Combining Machine Learning Research Articles

Zakat in metaverse? Evidence from cyberspace

Purpose This study aims to explore the recent state of zakat metaverse innovation from unstructured data available in cyberspace, i.e. YouTube, Instagram, X (Twitter), Facebook, LinkedIn and Google. Design/methodology/approach This study used “zakat metaverse” keywords to harvest unstructured data and analysed using a mixed-method approach. First step of the analysis applied quantitative text analytics via machine learning tool, followed by the final step of qualitative inductive analysis. Findings Quantitative text analytics identified keywords related to zakat metaverse innovation, whereas qualitative analysis explored the critical insights behind those keywords, presented in thematic interpretation. Research limitations/implications This study only used unstructured internet data, in which other relevant information may not be covered. Practical implications Shariah evaluation of zakat obligations from virtual assets requires the relevantisation of fiqh (Islamic jurisprudence) zakat, which opens future debates. Social implications Many zakat institutions operate in emerging economies where digital poverty occurs, and such zakat metaverse innovation would potentially contribute to this digital divide. The relevance of such innovation becomes a major question regarding its inclusivity. Originality/value This study combines machine learning analytics and qualitative analysis to explore the recent state of metaverse innovation in zakat administration.

Embedded physical constraints in machine learning to enhance vegetation phenology prediction

ABSTRACT Vegetation phenology plays a pivotal role in ecological processes on terrestrial surfaces and the interactions between the biosphere and atmospheric feedback. Current attempts to retrieve vegetation phenology have primarily depended on vegetation indices extracted from satellite remote sensing imagery. These approaches often neglect the underlying physical mechanisms associated with climatic factors, and there is a notable absence of evaluations and comparisons with field-observed phenology inventory data. To address these limitations, this paper proposes an innovative physical constraint neural networks (PCNNs) model that combines machine learning techniques with physical mechanisms to enhance the accuracy of vegetation phenology predictions. By incorporating meteorological variables into a machine learning model and by using the Moderate-Resolution Imaging Spectroradiometer (MODIS) dataset to identify the vegetation phenology of four vegetation types in North America, this study delved into the relationship between vegetation phenology and climate factors as well as its impacts on ecosystems. Our model demonstrated high accuracy compared to machine learning methods without physical mechanisms when validated by field observations from PhenoCam and the USA National Phenology Network (USA-NPN) spanning from 2001 to 2021. The results show that the overall root mean square error (RMSE) with physical constraints is reduced to 12.37 days, higher by 2.6 days than the machine learning method without physical constraints. We compared four vegetation types using different machine learning and traditional rule-based methods, deciduous vegetation (DV) exhibited the most favorable prediction results, with the RMSE and mean bias error (MBE) as low as 5.71 days and 4.06 days in PCNNs model, respectively. This was followed by evergreen needle-leaved forests and mixed forests with RMSE of 12.32 and 13.28 days, respectively. The stressed deciduous vegetation type had the worst prediction result of 19.86 days (RMSE), and the weighted index of agreement (WIA) attained a value of 0.68. These findings suggest that the model embedded with physical constraints significantly boosted prediction accuracy for four common vegetation types, particularly for DV, compared to the unconstrained ML model. It offers valuable insights into the incorporation of physical mechanisms within machine learning models. This research paves the way for substantial advancements in the field of land surface phenology, enabling more accurate and reliable predictions of vegetation phenology in various ecological and climatic contexts.

Research on Spatial Heterogeneity, Impact Mechanism, and Carbon Peak Prediction of Carbon Emissions in the Yangtze River Delta Urban Agglomeration

Urban agglomerations with a high economic activity and population density are key areas for carbon emissions and pioneers in achieving carbon peaking and the Sustainable Development Goals (SDGs). This study combines machine learning with an extended STIRPAT (Stochastic Impacts by Regression on Population, Affluence, and Technology) model to uncover the mechanisms driving carbon peaking disparities within these regions. It forecasts carbon emissions under different scenarios and develops indices to assess peaking pressure, reduction potential, and driving forces. The findings show significant carbon emission disparities among cities in the Yangtze River Delta, with a fluctuating downward trend over time. Technological advancement, population size, affluence, and urbanization positively impact emissions, while the effects of industrial structure and foreign investment are weakening. Industrially optimized cities lead in peaking, while others—such as late-peaking and economically radiating cities—achieve peaking only under the ER scenario. Cities facing population loss and demonstration cities fail to peak by 2030 in any scenario. The study recommends differentiated carbon peaking pathways for cities, emphasizing tailored targets, pathway models, and improved supervision. This research offers theoretical and practical insights for global urban agglomerations aiming to achieve early carbon peaking.

RERE-AS1 enhances the effect of CDK4/6 inhibitor Ribociclib and suppresses malignant phenotype in breast cancer via MEK/ERK pathway

BackgroundCurrently, there is a lack of biomarkers to identify breast cancer (BC) patients who would benefit from CDK4/6 inhibitors. This study combined machine learning (ML) algorithms based on transcriptomic data with both in vivo and in vitro experiments to identify therapeutic efficacy-related biomarkers of the CDK4/6 inhibitor ribociclib from the perspective of long non-coding RNA (lncRNA).MethodsWe used the Genomics of Drug Sensitivity in Cancer database along with the “oncoPredict” algorithm to calculate the half maximal inhibitory concentration (IC50) values for ribociclib based on transcriptome data. ML algorithms were utilized to select key lncRNAs related to ribociclib and to establish a model which could be used for selection of potential beneficiaries of ribociclib. Cellular experiments were conducted to validate the ML analysis and explore the potential biological mechanisms by which RERE-AS1 influences ribociclib efficacy and malignant phenotype of BC cells. Correlation analysis with clinical pathological factors, RT-qPCR experiments on tissue specimens, and pan-cancer analysis were carried out to explore the expression pattern, and the prognostic and diagnostic potential of RERE-AS1 in cancers.ResultsWe have identified 11 key ribociclib-related lncRNAs and constructed an artificial neural network model (ANNM) based on lncRNA. Cellular experiments demonstrated that overexpression of RERE-AS1 promoted the anti-tumor activity of ribociclib in BC cells. Furthermore, RERE-AS1 is crucial in suppressing the malignant traits of BC cells through the reduction of MEK and ERK phosphorylation levels. Patients with smaller primary tumors and lower pathological stage exhibited higher levels of RERE-AS1 expression. Lastly, a pan-cancer analysis revealed that RERE-AS1 exhibits distinctly abnormal expression patterns, prognostic significance, and clinical diagnostic value in BC, compared to other cancers.ConclusionsThe ANNM established through ML algorithms can serve as predictive indicators for the efficacy of ribociclib in BC patients. LncRNA RERE-AS1, a newly discovered biomarker, holds significant promise for diagnosis, treatment, and enhancing the therapeutic response to ribociclib in BC.Graphical

Global Genotype by Environment Prediction Competition Reveals That Diverse Modeling Strategies Can Deliver Satisfactory Maize Yield Estimates.

Predicting phenotypes from a combination of genetic and environmental factors is a grand challenge of modern biology. Slight improvements in this area have the potential to save lives, improve food and fuel security, permit better care of the planet, and create other positive outcomes. In 2022 and 2023 the first open-to-the-public Genomes to Fields (G2F) initiative Genotype by Environment (GxE) prediction competition was held using a large dataset including genomic variation, phenotype and weather measurements and field management notes, gathered by the project over nine years. The competition attracted registrants from around the world with representation from academic, government, industry, and non-profit institutions as well as unaffiliated. These participants came from diverse disciplines include plant science, animal science, breeding, statistics, computational biology and others. Some participants had no formal genetics or plant-related training, and some were just beginning their graduate education. The teams applied varied methods and strategies, providing a wealth of modeling knowledge based on a common dataset. The winner's strategy involved two models combining machine learning and traditional breeding tools: one model emphasized environment using features extracted by Random Forest, Ridge Regression and Least-squares, and one focused on genetics. Other high-performing teams' methods included quantitative genetics, machine learning/deep learning, mechanistic models, and model ensembles. The dataset factors used, such as genetics; weather; and management data, were also diverse, demonstrating that no single model or strategy is far superior to all others within the context of this competition.

Screening of Ionic Liquids for Efficient CO2 Cycloaddition Catalysis under Mild Condition: A Combined Machine Learning and DFT Approach

Screening of Ionic Liquids for Efficient CO₂ Cycloaddition Catalysis under Mild Condition: A Combined Machine Learning and DFT Approach

Analysis of Enzyme Interference Factors During Storage of Millet Based on Machine Learning

The color of millet beige is one of the important factors related to the quality of millet. As people's living standards continue to improve, consumers and the millet industry market are increasingly demanding quality. The fading problem of millet during storage has seriously affected the edible quality and commercial value of millet. However, there are few reports on the mechanism of millet storage deterioration so far. This paper combines machine learning algorithms to systematically study the relationship between lipoxygenase and the color of millet fading from the physiological and molecular levels. Moreover, this paper uses multiple sets of variables to conduct experimental research to explore the molecular mechanism of millet fading phenomenon, which not only provides a theoretical basis for delaying millet aging and ensuring millet storage quality, but also has important significance for the breeding of millet beige quality.

Determination of Geochemical Background and Baseline and Research on Geochemical Zoning in the Desert and Sandy Areas of China

Resources in deserts and sandy landscapes have potential for development, but existing surveys and sampling have not collected desert soil samples. As such, the geochemical background of these spaces remains unexplored due to the vastness and desolation of deserts. Therefore, researching the geochemical background values and geochemical baseline values of deserts is of long-term significance. Our research indicates that in addition to macrostructural environmental divisions, microelement geochemistry can also be used for geological unit zoning. In this paper, geochemical background and geochemical baseline values of 61 desert elements were calculated using the iterative method, frequency histograms method, and multifractal concentration-area method. It also analyzes the distribution characteristics of major, trace, and rare earth elements, and divides the 12 desert sand regions into different geochemical zones. This paper determines, for the first time, the geochemical background values of elements in Chinese deserts, filling the gap in the study of desert background values. By combining machine learning methods, different deserts have been divided into three geochemical zones. This research will greatly enhance our ability to interpret the geochemical distribution and evolutionary patterns of desert elements in China, and it has important scientific significance and practical value for desert research.

Application of machine learning for detecting and tracking turbulent structures in plasma fusion devices using ultra fast imaging

This study explores the application of machine learning techniques for detecting and tracking plasma filaments around the boundary of magnetically confined tokamak plasmas. Plasma filaments, also called blobs, are responsible for enhanced turbulent transport across magnetic field lines, and their accurate characterization is crucial for optimizing the performance of magnetic fusion devices. We present a novel approach that combines machine learning methods applied to data obtained from ultra-fast cameras, including YOLO (You Only Look Once) for object detection, semantic segmentation, and specific tracking methods. This approach enables fast and accurate detection and tracking of filaments while overcoming the limitations of conventional methods, which are time-consuming and prone to human subjectivity. A significant advance in our study lies in the development of a method for automatically labeling a large batch of data, which greatly facilitates the training of supervised machine learning algorithms. Using these techniques, we obtained promising results demonstrating a significant improvement over conventional tracking methods, achieving a detection accuracy of up to 98.8%, while reducing the inference time per frame by 15% to 31% compared to conventional Kalman filter tracking. These results open up new perspectives for investigating turbulent phenomena in tokamaks, and could have important implications for the development of controlled nuclear fusion.

More than words: the role of personality in shaping the timeliness of online reviews

Purpose The purpose of this paper is to investigate the potential influence of the big five personality traits − extraversion, neuroticism, agreeableness, openness and conscientiousness − on the time taken by travelers to submit online reviews after their hotel stay. Design/methodology/approach The study analyzed 83,235 TripAdvisor reviews from 415 hotels in six major US tourism cities using random forest algorithms and Poisson regression. The research investigated the influence of the big five personality traits on the time taken by travelers to submit online reviews post-hotel stay, merging personality psychology with consumer behavior research through a combination of machine learning and statistical analysis. Findings The findings reveal significant correlations between certain personality traits and the time taken to post online hotel reviews. Extraversion, neuroticism and agreeableness were found to be negatively correlated with response time, suggesting that individuals scoring higher in these traits tend to submit their reviews more quickly. Conversely, openness exhibited a positive correlation, indicating that those with higher levels of openness tend to delay their feedback. Conscientiousness showed no significant correlation with response time. Originality/value This study represents a novel approach to understanding the relationship between personality traits and online review behavior in the hospitality industry. By leveraging advanced machine learning techniques, such as random forest algorithms, alongside traditional statistical methods like Poisson regression, this research offers a unique perspective on the influence of personality on consumer behavior. The innovative application of these technologies to a large data set of TripAdvisor reviews provides fresh insights that can inform the development of personalized customer engagement strategies. The findings contribute to the growing body of literature on the intersection of personality psychology, consumer behavior and hospitality management in the digital age.

Machine Learning‐Guided Design of 10 nm Junctionless Gate‐All‐Around Metal Oxide Semiconductor Field Effect Transistors for Nanoscaled Digital Circuits

In this paper, we introduce an innovative design approach based on combined numerical simulations and machine learning (ML) analysis to investigate the design key parameters of ultra‐low scale junctionless gate‐all‐around (JLGAA) field‐effect transistor (FET) devices. To this end, precise 3D numerical models that incorporate quantum effects and ballistic transport are employed to simulate the current–voltage (I–V) characteristics of 10 nm‐scale JLGAA FET devices. The influence of design parameter variations and high‐k dielectric material on the subthreshold characteristics is thoroughly examined. Various ML algorithms were employed to analyze and classify the key design parameters influencing the subthreshold figures‐of‐merit (FoMs), the subthreshold swing (SS) factor and ION/IOFF ratio. The obtained results highlight that channel radius and channel doping design parameters are particularly important for affecting swing factor behavior. Similarly, these features also play a significant role in predicting and affecting ION/IOFF current ratio values. Additionally, machine learning is used to determine the optimal design parameters for each figure of merit (FoM) output value. In this context, the models effectively predicted both ION/IOFF current ratios and SS classification, with Naive Bayes achieving an accuracy of 90.8% for ION/IOFF and 92.6% for SS, showcasing the model's robustness in these classification tasks.

Identification of a 5-Plex Cytokine Signature that Differentiates Patients with Multiple Systemic Inflammatory Diseases.

Patients with non-infectious systemic inflammation may suffer from one of many diseases, including hyperinflammation (HI), autoinflammatory disorders (AID), and systemic autoimmune disease (AI). Despite their clinical overlap, the pathophysiology and patient management differ between these disorders. We aimed to investigate blood biomarkers able to discriminate between patient groups. We included 44 patients with active clinical and/or genetic systemic inflammatory disease (9 HI, 27 AID, 8 systemic AI) and 16 healthy controls. We quantified 55 serum proteins and combined multiple machine learning algorithms to identify five proteins (CCL26, CXCL10, ICAM-1, IL-27, and SAA) that maximally separated patient groups. High ICAM-1 was associated with HI. AID was characterized by an increase in SAA and decrease in CXCL10 levels. A trend for higher CXCL10 and statistically lower SAA was observed in patients with systemic AI. Principal component analysis and unsupervised hierarchical clustering confirmed separation of disease groups. Logistic regression modelling revealed a high statistical significance for HI (P = 0.001), AID, and systemic AI (P < 0.0001). Predictive accuracy was excellent for systemic AI (AUC 0.94) and AID (0.91) and good for HI (0.81). Further research is needed to validate findings in a larger prospective cohort. Results will contribute to a better understanding of the pathophysiology of systemic inflammatory disorders and can improve diagnosis and patient management.

Diagnosis of heart disease using an advanced triple hybrid algorithm combining machine learning techniques

Purpose This study aims to improve the detection and quantification of cardiac issues, which are a leading cause of mortality globally. By leveraging past data and using knowledge mining strategies, this study seeks to develop a technique that could assess and predict the onset of cardiac sickness in real time. The use of a triple algorithm, combining particle swarm optimization (PSO), artificial bee colony (ABC) and support vector machine (SVM), is proposed to enhance the accuracy of predictions. The purpose is to contribute to the existing body of knowledge on cardiac disease prognosis and improve overall performance in health care. Design/methodology/approach This research uses a knowledge-mining strategy to enhance the detection and quantification of cardiac issues. Decision trees are used to form predictions of cardiovascular disorders, and these predictions are evaluated using training data and test results. The study has also introduced a novel triple algorithm that combines three different combination processes: PSO, ABC and SVM to process and merge the data. A neural network is then used to classify the data based on these three approaches. Real data on various aspects of cardiac disease are incorporated into the simulation. Findings The results of this study suggest that the proposed triple algorithm, using the combination of PSO, ABC and SVM, significantly improves the accuracy of predictions for cardiac disease. By processing and merging data using the triple algorithm, the neural network was able to effectively classify the data. The incorporation of real data on various aspects of cardiac disease in the simulation further enhanced the findings. This research contributes to the existing knowledge on cardiac disease prognosis and highlights the potential of leveraging past data for strategic forecasting in the health-care sector. Originality/value The originality of this research lies in the development of the triple algorithm, which combines multiple data mining strategies to improve prognosis accuracy for cardiac diseases. This approach differs from existing methods by using a combination of PSO, ABC, SVM, information gain, genetic algorithms and bacterial foraging optimization with the Gray Wolf Optimizer. The proposed technique offers a novel and valuable contribution to the field, enhancing the competitive position and overall performance of businesses in the health-care sector.

Pt-Driven Stable and Metastable Ternary Compounds with Tunable Partial Flat Bands in Immiscible Si-Sn System.

A combined machine learning approach and first-principles calculations are employed to efficiently explore energetically favorable ternary compounds involving Pt and an immiscible Si-Sn pair of elements. We predict two stable Pt-Si-Sn ternary compounds (Pt2Si2Sn and Pt8SiSn2) and reproduce the experimentally synthesized Pt5SiSn. The crystal structures of Pt5SiSn and Pt8SiSn2 are composed of a SiPt8 cube and SnPt12 cuboctahedron. Based on these structural motifs, we design a series of Pt-Si-Sn ternary compounds that contain SiPt8 and SnPt12 units and discover some stable and metastable compounds. Partial flat bands near the Fermi level appear in these compounds, and the positions of partial flat bands move upward with increasing the Pt content. The partial flat bands are dominated by Pt atoms. These Pt-Si-Sn ternary compounds provide a platform to study flat band-related properties. Our results open a way to design new compounds in the ternary system.

Pioneering Klebsiella Pneumoniae Antibiotic Resistance Prediction With Artificial Intelligence-Clinical Decision Support System-Enhanced Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry: Retrospective Study.

The rising prevalence and swift spread of multidrug-resistant gram-negative bacteria (MDR-GNB), especially Klebsiella pneumoniae (KP), present a critical global health threat highlighted by the World Health Organization, with mortality rates soaring approximately 50% with inappropriate antimicrobial treatment. This study aims to advance a novel strategy to develop an artificial intelligence-clinical decision support system (AI-CDSS) that combines machine learning (ML) with matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS), aiming to significantly improve the accuracy and speed of diagnosing antibiotic resistance, directly addressing the grave health risks posed by the widespread dissemination of pan drug-resistant gram-negative bacteria across numerous countries. A comprehensive dataset comprising 165,299 bacterial specimens and 11,996 KP isolates was meticulously analyzed using MALDI-TOF MS technology. Advanced ML algorithms were harnessed to sculpt predictive models that ascertain resistance to quintessential antibiotics, particularly levofloxacin and ciprofloxacin, by using the amassed spectral data. Our ML models revealed remarkable proficiency in forecasting antibiotic resistance, with the random forest classifier emerging as particularly effective in predicting resistance to both levofloxacin and ciprofloxacin, achieving the highest area under the curve of 0.95. Performance metrics across different models, including accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and F1-score, were detailed, underlining the potential of these algorithms in aiding the development of precision treatment strategies. This investigation highlights the synergy between MALDI-TOF MS and ML as a beacon of hope against the escalating threat of antibiotic resistance. The advent of AI-CDSS heralds a new era in clinical diagnostics, promising a future in which rapid and accurate resistance prediction becomes a cornerstone in combating infectious diseases. Through this innovative approach, we answered the challenge posed by KP and other multidrug-resistant pathogens, marking a significant milestone in our journey toward global health security.

Improving the Performance of Screening MM/PBSA in Protein-Ligand Interactions via Machine Learning

Abstract Accurately estimating protein-ligand binding free energy is crucial for drug design and biophysics, yet remains a challenging task. In this study, we applied the screening molecular mechanics/Poisson Boltzmann surface area (MM/PBSA) method in combination with various machine learning techniques to compute the binding free energies of protein-ligand interactions. Our results demonstrate that machine learning outperforms direct screening MM/PBSA calculations in predicting protein-ligand binding free energies. Notably, the random forest (RF) method exhibited the best predictive performance, with a Pearson correlation coefficient (r p ) of 0.702 and a mean absolute error (MAE) of 1.379 kcal/mol. Furthermore, we analyzed feature importance rankings in the gradient boosting (GB), adaptive boosting (AdaBoost), and RF methods, and found that feature selection significantly impacted predictive performance. In particular, molecular weight (MW) and van der Waals (VDW) energies played a decisive role in the prediction. Overall, this study highlights the potential of combining machine learning methods with screening MM/PBSA for accurately predicting binding free energies in biosystems.

Enhancing and Explaining Energy Management in Smart Manufacturing Production via ML Within the Context of Industry 4.0

Objectives: This study aims to explore the integration of big data and machine learning to enhance energy efficiency in manufacturing, focusing on the opportunities presented by Industry 4.0 and cyber-physical production systems. Theoretical Framework: The research leverages supervised learning techniques to analyze and predict machine-specific energy consumption patterns, utilizing energy disaggregation as a foundational concept. Method: Various machine learning algorithms, including Lasso regression, Linear regression, and Decision Tree, will be employed to develop predictive models for energy usage. Additionally, Explainable Machine Learning (XML) techniques will be utilized to ensure interpretability and clarity in the prediction outcomes. Results and Discussion: The proposed framework aims to elucidate the relationship between equipment utility and energy consumption, providing comprehensible explanations that enhance decision-making processes within intelligent industrial environments. Research Implications: This work highlights the significant potential of XML in transforming machine learning applications in manufacturing, paving the way for improved energy efficiency and operational effectiveness. Originality/Value: The introduction of an innovative framework that combines machine learning with explainability in the context of energy consumption marks a valuable contribution to the fields of manufacturing and sustainable industry practices.

A Large Language Model and Qualitative Comparative Analysis-Based Study of Trust in E-Commerce

The primary goal of this study is to predict and analyze customer trust in e-commerce by leveraging neural computation within large language models (LLMs) alongside configurational approaches. We employ LLMs to predict trust levels based on customer reviews, applying artificial intelligence to analyze key aspects of the e-commerce experience, such as customer service, refund processes, item quality, and shipping. To extend beyond predictive performance, we integrate Qualitative Comparative Analysis (QCA) to identify the causal relationships between trust and various stages of the customer journey, including selection, delivery, and post-purchase support (recovery). This dual approach not only showcases the power of neural computation in predicting trust outcomes but also provides a deeper understanding of how specific configurations of customer experience elements contribute to either positive or negative trust. By combining machine learning techniques and QCA, this study contributes to the application of LLMs and configurational approaches, offering novel insights into the drivers of trust in e-commerce.

Transferable and Interpretable Prediction of Site-Specific Dehydrogenation Reaction Rate Constants with NMR Spectra.

Accurate and efficient determination of site-specific reaction rate constants over a wide temperature range remains challenging, both experimentally and theoretically. Taking the dehydrogenation reaction as an example, our study addresses this issue by an innovative combination of machine learning techniques and cost-effective NMR spectra. Through descriptor screening, we identified a minimal set of NMR chemical shifts that can effectively determine reaction rate constants. The constructed model performs exceptionally well on theoretical data sets and demonstrates impressive generalization capabilities, extending from small molecules to larger ones. Furthermore, this model shows outstanding performance when applied to limited experimental data sets, highlighting its robust applicability and transferability. Utilizing the Sure Independence Screening and Sparsifying Operator (SISSO) algorithm, we also present an interpretable rate constant-temperature-NMR (k-T-NMR) relationship with a mathematical formula. This study reveals the great potential of combining machine learning with easily accessible spectroscopic descriptors in the study of reaction kinetics, enabling the rapid determination of reaction rate constants and promoting our understanding of reactivity.

Combining array-assisted SERS microfluidic chips and machine learning algorithms for clinical leukemia phenotyping

The disease progression and treatment options of leukemia between different subtypes vary considerably, emphasizing the importance of phenotyping. However, early typing of leukemia remains challenging due to the lack of highly sensitive and specific analytical tools. Herein, we propose a SERS-based platform for the classification of acute lymphoblastic T-cell leukemia (T-ALL) and chronic myeloid leukemia (CML) through the combination of machine learning and microfluidic chips. The ordered arrays in microfluidic channels reshape the microscopic flow field and contacting interfaces, facilitating the uniform and efficient capture of tumor cells. To enable phenotypic analysis, spectrally orthogonal SERS aptamer nanoprobes were applied, providing composite spectral signatures of individual cells in accordance with surface protein expression. Further, machine learning algorithms were employed to analyze the SERS signatures automatically, resulting in an accuracy of 98.6 % for 73 clinical blood samples. The results demonstrate that this platform holds promising potential for clinical leukemia diagnosis and precision medicine.