Tree-based Machine Learning Models Research Articles

Machine Learning-Based Prediction of Suicidal Thinking in Adolescents by Derivation and Validation in 3 Independent Worldwide Cohorts: Algorithm Development and Validation Study.

Suicide is the second-leading cause of death among adolescents and is associated with clusters of suicides. Despite numerous studies on this preventable cause of death, the focus has primarily been on single nations and traditional statistical methods. This study aims to develop a predictive model for adolescent suicidal thinking using multinational data sets and machine learning (ML). We used data from the Korea Youth Risk Behavior Web-based Survey with 566,875 adolescents aged between 13 and 18 years and conducted external validation using the Youth Risk Behavior Survey with 103,874 adolescents and Norway's University National General Survey with 19,574 adolescents. Several tree-based ML models were developed, and feature importance and Shapley additive explanations values were analyzed to identify risk factors for adolescent suicidal thinking. When trained on the Korea Youth Risk Behavior Web-based Survey data from South Korea with a 95% CI, the XGBoost model reported an area under the receiver operating characteristic (AUROC) curve of 90.06% (95% CI 89.97-90.16), displaying superior performance compared to other models. For external validation using the Youth Risk Behavior Survey data from the United States and the University National General Survey from Norway, the XGBoost model achieved AUROCs of 83.09% and 81.27%, respectively. Across all data sets, XGBoost consistently outperformed the other models with the highest AUROC score, and was selected as the optimal model. In terms of predictors of suicidal thinking, feelings of sadness and despair were the most influential, accounting for 57.4% of the impact, followed by stress status at 19.8%. This was followed by age (5.7%), household income (4%), academic achievement (3.4%), sex (2.1%), and others, which contributed less than 2% each. This study used ML by integrating diverse data sets from 3 countries to address adolescent suicide. The findings highlight the important role of emotional health indicators in predicting suicidal thinking among adolescents. Specifically, sadness and despair were identified as the most significant predictors, followed by stressful conditions and age. These findings emphasize the critical need for early diagnosis and prevention of mental health issues during adolescence.

Monitoring saltwater intrusion to estuaries based on UAV and satellite imagery with machine learning models

Saltwater intrusion is a natural mixture process between watershed freshwater and seawater that frequently occurs in estuaries. Station-based monitoring of saltwater intrusion is time-consuming and labor-intensive. To enable quick monitoring of saltwater intrusion, this study developed new remote sensing algorithms for water surface salinity measurement using four decision tree-based machine learning models. These models were built based on simultaneously collected in-situ salinity data from a waterway in the Pearl River Delta and Unmanned Aerial Vehicle (UAV) hyperspectral images. A 10-fold cross-validation was applied to assess the performance of the models, with XGBoost outperforming the other three models (R2=0.93, RMSE = 0.88 psu). Then the developed model was employed for Sentinel-2 multispectral satellite images to invert the estuarine salinity distribution at a larger spatial scale. Results displayed the high performance of the machine learning models proposed in this study for mapping the salinity distribution in river channels, making it an efficient and practical technique for monitoring saltwater intrusion in river channels at a regional scale.

The Performance Analysis of Male Handball Players Using Tree-Based Machine Learning Models

The Performance Analysis of Male Handball Players Using Tree-Based Machine Learning Models

Proton versus photon adjuvant radiotherapy: a multicenter comparative evaluation of recurrence following spinal chordoma resection.

Chordomas are rare tumors of the skull base and spine believed to arise from the vestiges of the embryonic notochord. These tumors are locally aggressive and frequently recur following resection and adjuvant radiotherapy. Proton therapy has been introduced as a tissue-sparing option because of the higher level of precision that proton-beam techniques offer compared with traditional photon radiotherapy. This study aimed to compare recurrence in patients with chordomas receiving proton versus photon radiotherapy following resection by applying tree-based machine learning models. The clinical records of all patients treated with resection followed by adjuvant proton or photon radiotherapy for chordoma at Mayo Clinic were reviewed. Patient demographics, type of surgery and radiotherapy, tumor recurrence, and other variables were extracted. Decision tree classifiers were trained and tested to predict long-term recurrence based on unseen data using an 80/20 split. Fifty-three patients with a mean ± SD age of 55.2 ± 13.4 years receiving surgery and adjuvant proton or photon therapy to treat chordoma were identified; most patients were male. Gross-total resection was achieved in 54.7% of cases. Proton therapy was the most common adjuvant radiotherapy (84.9%), followed by conventional or external-beam radiation therapy (9.4%) and stereotactic radiosurgery (5.7%). Patients receiving proton therapy exhibited a 40% likelihood of having recurrence, significantly lower than the 88% likelihood observed in those treated with nonproton therapy. This was confirmed on logistic regression analysis adjusted for extent of tumor resection and tumor location, which revealed that proton adjuvant radiotherapy was associated with a decreased risk of recurrence (OR 0.1, 95% CI 0.01-0.71; p = 0.047) compared with photon therapy. The decision tree algorithm predicted recurrence with an accuracy of 90% (95% CI 55.5%-99.8%), with the lowest risk of recurrence observed in patients receiving gross-total resection with adjuvant proton therapy (23%). Following resection, adjuvant proton therapy was associated with a lower risk of chordoma recurrence compared with photon therapy. The described machine learning models were able to predict tumor progression based on the extent of tumor resection and adjuvant radiotherapy modality used.

Urban flood susceptibility mapping using frequency ratio and multiple decision tree-based machine learning models

Urban flood susceptibility mapping using frequency ratio and multiple decision tree-based machine learning models

Integrative analysis with microbial modelling and machine learning uncovers potential alleviators for ulcerative colitis

ABSTRACT Ulcerative colitis (UC) is a challenging form of inflammatory bowel disease, and its etiology is intricately linked to disturbances in the gut microbiome. To identify the potential alleviators of UC, we employed an integrative analysis combining microbial community modeling with advanced machine learning techniques. Using metagenomics data sourced from the Integrated Human Microbiome Project, we constructed individualized microbiome community models for each participant. Our analysis highlighted a significant decline in both α and β-diversity of strain-level microbial populations in UC subjects compared to controls. Distinct differences were also observed in the predicted fecal metabolite profiles and strain-to-metabolite contributions between the two groups. Using tree-based machine learning models, we successfully identified specific microbial strains and their associated metabolites as potential alleviators of UC. Notably, our experimental validation using a dextran sulfate sodium-induced UC mouse model demonstrated that the administration of Parabacteroides merdae ATCC 43,184 and N-acetyl-D-mannosamine provided notable relief from colitis symptoms. In summary, our study underscores the potential of an integrative approach to identify novel therapeutic avenues for UC, paving the way for future targeted interventions.

Predicting anxiety, depression, and insomnia among Bangladeshi university students using tree-based machine learning models.

Mental health problem is a rising public health concern. People of all ages, specially Bangladeshi university students, are more affected by this burden. Thus, the objective of the study was to use tree-based machine learning (ML) models to identify major risk factors and predict anxiety, depression, and insomnia in university students. A social media-based cross-sectional survey was employed for data collection. We used Generalized Anxiety Disorder (GAD-7), Patient Health Questionnaire (PHQ-9) and Insomnia Severity Index (ISI-7) scale for measuring students' anxiety, depression and insomnia problems. The tree-based supervised decision tree (DT), random forest (RF) and robust eXtreme Gradient Boosting (XGBoost) ML algorithms were used to build the prediction models and their predictive performance was evaluated using confusion matrix and receiver operating characteristic (ROC) curves. Of the 1250 students surveyed, 64.7% were male and 35.3% were female. The students' ages ranged from 18 to 26 years old, with an average age of 22.24 years (SD = 1.30). Majority of the students (72.6%) were from rural areas and social media addicted (56.6%). Almost 83.3% of the students had moderate to severe anxiety, 84.7% had moderate to severe depression and 76.5% had moderate to severe insomnia problems. Students' social media addiction, age, academic performance, smoking status, monthly family income and morningness-eveningness are the main risk factors of anxiety, depression and insomnia. The highest predictive performance was observed from the XGBoost model for anxiety, depression and insomnia. The study findings offer valuable insights for stakeholders, families and policymakers enabling a more profound comprehension of the pressing mental health disorders. This understanding can guide the formulation of improved policy strategies, initiatives for mental health promotion, and the development of effective counseling services within university campus. Additionally, our proposed model might play a critical role in diagnosing and predicting mental health problems among Bangladeshi university students and similar settings.

Determining pre-procedure fasting alert time using procedural and scheduling data.

Before a medical procedure requiring anesthesia, patients are required to not eat or drink non-clear fluids for 6h and not drink clear fluids for 2h. Fasting durations in standard practice far exceed these minimum thresholds due to uncertainties in procedure start time. The aim of this retrospective, observational study was to compare fasting durations arising from standard practice with different approaches for calculating the timepoint at which patients are instructed to stop eating and drinking. Scheduling data for procedures performed in the cardiac catheterization laboratory of an academic hospital in Canada (January 2020 to April 2022) were used. Four approaches utilizing machine learning (ML) and simulation were used to predict procedure start times and calculate when patients should be instructed to start fasting. Median fasting duration for standard practice was 10.08h (IQR 3.5) for both food and clear fluids intake. The best performing alternative approach, using tree-based ML models to predict procedure start time, reduced median fasting from food/non-clear fluids to 7.7h (IQR 2) and clear liquids fasting to 3.7h (IQR 2.4). 97.3% met the minimum fasting duration requirements (95% CI 96.9% to 97.6%). Further studies are required to determine the effectiveness of operationalizing this approach as an automated fasting alert system.

Abstract 7352: ScanCT: A tree-based machine learning model to detect single-cell genomic features associated with clinical outcomes

Abstract Single-cell technologies represent a revolutionary approach to resolving cell-type heterogeneity, identifying cells in specialized states, and detecting rare disease-associated cells. With the cost of single-cell technology decreasing substantially, its integration into clinical studies is gaining momentum. A new computational tool is needed to accommodate different single-cell genomics and clinical data formats while accounting for unwanted confounders. The study aims to develop a tree-based machine learning model to leverage the unprecedented resolution of single-cell multi-omics data for delineating the genomic and phenotypic drivers behind diverse immunotherapy responses. The proposed model is called single-cell analysis of Clinical Tree (scanCT), inspired by the Generalized Unbiased Interaction Detection and Estimation method for unbiased gene and protein feature selection and easy interpretation. The scanCT model learns from the data to select the genomic feature that best splits the cells from distinct clinical responses for each tree node. The confounding factors will be regressors in the nodes but not be used for branch splitting, while gene and protein features of interest will split the tree but not enter the regression model in each node. scanCT is built to be free from the biased selection towards variables of a larger number of categories or values. With tree-pruning and cross-validation, scanCT overcomes the over-fitting issue and enhances model generalization, especially for clinical studies with limited patients. Particularly, scanCT naturally fits the hierarchical cell type relationship and handles marker gene and protein interaction effects efficiently. Our approach was tested on single-cell datasets from B-cell malignancy patients undergoing Chimeric Antigen Receptor (CAR)-T cell therapy. The results from the scanCT are highly interpretable. For instance, each branch is a gene-protein combination profile, and cells are naturally partitioned by clinical association. The linear regressions at each leaf node are the clinical predictions for cells following the splitting criteria. The regression intercept is an average estimation of toxicity (e.g., neurotoxicity) or efficacy after controlling for confounder (e.g., tumor burden). scanCT accommodates categorical or continuous clinical response and survival data and is robust to missing values, a frequent challenge in oncological studies. scanCT represents a significant step forward in single-cell data analysis, which merges complex genotypic and phenotypic information with clinical outcomes. The efficacy and toxicity-associated genomic signatures will inform new manufacturing strategies to optimize CAR-T cell therapy products. The model and clinical association detections are expected to go beyond the B-cell malignancy field to benefit the broader cancer research community. Citation Format: Ye Zheng, Long Nguyen, Peigen Zhou, Alexandre V. Hirayama. ScanCT: A tree-based machine learning model to detect single-cell genomic features associated with clinical outcomes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 7352.

Developing Tree-Based Machine Learning Models for Estimating the Pile Setup Parameter for Clay Soils

Piles driven into cohesive soils usually experience increases in capacity with time, known as pile setup phenomenon. Several empirical methods have been developed to estimate the setup parameter (A), such as the well-known Skov and Denver equation. Parameter A is crucial in predicting pile setup behavior. In this study, tree-based machine learning (ML) models such as random forest (RF) and gradient boosted tree (GBT) were applied for better estimation of the setup parameter. A database consisting of setup data from 12 instrumented piles tested at different times, and corresponding cone penetration test (CPT) and soil boring data were collected. The soil properties (i.e., undrained shear strength, plasticity index, over consolidation ratio, and coefficient of consolidation) and CPT data (cone tip resistance, sleeve friction) of clayey soil layers at pile locations were utilized to develop the ML models. Three types of tree-based ML model were developed for predicting the setup parameter, A, using CPT and soil boring data. A comparison was made between the developed ML models based on soil properties, ML models based on CPT data, and an artificial neural network (ANN) model proposed in a previous study using the same dataset. Furthermore, the best performing ML models were compared with two nonlinear regression models recommended in a previous study using the same dataset for estimating the setup parameter. The results of this research clearly demonstrated the superior prediction capability of the tree-based ML models, particularly the GBT model over the ANN and the two nonlinear regression models in evaluating the pile setup parameter.

Interpretable machine learning-based individual analysis of acute kidney injury in immune checkpoint inhibitor therapy.

Acute kidney injury (AKI) is a critical complication of immune checkpoint inhibitor therapy. Since the etiology of AKI in patients undergoing cancer therapy varies, clarifying underlying causes in individual cases is critical for optimal cancer treatment. Although it is essential to individually analyze immune checkpoint inhibitor-treated patients for underlying pathologies for each AKI episode, these analyses have not been realized. Herein, we aimed to individually clarify the underlying causes of AKI in immune checkpoint inhibitor-treated patients using a new clustering approach with Shapley Additive exPlanations (SHAP). We developed a gradient-boosting decision tree-based machine learning model continuously predicting AKI within 7 days, using the medical records of 616 immune checkpoint inhibitor-treated patients. The temporal changes in individual predictive reasoning in AKI prediction models represented the key features contributing to each AKI prediction and clustered AKI patients based on the features with high predictive contribution quantified in time series by SHAP. We searched for common clinical backgrounds of AKI patients in each cluster, compared with annotation by three nephrologists. One hundred and twelve patients (18.2%) had at least one AKI episode. They were clustered per the key feature, and their SHAP value patterns, and the nephrologists assessed the clusters' clinical relevance. Receiver operating characteristic analysis revealed that the area under the curve was 0.880. Patients with AKI were categorized into four clusters with significant prognostic differences (p = 0.010). The leading causes of AKI for each cluster, such as hypovolemia, drug-related, and cancer cachexia, were all clinically interpretable, which conventional approaches cannot obtain. Our results suggest that the clustering method of individual predictive reasoning in machine learning models can be applied to infer clinically critical factors for developing each episode of AKI among patients with multiple AKI risk factors, such as immune checkpoint inhibitor-treated patients.

The application of machine learning methods for prediction of heavy metal by activated carbons, biochars, and carbon nanotubes

Carbonaceous materials are commonly used as adsorbents for heavy metals. The determination of the adsorption capacity needs time and energy, and the key factors affecting the adsorption capacity have not been determined. Therefore, a new and efficient method is needed to predict the adsorption capacity and explore the decisive factors in the adsorption process. In this study, three tree-based machine learning models (i.e., random forest, gradient boosting decision tree, and extreme gradient boosting) were developed to predict the adsorption capacity of eight heavy metals (i.e., As, Cd, Cr, Cu, Hg, Ni, Pb, and Zn) on activated carbons, biochars, and carbon nanotubes using 3674 data points extracted from 151 journal articles. After a comprehensive comparison, the gradient boosting decision tree had the best performance for a combined model based on all data (R2 = 0.9707, RMSE = 0.1420). Moreover, independent models were developed for three datasets classified by the adsorbent and eight datasets classified by the heavy metals. In addition, a graphical user interface was built to predict the adsorption capacity of heavy metals. This study provides a novel strategy and convenient tool for the removal of heavy metals and can help to improve the removal efficiency of heavy metals to build a healthier world.

Efficient neural network- and tree-based machine learning models for predicting shear capacity of RC slender walls

Efficient neural network- and tree-based machine learning models for predicting shear capacity of RC slender walls

Assessing influential factors of Chinese industrial aqueous cadmium emissions based on machine learning and shapley additive explanations

As the main contributor of cadmium (Cd) pollution in China, industrial aqueous Cd emissions are influenced by both industrial and socioeconomic characteristics owing to their “environmental service” nature. It is essential to understand the relationship between these characteristics and industrial aqueous Cd emissions during industrialization. In this study, a data-driven framework was proposed to reveal these relationships. The framework comprises two key steps. i) Three state-of-the-art tree-based machine learning models, including LightGBM, Gradient Boosting Decision Tree and Random Forest, were trained to capture the relationships among variables. ii) Shapley additive explanation was utilized to decompose the contribution of the characteristics to the prediction for each sample. The trained LightGBM model demonstrated the most superior performance in predicting industrial aqueous Cd emissions in test datasets (R2=0.881±0.0431， RMSE=58.260±11.839， MAE=27.743±3.274), significantly overperforming traditional linear regression model. Our analysis revealed that during our sample period, the influence of industrial characteristics on regional industrial aqueous Cd emissions was approximately 1.796 times greater than that of socioeconomic characteristics. Shifts in the characteristics of the non-ferrous metal industry contributed to approximately 74% of the average increase in industrial aqueous Cd emissions in China approximately. Regional emissions were found to be positively affected by average size and density but negatively affected by the operation duration of local non-ferrous metal industrial firms. Further driver analysis showed that the growth trajectory of emissions in China can be split into three stages based on the main drivers and growth rate: 2000–2002, 2003–2004, and 2005–2007. The proposed framework overcomes the limitations of the previous methods in terms of application scope, potential factor considerations, and regression structure predefinition. Our analysis implies that policymakers should proactively adjust industrial policies and emissions regulations in response to market shocks and industrial shifts, to a better management of industrial aqueous Cd emissions.

Prediction of Vehicle-induced Air Pollution based on Advanced Machine Learning Models

Vehicle-induced air pollution is an important issue in the 21st century, posing detrimental effects on human health. Prediction of vehicle-emitted air pollutants and evaluation of the diverse factors that contribute to them are of the utmost importance. This study employed advanced tree-based machine learning models to predict vehicle-induced air pollutant levels, with a particular focus on fine particulate matter (PM2.5). In addition to a benchmark statistical model, the models employed were Gradient Boosting (GB), Light Gradient Boosting Machine (LGBM), Extreme Gradient Boosting (XGBoost), Extra Tree (ET), and Random Forest (RF). Regarding the evaluation of PM2.5 predictions, the ET model outperformed the others, as shown by MAE of 1.69, MSE of 5.91, RMSE of 2.43, and R2 of 0.71. Afterward, the optimal ET models were interpreted using SHAP analysis to overcome the ET model's lack of explainability. Based on the SHAP analysis, it was determined that temperature, humidity, and wind speed emerged as the primary determinants in forecasting PM2.5 levels.

Robust mortality prediction on a recirculating aquaculture system.

Aquaculture presents itself as one of the most rapidly developing means of sustainable production of animal protein to feed ever-growing populations. Recirculating aquaculture systems offer higher control and fewer inconveniences than traditional systems, making them an attractive option for fish production. Although the sector's digitalization is in its early stages, its application should increase its rentability while conserving the environment. This paper aims to promote the sector's evolution by assessing parameter importance in mortality with tree-based machine learning models, verifying the method's natural robustness and how it compares to a specially devised one, and at the same time evaluating the concept's relevance in predicting categorical mortality values. In particular, to better understand the aquaculture production process through a systematic data evaluation, an exploration based on real-time data acquisition is fully needed. Moreover, algorithm robustness is a key ingredient in this application since measurements are greatly affected by errors. This invalidates the application of traditional machine learning methods, where models are sensitive to production data variations and sensor noise. The study found the parameters that play relevant roles in the production phases, such as pH and nitrate concentration. While the obtained predictive metrics are still sub-optimal, further enhancements could be achieved through rigorous analysis of feature engineering, fine-tuning model hyperparameters, and exploring more advanced algorithms. Additionally, incorporating larger and more diverse datasets, refining data pre-processing techniques, and iteratively optimizing the model architecture may contribute to significant improvements in predictive performance. Despite that, the impact costs of using adjusted machine learning metrics are clear, as are the importance of data rounding in pre-processing and directions for improvement regarding data acquisition and transformation.

Assessing the virtual reality perspectives and self-directed learning skills of nursing students: A machine learning-enhanced approach

AimThis study aims to investigate nursing students’ perspectives on virtual reality technologies and their self-directed learning skills, specifically focusing on how these variables interact and influence each other in the context of nursing education. We also discern potential disparities in these skills based on descriptive characteristics, using both traditional statistical and advanced machine learning approaches for a comprehensive analysis. BackgroundRapid developments in technology, particularly during the Covid-19 pandemic, have brought virtual reality technologies to the forefront of nursing education. However, there is a gap in understanding how nursing students’ perceptions of these technological relate to their development of self-directed learning skills. DesignA descriptive and cross-sectional study design is employed to both quantify nursing students’ perspectives on virtual reality in their education and assess their self-directed learning skills. This approach integrates traditional statistical methods with advanced machine learning techniques, with the intention of offering a comprehensive and nuanced analysis to inform future teaching strategies in nursing. MethodsThe study used a blend of survey scales and a tree-based machine learning model to measure and analyze nursing students’ views, attitudes and self-directed learning levels. This dual approach allows for a more detailed assessment of the factors influencing self-directed learning abilities. Traditional statistical techniques were also applied to assess the reliability of the machine learning findings. ResultsFindings reveal that nursing students generally held positive views towards virtual reality technologies and exhibited a high level of self-directed learning skills. Notable differences in self-directed learning skills were influenced by gender on the overall scale (p <0.001), with male students scoring higher than their female counterparts in both specific sub-dimensions and on the overall scale, but not by academic year. The machine learning analysis provided deeper insights into these variations, highlighting subtle distinctions in student demographics that traditional statistical methods did not fully capture. ConclusionsThe study offers valuable insights into interconnected nature of nursing students’ views on virtual reality technologies and their self-directed learning skills. The results support the integration of virtual reality in nursing curriculum programs and underscore the importance of customizing teaching strategies based on insights gained from machine learning analyses. This approach has the potential to substantially improve both the learning experience and the overall quality of nursing education.

Hybrid decision tree-based machine learning models for diabetes prediction

Hybrid decision tree-based machine learning models for diabetes prediction

Unveiling the driving patterns of carbon prices through an explainable machine learning framework: Evidence from Chinese emission trading schemes

Effectively modeling carbon prices while maintaining interpretability is essential, given the potential risks associated with unexpected price fluctuations. To this end, this study proposes an explainable machine learning (XML) framework to predict and explain carbon prices in China’s three representative carbon markets: Shenzhen, Hubei, and Beijing. Leveraging the strengths of tree-based machine learning models and Tree SHAP algorithms, we unveil global and local explanations in the driving patterns of carbon prices. Our findings indicate that the distribution of local explanatory effects exhibits asymmetric and long-tailed characteristics. Notably, the top global drivers in Shenzhen, Hubei, and Beijing are the photovoltaic price index, coal prices, and the industrial added value of electricity sectors, respectively. Furthermore, we uncover the nonlinear impacts of key drivers on individual carbon price predictions, and identify three key interaction patterns through the calculation of SHAP interaction values. Lastly, we evaluate the explainable performance of various XML benchmarks to validate the superiority of our XML framework, as well as demonstrate its economic significance via a switching trading strategy. Our research provides a reference for mitigating large price fluctuations in local carbon price samples, thereby promoting the stability of carbon markets.

Cluster-based prognostication in glioblastoma: Unveiling heterogeneity based on diffusion and perfusion similarities.

While the association between diffusion and perfusion MRI and survival in glioblastoma is established, prognostic models for patients are lacking. This study employed clustering of functional imaging to identify distinct functional phenotypes in untreated glioblastomas, assessing their prognostic significance for overall survival. A total of 289 patients with glioblastoma who underwent preoperative multimodal MR imaging were included. Mean values of apparent diffusion coefficient (ADC) normalized relative cerebral blood volume (nrCBV), and relative cerebral blood flow (rCBF) were calculated for different tumor compartments and the entire tumor. Distinct imaging patterns were identified using Partition Around Medoids (PAM) clustering on the training dataset, and their ability to predict overall survival was assessed. Additionally, tree-based machine-learning models were trained to ascertain the significance of features pertaining to cluster membership. Using the training dataset (231/289) we identified two stable imaging phenotypes through PAM clustering with significantly different overall survival (OS). Validation in an independent test set revealed a high-risk group with a median OS of 10.2 months and a low-risk group with a median OS of 26.6 months (p=0.012). Patients in the low-risk cluster had high diffusion and low perfusion values throughout, while the high-risk cluster displayed the reverse pattern. Including cluster membership in all multivariate Cox regression analyses improved performance (p≤ 0.004 each). Our research demonstrates that data-driven clustering can identify clinically relevant, distinct imaging phenotypes, highlighting the potential role of diffusion and perfusion MRI in predicting survival rates of glioblastoma patients.