Classification And Regression Tree Algorithm Research Articles

Prediction of response to promising first-line chemotherapy in ovarian cancer patients with residual peritoneal tumors: practical biomarkers and robust multiplex models.

Platinum/taxane (TC) chemotherapy with debulking surgery stays the mainstay of the treatment in ovarian cancer patients with peritoneal metastasis, and recently its novel modality, intraperitoneal carboplatin with dose-dense paclitaxel (ddTCip), was shown to have greater therapeutic impact. Nevertheless, the response varies among patients and consequent recurrence, or relapse often occurs. Discovery of therapeutic response predictor to ddTCip and/or TC therapy is eagerly awaited to improve the treatment outcome. Using datasets in 76 participants in our ddTCip study and published databases on patients received TC therapy, we first validated a total of 75 previously suggested markers, sought out more active biomarkers through the association analyses of genome-wide transcriptome and genotyping data with progression-free survival (PFS) and adverse events, and then developed multiplex statistical prediction models for PFS and toxicity by mainly using multiple regression analysis and the classification and regression tree (CART) algorithm. The association analyses revealed that SPINK1 could be a possible biomarker of ddTCip efficacy, while ABCB1 rs1045642 and ERCC1 rs11615 would be a predictor of hematologic toxicity and peripheral neuropathy, respectively. Multiple regression analyses and CART algorithm finally provided a potent efficacy prediction model using 5 gene expression data and robust multiplex toxicity prediction models-CART models using a total of 4 genotype combinations and multiple regression models using 15 polymorphisms on 12 genes. Biomarkers and multiplex models composed here could work well in the response prediction of ddTCip and/or TC therapy, which might contribute to realize optimal selection of the key therapy.

Building semi-supervised decision trees with semi-cart algorithm

AbstractDecision trees are a fundamental statistical learning tool for addressing classification and regression problems through a recursive partitioning approach that effectively accommodates numerical and categorical data [1, 2]. The Classification and regression tree (CART) algorithm underlies modern Boosting methodologies such as Gradient boosting machine (GBM), Extreme gradient boosting (XGBoost), and Light gradient boosting machine (LightGBM). However, the standard CART algorithm may require improvement due to its inability to learn from unlabeled data. This study proposes several modifications to incorporate test data into the training phase. Specifically, we introduce a method based on Graph-based semi-supervised learning called “Distance-based Weighting,” which calculates and removes irrelevant records from the training set to accelerate the training process and improve performance. We present Semi-supervised classification and regression tree (Semi-Cart), a new implementation of CART that constructs a decision tree using weighted training data. We evaluated its performance on thirteen datasets from various domains. Our results demonstrate that Semi-Cart outperforms standard CART methods and contributes to statistical learning.

Potential and performance for classifying Earth surface only with ICESat-2 altimetric data

The huge volume of data from the Ice, Cloud and land Elevation Satellite-2 (ICESat-2), designed for mapping polar ice, sea ice, and continental vegetation, requires a highly automated data analysis and reliable terrain classification. In particular, we have developed a method to identify 4 distinct terrain categories in observed terrain, namely ocean, land, sea ice, and ice sheets. This study performed the following efforts: first, the spatial distribution characteristics for each of the 4 categories within individual ICESat-2 “major frames” along the orbit were extracted; second, these features were fed into Classification and Regression Tree (CART) and Random Forest (RF) for training; and lastly, post-processing enhancement was used to improve the classification results. Based on the 76,891 major frame samples (10,764,740 m along track) acquired via various ICESat-2 datasets, the accuracy of the two model were calculated using ten-fold cross-validation. The results indicate that the RF algorithm obtained higher classification accuracy (average accuracy [AA] = 0.9353, overall accuracy [OA] = 0.9342, and Cohen’s Kappa coefficient [kappa] = 0.9122) when compared with the CART algorithm (AA = 0.9066, OA = 0.9057, and kappa = 0.8743). Overall, our approach can effectively reduce the workload of human field investigation or visual inspection of altimetry data, improve the accuracy for Earth surface classification, and add to the variety of ways to obtain global surface information from ICESat-2 data.

A CHEAPLY NON-DESTRUCTIVE TECHNIQUE TO ESTIMATE HONEY QUALITY: THERMAL IMAGING AND MACHINE LEARNING

The aim of this study was to estimate honey quality based on proline and Brix content using a thermal imaging and machine learning algorithm. The proline, Brix and color properties of twenty honey samples were determined. Proline and Brix values were classified and estimated using the classification and regression tree (CART) algorithm. The mean proline and Brix content in honey samples was 678.83±192.16 mg/kg and 83.2±0.79%, respectively. CART analysis revealed that high proline honey samples had L values above 48.143 and b* values below 35.416. In contrast, honey samples with low Brix values were characterized by L and a* values below 55.860 and 53.660, respectively, and were identified as freshly harvested. The CART algorithm classified the proline and Brix values with an accuracy of 95% and 100%, respectively (p< 0.001). As a result, whitish, bluish, blackish and greenish honeys are of higher quality due to high proline and low Brix content. However, to accurately assess honey quality based on its color traits, comprehensive studies with more honey samples and origin, are required.

Object Based Classification in Google Earth Engine Combining SNIC and Machine Learning Methods (Case Study: Lake Köyceğiz)

Köyceğiz Lake is one of our country’s most critical coastal barrier lakes, rich in sulfur, located at the western end of the Mediterranean Region. Köyceğiz Lake, connected to the Mediterranean via the Dalyan Strait, is one of the 7 lakes in the world with this feature. In this study, water change analysis of Köyceğiz Lake was carried out by integrating the Object-Based Image Classification method with CART (Classification and Regression Tree), RF (Random Forest), and SVM (Support Vector Machine) algorithms, which are machine learning algorithms. SNIC (Simple Non-iterative Clustering) segmentation method was used, which allows a detailed analysis at the object level by dividing the image into super pixels. Sentinel 2 Harmonized images of the study area were obtained from the Google Earth Engine (GEE) platform for 2019, 2020, 2021, and 2022,and all calculations were made in GEE. When the classification accuracies of four years were examined, it was seen that the classification accuracies(OA, UA, PA, and Kappa) of the lake water area were above 92%, F-score was above 0.98 for all methods using the object-based classification method obtained by the combination of the SNIC algorithm and CART, RF, and SVM machine learning algorithms. It has been determined that the SVM algorithm has higher evaluation metrics in determining the lake water area than the CART and RF methods.

Land-use classification of Malaysian soils by ultra-high performance liquid chromatography (UHPLC)-based untargeted data combined with chemometrics for forensic provenance

Soil has become the most precious trace evidence attributed to its transference and persistence, but little is known about the capability of non-volatile organic fractions of soil in predicting land-use class. Hence, this study devoted to discriminating soils into four land-use classes based on chromatographic data and predictive modelling. Nine sites representing four different land-use classes were selected for collecting 73 soil samples via a grid method. Then, all the soil samples were dried, homogenized, sieved and eventually extracted using acetonitrile. The extracts were further analyzed via an isocratic elution in an ultra-high performance liquid chromatography (UHPLC) system. The pixel-level chromatograms were carefully assessed via diverse data preprocessing (DP) methods including baseline correction, normalization and derivative algorithms. Furthermore, the retention time (RT) window was also segmented into several sub-windows and thoroughly evaluated. Prediction accuracy of classification and regression tree (CART) modelling as estimated using 100 subsets of training samples on the corresponding testing and blind samples were ranked using TOPSIS method for identifying the most desired sub-window and DP strategy. The sub-RT window covering 0–5 min preprocessed via modified polynomial fitting algorithm followed by vector normalization emerged to be the most desired sub-dataset. The best sub-dataset modelled using CART and partial least squares-discriminant analysis (PLS2-DA) algorithms achieved 79.5 % and 97.4 % accuracy in predicting the land-use classes of 13 mixed samples. In conclusion, UHPLC-based fingerprint technique coupled with predictive modelling shows great potential in inferring land-use class of trace amount of soil.

Crop Field Boundary Detection and Classification using Machine Learning

Crop classification and detection of crop field boundaries empower farmers which helps the agricultural businesses to estimate crop field dimensions and yields accurately. Our research focuses on estimating crucial agricultural inputs such as seeds, pesticides, insecticides, and fertilizers to enhance overall production. The conventional method of manually identifying field boundaries is both time-consuming with labour-intensive. In contrast, our study harnesses data from diverse satellites such as Sentinel, Landsat, and MODIS, encompassing valuable land usage information. By integrating this data with machine learning algorithms, we achieve real-time monitoring of crop fields through effective classification and boundary identification. For the classification of crop fields within our study area, we recommend employing the Classification and Regression Tree (CART) algorithm. Additionally, we leverage normalized difference indices, such as the Normalized Difference Vegetation Index (NDVI) and the Normalized Difference Water Index (NDWI), as features for classification. We compare these features with Support Vector Machine (SVM) and Random Forest (RF) algorithms. Subsequently, we utilize the Canny edge detection technique to identify boundaries within the classified crop areas. Notably, our approach utilizes the Google Earth Engine (GEE) as a primary platform for extracting features, conducting data training, and visualizing information. The proposed algorithm yields impressive results with a high level of accuracy. Notably, the CART algorithm achieves a remarkable accuracy rate of 96.1%. Furthermore, we incorporate NDWI-based Canny edge detection into our methodology. The outcomes convincingly underscore the practicality and applicability of our research in real-world scenarios.

Unraveling the Effect of Compositional Ratios on the Kesterite Thin-Film Solar Cells Using Machine Learning Techniques

In the Kesterite family, the Cu2ZnSn(S,Se)4 (CZTSSe) thin-film solar cells (TFSCs) have demonstrated the highest device efficiency with non-stoichiometric cation composition ratios. These composition ratios have a strong influence on the structural, optical, and electrical properties of the CZTSSe absorber layer. So, in this work, a machine learning (ML) approach is employed to evaluate effect composition ratio on the device parameters of CZTSSe TFSCs. In particular, the bi-metallic ratios like Cu/Sn, Zn/Sn, Cu/Zn, and overall Cu/(Zn+Sn) cation composition ratio are investigated. To achieve this, different machine learning algorithms, such as decision trees (DTs) and classification and regression trees (CARTs), are used. In addition, the output performance parameters of CZTSSe TFSCs are predicted by both continuous and categorical approaches. Artificial neural networks (ANN) and XGBoost (XGB) algorithms are employed for the continuous approach. On the other hand, support vector machine and k-nearest neighbor’s algorithms are also used for the categorical approach. Through the analysis, it is observed that the DT and CART algorithms provided a critical composition range well suited for the fabrication of highly efficient CZTSSe TFSCs, while the XGB and ANN showed better prediction accuracy among the tested algorithms. The present work offers valuable guidance towards the integration of the ML approach with experimental studies in the field of TFSCs.

Does decision tree analysis predict oral ulcer activity-related factors in patients with Behçet's syndrome?

The study aimed to identify the interactions among treatment protocols and oral ulcer activity related factors in patients with Behçet's syndrome (BS) using the Classification and Regression Tree (CART) algorithm. In this cross-sectional study, 979 patients with BS were included from16 centres in Turkey, Jordan, Brazil and the United Kingdom. In the CART algorithm, activities of oral ulcer (active vs. inactive), genital ulcer (active vs. inactive), cutaneous involvement (active vs. inactive), musculoskeletal involvement (active vs. inactive), gender (male vs. female), disease severity (mucocutaneous and musculoskeletal involvement vs. major organ involvement), smoking habits (current smoker vs. non-smoker), tooth brushing habits (irregular vs. regular), were input variables. The treatment protocols regarding immunosuppressive (IS) or non-IS medications were the target variable used to split from parent nodes to purer child nodes in the study. In mucocutaneous and musculoskeletal involvement (n=538), the ratio of IS use was higher in patients with irregular toothbrushing (ITB) habits (27.1%) than in patients with regular toothbrushing (RTB) habits (14.2%) in oral ulcer activity. In major organ involvement (n=441), male patients with ITB habits were more likely treated with IS medications compared to those with RTB habits (91.6% vs. 77.6%, respectively). Male BS patients on IS who have major organ involvement and oral ulcer activity with mucocutaneous and musculoskeletal involvement have irregular toothbrushing habits. Improved oral hygiene practices should be considered to be an integral part for implementing patient empowerment strategies for BS.

Predicting body weight of South African Sussex cattle at weaning using multivariate adaptive regression splines and classification and regression tree data mining algorithms

ABSTRACT The use of multivariate adaptive regression splines (MARS) and classification and regression tree (CART) to estimate the live body weight at weaning age of the Sussex cattle breed remain poorly understood in South Africa. This study was conducted to examine the effect of linear body measurements on body weight at weaning using MARS and CART algorithms. The body weight and linear body measurements included sternum height, withers height, heart girth, hip height, body length, rump length and rump width were collected from 101 Sussex cattle (female = 57 and male = 44) at weaning. Goodness of fit criterions was used to select the best data mining algorithms. The results showed that MARS showed higher predictive performance in the criteria as compared to CART algorithm. The findings of the study suggest that MARS algorithm can be used to estimate the BW at weaning age in Sussex cattle breed. These findings might be helpful to cattle farmers in the selection criterions of breeding stock at weaning age.

Decision Tree Models for Predicting the Effect of Electronic Waste on Human Health

Informal processing of electronic waste has become one of the commonest sources of employment in developing countries which has contracted a great impact on human health due to the improper disposal of the heavy metals found in these waste materials. Several research works have been conducted to predict e-waste generation and management. Unfortunately, there is no study to predict the disease associated with the activities of informal e-waste products and their disposal. This study predicts the categorized disease of a person working and/or living at an electronic waste dump site based on their activities and their lifestyle using decision tree algorithms. The categorized diseases are skin, respiratory and reproductive diseases. The work compared the performance of C4.5 algorithm which used the Chi-squared test for tree pruning to handle overfitting with the Classification and Regression Tree (CART) algorithm which used tree depth control to handle overfitting. The C4.5 algorithm proved to be more effective than the CART algorithm. The study recommends that whenever two or more algorithms can be used to handle the same problem in principle, they should all be used and their results be compared.

Penerapan Metode CART Dalam Klasifikasi Jurusan Siswa Baru

SMK Negeri 3 Pontianak is one of the vocational education schools in Pontianak City. Every new academic year SMK Negeri 3 Pontianak accepts around 320 new students. The large number of prospective new students makes the majoring process carried out by the school become less effective and takes a long time to determine majors for new students. With a system that can classify new student majors, it helps in the process of determining student majors. This study uses the Classification and Regression Trees (CART) algorithm for the classification process in determining majors for new students. The assessment indicators used for classification consist of interest, MTK (US) school exam scores, school exam IPA scores, school exam Indonesian language scores, math report cards, science report cards, social science report cards, Indonesian language report cards, and English report cards. Classification of majors at SMK 3 Pontianak consists of accounting, office, marketing, and hospitality majors. The amount of data used is 320 data which is divided into 224 training data and 96 test data. The CART algorithm generates decision trees, rules, and new student majors that have been classified. Based on the test results using the confusion matrix, the system accuracy results are 84.38%.

Performance Analysis of Classification and Regression Tree (CART) Algorithm in Classifying Male Fertility Levels with Mobile-Based

Fertility is the ability to produce offspring in a man or the ability of the reproductive organs to work optimally in fertilization. Fertility rates have declined drastically in the last fifty years. Machine Learning is a field devoted to understanding and building learning methods. This study will use machine learning algorithms to classify male fertility levels, namely the Classification and Regression Tree (CART) algorithm and the K-Fold Cross Validation validation method. The fertility dataset used in this study was obtained from the UCI Machine Learning website, with a total of 100 data and the variables used are Age, Childish diseases, Accident or serious trauma, Surgical intervention, High fevers in the last year, Frequency of alcohol consumption, Smoking habit, Number of hours spent sitting per day and Diagnosis. K-Fold Cross Validation can be used together with CART to measure the performance of the CART model on different data, so as to avoid overfitting or underfitting the CART model. Based on the calculation of the CART algorithm and the K-Fold Cross Validation validation method (K = 1 to K = 9), the average accuracy value for training data is 98.70% and the average accuracy value for testing data is 81.16%. The results of this study have proven that the CART algorithm can be used to classify the level of fertility in men well. In addition, the classification model formed can be implemented into a mobile application (android) so that it is easy to use and understand.

Comparison of Error Rate Prediction Methods in Classification Modeling with Classification and Regression Tree (CART) Methods for Balanced Data

CART (Classification and Regression Tree) is one of the classification algorithms in the decision tree method. The model formed in CART is a tree consisting of root nodes, internal nodes, and terminal nodes. After the model is formed, it is necessary to calculate the accuracy of the model. The aims is to see the performance of the model. The accuracy of this model can be done by calculating the predicted error rate in the model. The error rate prediction method works by dividing the data into training data and testing data. There are three methods in the error rate prediction method, such as Leave One Out Cross Validation (LOOCV), Hold Out (HO), and K-Fold Cross Validation. These methods have different performance in dividing data into training data and testing data, so there are advantages and disadvantages to each method. Therefore, a comparison was made for the three error rate prediction methods with the aim of determining the appropriate method for the CART algorithm. This comparison was made by considering several factors, for instance variations in the mean, number of variables, and correlations in normal distributed random data. The results of the comparison will be observed using a boxplot by looking at the median error rate and the lowest variance. The results of this study indicate that the K-Fold Cross Validation has the median error rate and the lowest variance, so the most suitable error prediction method used for the CART method is the K-Fold Cross Validation method.

The Comparison of Ketokonazol and Hypochlorous Acid (Hocl) Applications for the Treatment of The Fungal Infections (Dermatophytosis)

Dermatophytosis is a mycotic disease of the skin that is resistant to treatment. The aim of this study was to investigate the treatment efficacy of a novel antimicrobial agent, Hypochlorous Acid (HOCL), on dermatophytosis of cats and dogs, in comparison with Ketaconazole.
 In this study, a total of 76 animals (26 cats and 50 dogs) without any disease other than skin fungal infection after clinical, hematological, biochemical, microscopic and Wood's lamp examinations were used. Subjects were randomly assigned to two equal treatment groups within their own species as HOCL: HP and Ketaconazole: KT. Naked eye inspection results were collected on the 8th, 11th and 15th days of all patients. The study was terminated on the 15th day by collecting the Wood's lamp and microscopic examination data together with the last inspection finding. Findings were analyzed statistically with chi-square and CART (Classification and Regression Tree) algorithm test.
 Inspection results of the treatment groups exhibited significant recovery over time (day 8, 11 and 15) for both species (p0.05). According to the microscopic examination results, a significant statistical difference was observed between the HP and KT groups (p0.05).
 As a result, it was concluded that HOCl has an effect on dermatophytosis of cat and dogs, although not as much as Ketaconazole, but further studies are needed to reveal the results more clearly.

Unraveling the importance of fabrication parameters of copper oxide-based resistive switching memory devices by machine learning techniques

In the present study, various statistical and machine learning (ML) techniques were used to understand how device fabrication parameters affect the performance of copper oxide-based resistive switching (RS) devices. In the present case, the data was collected from copper oxide RS devices-based research articles, published between 2008 to 2022. Initially, different patterns present in the data were analyzed by statistical techniques. Then, the classification and regression tree algorithm (CART) and decision tree (DT) ML algorithms were implemented to get the device fabrication guidelines for the continuous and categorical features of copper oxide-based RS devices, respectively. In the next step, the random forest algorithm was found to be suitable for the prediction of continuous-type features as compared to a linear model and artificial neural network (ANN). Moreover, the DT algorithm predicts the performance of categorical-type features very well. The feature importance score was calculated for each continuous and categorical feature by the gradient boosting (GB) algorithm. Finally, the suggested ML guidelines were employed to fabricate the copper oxide-based RS device and demonstrated its non-volatile memory properties. The results of ML algorithms and experimental devices are in good agreement with each other, suggesting the importance of ML techniques for understanding and optimizing memory devices.

Do Innovation Metrics Reflect Sustainable Policy Making in Europe? A Comparative Study Case on the Carpathian and Alpine Mountain Regions

This paper questions the evaluation of innovation systems and innovation measurements and the effectiveness of innovation policies applied at the territorial level by assessing whether the existing European regional scoreboard is effective in providing accurate inputs for decision-makers in mountainous regions. The aim of the research is to provide, through comparative analysis by using statistical multi-methods of two mountainous macro-regions (the Alps and the Carpathians), a possible and available path to develop novel perspectives and alternative views on innovation systems’ performance for informed and territorial-based policy making by using the indicators of the Regional Innovation Scoreboard. The methodology used includes descriptive statistics, chi-square bivariate test, Student’s t test, one-way ANOVA with Bonferroni post hoc multiple comparisons, multilinear regression analysis, and decision tree with CRT (classification and regression trees) algorithm. Our results emphasize the similarities and differences between the Alpine and Carpathian mountain regions, find the best predictors for each mountain region, and provide a scientific basis for the development of a holistic approach linking measurement theory, innovation systems, innovation policies, and their territorial approach toward sustainable development of mountain areas. The paper’s contribution is relevant in the context of remote, rural, and mountain areas, which are usually left behind in terms of innovation chances and in the context of the COVID-19 aftermath with budget constraints. The present results are pertinent for designing effective smart specialization strategies in these regions due to the difficulties that most remote areas and less developed regions are facing in developing innovation policies.

Skin cancer detection using effective optical parameters and the classification and regression tree algorithm: A novel framework

Early detection of skin cancer matters because diagnosis, prognosis and treatment plan differ for each skin cancer type at their stages. Medical imaging taking the advantages of the non-invasive and non-ionizing polarized light is emerging as a tool for the development of screening and diagnotic tests. In this work, we proposed a novel framework to classify human melanoma and nonmelanoma skin cancer using the Classification and Regression Tree algorithm (CART). The samples were prepared from twenty-four non-melanoma skin cancer samples (consisting of twelve squamous cell carcinoma and twelve basal cell carcinoma samples); and three melanoma skin cancer samples. We calculated ten optical parameters from anisotropic biological tissues, namely the LB orientation angle (α), the LB phase retardance (β), the CB optical rotation angle (γ), the LD orientation angle (θd), the linear dichroism (D), the circular dichroism (R), the degrees of linear depolarization (e1 and e2), the degree of circular depolarization (e3), and the depolarization index (∆) using Stokes-Mueller matrix formalism. All effective optical parameters of biological tissue were then input into the CART classifier as predictors. The model yielded an accuracy of 92.6%, which is desirable for any robust and interpretable classification model. The results showed that for biological tissue samples, linear polarization properties dominate over circular ones due to the cellular microstructural composition of tissue, especially under anomalous growth as seen in skin cancer. This novel framework can potentially assist physicians in making timely and well-informed medical decisions.

A new classification and regression tree algorithm: Improved diagnostic sensitivity for HCC ≤ 3.0 cm using gadoxetate disodium-enhanced MRI

PurposeTo develop and validate an effective algorithm, based on classification and regression tree (CART) analysis and LI-RADS features, for diagnosing HCC ≤ 3.0 cm with gadoxetate disodium‑enhanced MRI (Gd-EOB-MRI). MethodWe retrospectively included 299 and 90 high-risk patients with hepatic lesions ≤ 3.0 cm that underwent Gd-EOB-MRI from January 2018 to February 2021 in institution 1 (development cohort) and institution 2 (validation cohort), respectively. Through binary and multivariate regression analyses of LI-RADS features in the development cohort, we developed an algorithm using CART analysis, which comprised the targeted appearance and independently significant imaging features. On per-lesion basis, we compared the diagnostic performances of our algorithm, two previously reported CART algorithms, and LI-RADS LR-5 in development and validation cohorts. ResultsOur CART algorithm, presenting as a decision tree, included targetoid appearance, HBP hypointensity, nonrim arterial phase hyperenhancement (APHE), and transitional phase hypointensity plus mild-moderate T2 hyperintensity. For definite HCC diagnosis, the overall sensitivity of our algorithm (development cohort 93.2%, validation cohort 92.5%; P < 0.006) was significantly higher than those of Jiang’s algorithm modified LR-5 (defined as targetoid appearance, nonperipheral washout, restricted diffusion, and nonrim APHE) and LI-RADS LR-5, with the comparable specificity (development cohort: 84.3%, validation cohort: 86.7%; P ≥ 0.006). Our algorithm, providing the highest balanced accuracy (development cohort: 91.2%, validation cohort: 91.6%), outperformed other criteria for identifying HCCs from non-HCC lesions. ConclusionsIn high-risk patients, our CART algorithm developed with LI-RADS features showed promise for the early diagnosis of HCC ≤ 3.0 cm with Gd-EOB-MRI.

A Simplified Clinical Prediction Rule for Prognosticating Dependent Daily Living in Patients with Thoracic Spinal Cord Injury: A Multicenter Nationwide Japan Registry Study

Introduction: Clinical prediction rule (CPR) using decision tree analysis is able to show the branching of the variables under consideration in a clear, hierarchical manner, including specific reference values, which can be used as classifiers in clinical practice. However, CPRs developed by decision tree analysis for predicting the degree of independent living of patients with thoracic spinal cord injury (SCI) are few. The purpose of this study was to develop a simplified CPR for prognosticating dependent daily living in patients with thoracic SCI. Methods: We extracted data on patients with thoracic SCI from a national multicenter registry database, the Japan Rehabilitation Database (JRD). All patients with thoracic SCI who were hospitalized within 30 days after the injury onset were included. The independent living was categorized in the JRD as follows: independent socially, independent at home, needing care at home, independent at the facility, and needing care at the facility. These categories were used as the objective variables in classification and regression tree (CART) analysis. The CART algorithm was applied to develop the CPR for predicting whether patients with thoracic SCI achieve independent living at hospital discharge. Results: Three hundred ten patients with thoracic SCI were included in the CART analysis. The CART model identified, in a hierarchical order, patient’s age, residual function level, and the bathing sub-score of Functional Independence Measure as the top three factors with moderate classification accuracy and area under the curve. Conclusions: We developed a simplified, moderately accurate CPR for predicting whether patients with thoracic SCI achieve independent living at hospital discharge.