrospectively collected 88 schwannomas (52 males and 36 females) and 51 meningiomas (10 males and 41 females) who underwent magnetic resonance imaging (MRI) examinations prior to the surgery. Simple clinical data and MRI imaging features, including age, sex, tumor location and size, T1-weighted images (T1WI) and T2-weighted images (T2WI) signal characteristics, degree and pattern of enhancement, dural tail sign, ginkgo leaf sign, and intervertebral foramen widening (IFW), were reviewed. Finally, a CART model and RF model were established based on the aforementioned features to evaluate their effectiveness in differentiating between the two types of tumors. Meanwhile, we also compared the performance of the ML models to the radiologists. The receiver operating characteristic (ROC) curve, accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were used to evaluate the models and clinicians' discrimination performance. Results: Our investigation reveals significant variations in ten out of 11 variables in the training group and five out of 11 variables in the test group when comparing schwannomas and meningiomas (P<0.05). Ultimately, the CART model incorporated five variables: enhancement pattern, the presence of IFW, tumor location, maximum diameter, and T2WI signal intensity (SI). The RF model combined all 11 variables. The CART model, RF model, radiologist 1, and radiologist 2 achieved an area under the curve (AUC) of 0.890, 0.956, 0.681, and 0.723 in the training group, and 0.838, 0.922, 0.580, and 0.659 in the test group, respectively. Conclusions: The RF prediction model exhibits more exceptional performance than an experienced radiologist in discriminating intraspinal schwannomas from meningiomas. The RF model seems to be better in discriminating the two tumors than the CART model.

Further Clarification of Pain Management Complexity in Radiotherapy: Insights from Modern Statistical Approaches.

BACKGROUND: The primary objective of this study was to assess the adequacy of analgesic care in radiotherapy (RT) patients, with a secondary objective to identify predictive variables associated with pain management adequacy using a modern statistical approach, integrating the Least Absolute Shrinkage and Selection Operator (LASSO) algorithm and the Classification and Regression Tree (CART) analysis. METHODS: This observational, multicenter cohort study involved 1387 patients reporting pain or taking analgesic drugs from 13 RT departments in Italy. The Pain Management Index (PMI) served as the measure for pain control adequacy, with a PMI score < 0 indicating suboptimal management. Patient demographics, clinical status, and treatment-related factors were examined to discern the predictors of pain management adequacy. RESULTS: Among the analyzed cohort, 46.1% reported inadequately managed pain. Non-cancer pain origin, breast cancer diagnosis, higher ECOG Performance Status scores, younger patient age, early assessment phase, and curative treatment intent emerged as significant determinants of negative PMI from the LASSO analysis. Notably, pain management was observed to improve as RT progressed, with a greater discrepancy between cancer (33.2% with PMI < 0) and non-cancer pain (73.1% with PMI < 0). Breast cancer patients under 70 years of age with non-cancer pain had the highest rate of negative PMI at 86.5%, highlighting a potential deficiency in managing benign pain in younger patients. CONCLUSIONS: The study underscores the dynamic nature of pain management during RT, suggesting improvements over the treatment course yet revealing specific challenges in non-cancer pain management, particularly among younger breast cancer patients. The use of advanced statistical techniques for analysis stresses the importance of a multifaceted approach to pain management, one that incorporates both cancer and non-cancer pain considerations to ensure a holistic and improved quality of oncological care.

Identification and prediction of association patterns between nutrient intake and anemia using machine learning techniques: results from a cross-sectional study with university female students from Palestine.

PURPOSE: This study utilized data mining and machine learning (ML) techniques to identify new patterns and classifications of the associations between nutrient intake and anemia among university students. METHODS: We employed K-means clustering analysis algorithm and Decision Tree (DT) technique to identify the association between anemia and vitamin and mineral intakes. We normalized and balanced the data based on anemia weighted clusters for improving ML models' accuracy. In addition, t-tests and Analysis of Variance (ANOVA) were performed to identify significant differences between the clusters. We evaluated the models on a balanced dataset of 755 female participants from the Hebron district in Palestine. RESULTS: Our study found that 34.8% of the participants were anemic. The intake of various micronutrients (i.e., folate, Vit A, B5, B6, B12, C, E, Ca, Fe, and Mg) was below RDA/AI values, which indicated an overall unbalanced malnutrition in the present cohort. Anemia was significantly associated with intakes of energy, protein, fat, Vit B1, B5, B6, C, Mg, Cu and Zn. On the other hand, intakes of protein, Vit B2, B5, B6, C, E, choline, folate, phosphorus, Mn and Zn were significantly lower in anemic than in non-anemic subjects. DT classification models for vitamins and minerals (accuracy rate: 82.1%) identified an inverse association between intakes of Vit B2, B3, B5, B6, B12, E, folate, Zn, Mg, Fe and Mn and prevalence of anemia. CONCLUSIONS: Besides the nutrients commonly known to be linked to anemia-like folate, Vit B6, C, B12, or Fe-the cluster analyses in the present cohort of young female university students have also found choline, Vit E, B2, Zn, Mg, Mn, and phosphorus as additional nutrients that might relate to the development of anemia. Further research is needed to elucidate if the intake of these nutrients might influence the risk of anemia.

Dissecting Alzheimer's Disease Risk in Asian American Elders: A Classification and Regression Tree Approach.

Background: Alzheimer's disease (AD) poses a growing public health challenge, particularly with an aging population. While extensive research has explored the relationships between AD, socio-demographic factors, and cardiovascular risk factors, a notable gap exists in understanding these connections within the Asian American elderly population. Objective: This study aims to address this gap by employing the Classification and Regression Tree (CART) approach to investigate the intricate interplay of socio-demographic variables, cardiovascular risk factors, sleep patterns, prior antidepressant use, and AD among Asian American elders. Methods: Data from the 2017 Uniform Data Set, provided by the National Alzheimer's Coordinating Center, were analyzed, focusing on a sample of Asian American elders (n = 4,343). The analysis utilized the Classification and Regression Tree (CART) approach. Results: CART analysis identified critical factors, including levels of independence, specific age thresholds (73.5 and 84.5 years), apnea, antidepressant use, and body mass index, as significantly associated with AD risk. Conclusions: These findings have far-reaching implications for future research, particularly in examining the roles of gender, cultural nuances, socio-demographic factors, and cardiovascular risk elements in AD within the Asian American elderly population. Such insights can inform tailored interventions, improved healthcare access, and culturally sensitive policies to address the complex challenges posed by AD in this community.

A Day 14 Endpoint for Acute GVHD Clinical Trials.

The overall response rate (ORR) 28 days after treatment has been adopted as the primary endpoint for clinical trials of acute graft versus host disease (GVHD). However, physicians often need to modify immunosuppression earlier than day (D) 28, and non-relapse mortality (NRM) does not always correlate with ORR at D28. We studied 1144 patients that received systemic treatment for GVHD in the Mount Sinai Acute GVHD International Consortium (MAGIC) and divided them into a training set (n=764) and a validation set (n=380). We used a recursive partitioning algorithm to create a Mount Sinai model that classifies patients into favorable or unfavorable groups that predicted 12 month NRM according to overall GVHD grade at both onset and D14. In the Mount Sinai model grade II GVHD at D14 was unfavorable for grade III/IV GVHD at onset and predicted NRM as well as the D28 standard response model. The MAGIC algorithm probability (MAP) is a validated score that combines the serum concentrations of suppression of tumorigenicity 2 (ST2) and regenerating islet-derived 3-alpha (REG3α) to predict NRM. Inclusion of the D14 MAP biomarker score with the D14 Mount Sinai model created three distinct groups (good, intermediate, poor) with strikingly different NRM (8%, 35%, 76% respectively). This D14 MAGIC model displayed better AUC, sensitivity, positive and negative predictive value, and net benefit in decision curve analysis compared to the D28 standard response model. We conclude that this D14 MAGIC model could be useful in therapeutic decisions and may offer an improved endpoint for clinical trials of acute GVHD treatment.

Interactions of genetic variations in FAS, GJB2 and PTPRN2 are associated with noise-induced hearing loss: a case-control study in China.

BACKGROUND: This study aimed to screen and validate noise-induced hearing loss (NIHL) associated single nucleotide polymorphisms (SNPs), construct genetic risk prediction models, and evaluate higher-order gene-gene, gene-environment interactions for NIHL in Chinese population. METHODS: First, 83 cases and 83 controls were recruited and 60 candidate SNPs were genotyped. Then SNPs with promising results were validated in another case-control study (153 cases and 252 controls). NIHL-associated SNPs were identified by logistic regression analysis, and a genetic risk model was constructed based on the genetic risk score (GRS), and classification and regression tree (CART) analysis was used to evaluate interactions among gene-gene and gene-environment. RESULTS: Six SNPs in five genes were significantly associated with NIHL risk (p < 0.05). A positive dose-response relationship was found between GRS values and NIHL risk. CART analysis indicated that strongest interaction was among subjects with age ≥ 45 years and cumulative noise exposure ≥ 95 [dB(A)·years], without personal protective equipment, and carried GJB2 rs3751385 (AA/AB) and FAS rs1468063 (AA/AB) (OR = 10.038, 95% CI = 2.770, 47.792), compared with the referent group. CDH23, FAS, GJB2, PTPRN2 and SIK3 may be NIHL susceptibility genes. CONCLUSION: GRS values may be utilized in the evaluation of the cumulative effect of genetic risk for NIHL based on NIHL-associated SNPs. Gene-gene, gene-environment interaction patterns play an important role in the incidence of NIHL.

Assessing Derawan Island's Coral Reefs over Two Decades: A Machine Learning Classification Perspective.

This study aims to understand the dynamic changes in the coral reef habitats of Derawan Island over two decades (2003, 2011, and 2021) using advanced machine learning classification techniques. The motivation stems from the urgent need for accurate, detailed environmental monitoring to inform conservation strategies, particularly in ecologically sensitive areas like coral reefs. We employed non-parametric machine learning algorithms, including Random Forest (RF), Support Vector Machine (SVM), and Classification and Regression Tree (CART), to assess spatial and temporal changes in coral habitats. Our analysis utilized high-resolution data from Landsat 9, Landsat 7, Sentinel-2, and Multispectral Aerial Photos. The RF algorithm proved to be the most accurate, achieving an accuracy of 71.43% with Landsat 9, 73.68% with Sentinel-2, and 78.28% with Multispectral Aerial Photos. Our findings indicate that the classification accuracy is significantly influenced by the geographic resolution and the quality of the field and satellite/aerial image data. Over the two decades, there was a notable decrease in the coral reef area from 2003 to 2011, with a reduction to 16 hectares, followed by a slight increase in area but with more heterogeneous densities between 2011 and 2021. The study underscores the dynamic nature of coral reef habitats and the efficacy of machine learning in environmental monitoring. The insights gained highlight the importance of advanced analytical methods in guiding conservation efforts and understanding ecological changes over time.

Diagnostic value of 18F-fluorodeoxyglucose positron emission tomography and computed tomography for differentiating polymyalgia rheumatica and rheumatoid arthritis: Using classification and regression tree analysis.

OBJECTIVES: Determining which sites were important to differentiate polymyalgia rheumatica (PMR) from rheumatoid arthritis (RA) using 18F-fluorodeoxyglucose (FDG) positron emission tomography and computed tomography (PET-CT) is challenging. METHODS: Patients with PMR or RA who were undergoing PET-CT were recruited at two mutual-aid hospitals in Japan between 2009 and 2018. Classification and regression tree (CART) analyses were performed to identify FDG uptake patterns that differentiated PMR from RA. RESULTS: We enrolled 35 patients with PMR and 46 patients with RA. Univariate CART analysis showed that FDG uptake in the shoulder joints, spinous processes of the lumbar vertebrae, pubic symphysis, sternoclavicular joints, ischial tuberosities, greater trochanters, and hip joints differentiated PMR from RA. Multivariate CART analysis revealed that FDG uptake by at least one of the ischial tuberosities had the highest diagnostic value for distinguishing PMR from RA (sensitivity, 77.1%; specificity, 82.6%). We performed the same CART analysis to patients who had not undergone treatment (PMR, n = 28; RA, n = 9). Similar results were obtained, and sensitivity and specificity were increased (sensitivity, 89.3%; specificity, 88.8%). CONCLUSIONS: In PET-CT, FDG uptake by at least one of the ischial tuberosities best discriminates between PMR and RA.

Developing radiology diagnostic tools for pulmonary fibrosis using machine learning methods.

BACKGROUND: Accurate and prompt diagnosis of the different patterns for pulmonary fibrosis is essential for patient management. However, accurate diagnosis of the specific pattern is challenging due to overlapping radiographic characteristics. MATERIALS AND METHODS: We conducted a retrospective chart review utilizing two machine learning methods, classification and regression tree and Bayesian additive regression tree, to select the most important radiographic features for diagnosing the three most common fibrosis patterns and created an online diagnostic app for convenient implementation. RESULTS: Four hundred patients (median age of 67 with inter quartile range 58-73; 200 males) were included in the study. Peripheral distribution, homogeneity, lower lobe predominance and mosaic attenuation of fibrosis are the four most important features identified. Bayesian additive regression tree demonstrates better performance than classification and regression tree in diagnosis prediction and provides the predicted probability of each diagnosis with uncertainty intervals for each combination of features. CONCLUSION: The model and app built with Bayesian additive regression tree can be used as an effective tool in assisting radiologists in the diagnostic process of pulmonary fibrosis pattern recognition.

Development of a clinical prediction rule for mobility status at discharge in patients with total knee arthroplasty: Using a decision tree model.

BACKGROUND: Total knee arthroplasty (TKA) is an effective treatment to improve mobility in patients with severe knee osteoarthritis. However, some patients continue to have poor mobility after surgery. The preoperative identification of patients with poor mobility after TKA allows for better treatment selection and appropriate goal setting. The purpose of this study was to develop a clinical prediction rule (CPR) to predict mobility after TKA. METHODS: This study included patients undergoing primary TKA. Predictors of outcome included patient characteristics, physical function, and psychological factors, which were measured preoperatively. The outcome measure was the Timed Up and Go test, which was measured at discharge. Patients with a score of ≥11 s were considered having a low-level of mobility. The classification and regression tree methodology of decision tree analysis was used for developing a CPR. RESULTS: Of the 101 cases (mean age, 72.2 years; 71.3 % female), 26 (25.7 %) were classified as low-mobility. Predictors were the modified Gait Efficacy Scale, age, knee pain on the operated side, knee extension range of motion on the non-operated side, and Somatic Focus, a subscale of the Tampa Scale for Kinesiophobia (short version). The model had a sensitivity of 50.0 %, a specificity of 98.7 %, a positive predictive value of 92.9 %, a positive likelihood ratio of 37.5, and an area under the receiver operating characteristic curve of 0.853. CONCLUSION: We have developed a CPR that, with some accuracy, predicts the mobility outcomes of patients after TKA. This CPR may be useful for predicting postoperative mobility and clinical goal setting.

Risk profiles for smoke behavior in COVID-19: a classification and regression tree analysis approach.

BACKGROUND: COVID-19 pandemic emerged worldwide at the end of 2019, causing a severe global public health threat, and smoking is closely related to COVID-19. Previous studies have reported changes in smoking behavior and influencing factors during the COVID-19 period, but none of them explored the main influencing factor and high-risk populations for smoking behavior during this period. METHODS: We conducted a nationwide survey and obtained 21,916 valid data. Logistic regression was used to examine the relationships between each potential influencing factor (sociodemographic characteristics, perceived social support, depression, anxiety, and self-efficacy) and smoking outcomes. Then, variables related to smoking behavior were included based on the results of the multiple logistic regression, and the classification and regression tree (CART) method was used to determine the high-risk population for increased smoking behavior during COVID-19 and the most profound influencing factors on smoking increase. Finally, we used accuracy to evaluated the performance of the tree. RESULTS: The strongest predictor of smoking behavior during the COVID-19 period is acceptance degree of passive smoking. The subgroup with a high acceptation degree of passive smoking, have no smokers smoked around, and a length of smoking of ≥ 30 years is identified as the highest smoking risk (34%). The accuracy of classification and regression tree is 87%. CONCLUSION: The main influencing factor is acceptance degree of passive smoking. More knowledge about the harm of secondhand smoke should be promoted. For high-risk population who smoke, the "mask protection" effect during the COVID-19 pandemic should be fully utilized to encourage smoking cessation.

_Combining return to sport, psychological readiness, body mass, hamstring strength symmetry, and hamstring/quadriceps ratio increases the risk of a second anterior cruciate ligament injury.

PURPOSE: To investigate the combinations of variables that comprise the biopsychosocial model domains to identify clinical profiles of risk and protection of second anterior cruciate ligament injury. METHODS: One hundred and forty-five patients for return-to-sport testing after anterior cruciate ligament (ACL) reconstruction (ACLR) were contacted, and 97 were deemed eligible. All were evaluated between 6 and 24 months and followed up for 2 years. Participants answered the International Knee Documentation Committee (IKDC) and Anterior Cruciate Ligament-Return to Sport after Injury Scale (ACL-RSI), performed the postural stability assessment using the Biodex Balance System, and assessed muscle strength at 60° and 300°/s on the isokinetic dynamometer. Personal factors (age, gender, body mass index), body structures (graft type and concomitant injuries), and environmental factors (time between surgery and evaluation) were also collected. The participants were asked about the occurrence of a second ACL injury and return to sport after 2 years of follow-up. Classification and regression tree (CART) analysis was used to determine predictors of a second ACL injury. The receiver operating characteristic (ROC) curve was performed to verify the accuracy of the CART analysis, in addition to the sensitivity, specificity, and relative risk (RR) of the model. RESULTS: Of the initial 97 participants, 88 (89.8%) responded to follow-up and 14 (15.9%) had a second ACL injury (11 graft ruptures and three contralateral ACL). CART analysis identified the following variables as predictors of second ACL injury: return to sport, hamstring strength symmetry at 300°/s, ACL-RSI score, hamstrings/quadriceps ratio at 60°/s, and body mass index (BMI). CART correctly identified 9 (64.3%) of the 14 participants who were reinjured and 71 (95.9%) of the 74 participants who were not. The total correct classification was 90.9%. The area under the ROC curve was 0.88 (95% CI 0.72-0.99; p < 0.001), and the model showed a sensitivity of 75% (95% CI 42.8-94.5), specificity of 93.4% (95% CI 85.3-97.8), and RR of 15.9 (95% CI 4.9-51.4; p < 0.0001). CONCLUSION: The combination of hamstring strength symmetry, hamstring/quadriceps ratio (body functions); return to sport (activity and participation); psychological readiness; and BMI (personal factors) could identify three clinical risk profiles for a second ACL injury with good accuracy. LEVEL OF EVIDENCE: IV.

Quantifying conditioned place preference: a review of current analyses and a proposal for a novel approach.

Conditioned place preference (CPP) is used to measure the conditioned rewarding effects of a stimulus, including food, drugs, and social interaction. Because various analytic approaches can be used to quantify CPP, this can make direct comparisons across studies difficult. Common methods for analyzing CPP involve comparing the time spent in the CS+ compartment (e.g., compartment paired with drug) at posttest to the time spent in the CS+ compartment at pretest or to the CS- compartment (e.g., compartment paired with saline) at posttest. Researchers can analyze the time spent in the compartment(s), or they can calculate a difference score [(CS+post - CS+pre) or (CS+post - CS-post)] or a preference ratio (e.g., CS+post/(CS+post + CS-post)). While each analysis yields results that are, overall, highly correlated, there are situations in which different analyses can lead to discrepant interpretations. The current paper discusses some of the limitations associated with current analytic approaches and proposes a novel method for quantifying CPP, the adjusted CPP score, which can help resolve the limitations associated with current approaches. The adjusted CPP score is applied to both hypothetical and previously published data. Another major topic covered in this paper is methodologies for determining if individual subjects have met criteria for CPP. The paper concludes by highlighting ways in which researchers can increase transparency and replicability in CPP studies.

RR Interval-based Atrial Fibrillation Detection using Traditional and Ensemble Machine Learning Algorithms.

Atrial fibrillation (AF) is a life threatening disease and can cause stroke, heart failure, and sometimes death. To reduce the rate of mortality and morbidity due to increased prevalence of AF, early detection of the same becomes a prior concern. Traditional machine learning (TML) algorithms and ensemble machine learning (EML) algorithms are proposed to detect AF in this article. The performances of both these methods are compared in this study. Methodology involves computation of RR interval features extracted from electrocardiogram and its classification into: normal, AF, and other rhythms. TML techniques such as Classification and Regression Tree, K Nearest Neighbor, C4.5, Iterative Dichotomiser 3, Support Vector Machine and EML classifier such as Random Forest (RF), and Rotation Forest are used for classification. The proposed method is evaluated using PhysioNet challenge 2017. During the tenfold cross validation, it is observed that RF classifier provided good classification accuracy of 99.10% with area under the curve of 0.998. Apart from contributing a new methodology, the proposed study also experimentally proves higher performance with ensemble learning method, RF. The methodology has many applications in health care management systems including defibrillators, cardiac pacemakers, etc.

The ReadFree tool for the identification of poor readers: a validation study based on a machine learning approach in monolingual and minority-language children.

In this study, we validated the "ReadFree tool", a computerised battery of 12 visual and auditory tasks developed to identify poor readers also in minority-language children (MLC). We tested the task-specific discriminant power on 142 Italian-monolingual participants (8-13 years old) divided into monolingual poor readers (N = 37) and good readers (N = 105) according to standardised Italian reading tests. The performances at the discriminant tasks of the "ReadFree tool" were entered into a classification and regression tree (CART) model to identify monolingual poor and good readers. The set of classification rules extracted from the CART model were applied to the MLC's performance and the ensuing classification was compared to the one based on standardised Italian reading tests. According to the CART model, auditory go-no/go (regular), RAN and Entrainment100bpm were the most discriminant tasks. When compared with the clinical classification, the CART model accuracy was 86% for the monolinguals and 76% for the MLC. Executive functions and timing skills turned out to have a relevant role in reading. Results of the CART model on MLC support the idea that ad hoc standardised tasks that go beyond reading are needed.

Machine learning decision tree models for multiclass classification of common malignant brain tumors using perfusion and spectroscopy MRI data.

Background: To investigate the contribution of machine learning decision tree models applied to perfusion and spectroscopy MRI for multiclass classification of lymphomas, glioblastomas, and metastases, and then to bring out the underlying key pathophysiological processes involved in the hierarchization of the decision-making algorithms of the models. Methods: From 2013 to 2020, 180 consecutive patients with histopathologically proved lymphomas (n = 77), glioblastomas (n = 45), and metastases (n = 58) were included in machine learning analysis after undergoing MRI. The perfusion parameters (rCBVmax, PSRmax) and spectroscopic concentration ratios (lac/Cr, Cho/NAA, Cho/Cr, and lip/Cr) were applied to construct Classification and Regression Tree (CART) models for multiclass classification of these brain tumors. A 5-fold random cross validation was performed on the dataset. Results: The decision tree model thus constructed successfully classified all 3 tumor types with a performance (AUC) of 0.98 for PCNSLs, 0.98 for GBM and 1.00 for METs. The model accuracy was 0.96 with a RSquare of 0.887. Five rules of classifier combinations were extracted with a predicted probability from 0.907 to 0.989 for that end nodes of the decision tree for tumor multiclass classification. In hierarchical order of importance, the root node (Cho/NAA) in the decision tree algorithm was primarily based on the proliferative, infiltrative, and neuronal destructive characteristics of the tumor, the internal node (PSRmax), on tumor tissue capillary permeability characteristics, and the end node (Lac/Cr or Cho/Cr), on tumor energy glycolytic (Warburg effect), or on membrane lipid tumor metabolism. Conclusion: Our study shows potential implementation of machine learning decision tree model algorithms based on a hierarchical, convenient, and personalized use of perfusion and spectroscopy MRI data for multiclass classification of these brain tumors.

An evaluation of microcrystalline cellulose attributes affecting compaction-induced pellet coat damage through a multi-faceted analysis.

Pellet coat damage in multi-unit pellet system (MUPS) tablets has previously been studied and addressed with limited success. The effects of lactose filler material attributes on pellet coat damage have been relatively well-studied but a similar understanding of microcrystalline cellulose (MCC) is lacking notwithstanding its high cushioning potential. Hence, the relationships between MCC attributes and pellet coat damage were investigated. Single pellet in minitablets (SPIMs) were used to isolate pellet-filler effects and reveal the under-unexplored impact of risk factors found in MUPS tablets. MUPS tablets and SPIMs were prepared with various grades of MCC and pellets with an ethylcellulose or acrylic coat at various compaction pressures. Subsequently, the extent of pellet coat damage was determined by dissolution test and quantified using two indicators to differentiate the nature of the damage. A multi-faceted analytical approach incorporated linear regression, correlations and a classification and regression tree algorithm and evaluated how MCC attributes, such as flowability, particle size and plastic deformability, exert various influences on the extent of ethylcellulose and acrylic pellet coat damage. This analysis improved the understanding of the different mechanisms by which pellet coat damage to these two polymer types occurs which can help enhance future pellet coat damage mitigation strategies.

Adding automated decision-tree models to multiparametric MRI for parotid tumours improves clinical performance.

PURPOSE: Therapeutic management of parotid gland tumours depends on their histological type. To aid its characterisation, we sought to develop automated decision-tree models based on multiparametric magnetic resonance imaging (MRI) parameters and to evaluate their added diagnostic value compared with morphological sequences. METHODS: 206 MRIs from 206 patients with histologically proven parotid gland tumours were included from January 2009 to January 2018. Multiparametric MRI findings (including parameters derived from diffusion-weighted imaging [DWI] and dynamic contrast-enhanced [DCE]) were used to build predictive classification and regression tree (CART) models for each histological type. All MRIs were read twice: first, based on morphological sequence findings only, and second, with the addition of multiparametric sequences and CART findings. The diagnostic performance between these two readings was compared using ROC curves. RESULTS: Compared to morphological sequences alone, the addition of multiparametric analysis significantly increased the diagnostic performance for all histological types (p < 0.001 to p = 0.011), except for lymphomas, where the increase was not significant (AUC 1.00 vs. 0.99, p = 0.066). ADCmean was the best parameter to identify pleomorphic adenomas, carcinomas and lymphomas with respective cut-offs of 1.292 × 10-3 mm2/s, 1.181 × 10-3 mm2/s and 0.611 × 10-3 mm2/s, respectively. × 10-3 mm2/s. The mean extracellular-extravascular space coefficient was the best parameter to Warthin tumours from the others, with a cut-off of 0.07. CONCLUSIONS: The addition of decision tree prediction models based on multiparametric sequences improves the non-invasive diagnostic performance of parotid gland tumours. ADC and extracellular-extravascular space coefficient are the two best parameters for decision making.

What cultural values determine student self-efficacy? An empirical study for 42 countries and economies.

Self-efficacy is a vital personal characteristic for student success. However, the challenge of cross-cultural comparisons remains as scalar invariance is hard to be satisfied. Also, it is unclear how to contextually understand student self-efficacy in light of cultural values in different countries. This study implements a novel alignment optimization method to rank the latent means of student self-efficacy of 308,849 students in 11,574 schools across 42 countries and economies that participated in the 2018 Program in International Student Assessment. We then used classification and regression trees to classified countries with differential latent means of student self-efficacy into groups according to Hofstede's six cultural dimensions theory. The results of the alignment method recovered that Albania, Colombia, and Peru had students with the highest mean self-efficacy, while Slovak Republic, Moscow Region (RUS), and Lebanon had the lowest. Moreover, the CART analysis indicated a low student self-efficacy for countries presenting three features: (1) extremely high power distance; (2) restraint; and (3) collectivism. These findings theoretically highlighted the significance of cultural values in shaping student self-efficacy across countries and practically provided concrete suggestions to educators on which countries to emulate such that student self-efficacy could be promoted and informed educators in secondary education institutes on the international expansion of academic exchanges.