Integrated Machine Learning Approach Research Articles

Strategic reuse of rapid antigen tests for coagulation status assessment: an integrated machine learning approach

Addressing the pressing demand for rapid and inexpensive coagulation testing in cardiovascular care, this study introduces a novel application of repurposed COVID-19 rapid antigen tests (RATs) as paper-based lateral flow assays (LFAs) combined with machine learning for coagulation status evaluation. By further developing a mobile app prototype, we present a platform that enables clinicians to perform immediate and accurate anticoagulant dosing adjustments using existing post-pandemic resources. Our proof-of-concept employs a random forest machine learning classifier to interpret image feature variations on RAT NC membrane, correlating red blood cell (RBC) wicked diffusion distance in recalcified citrated whole blood with changes in coagulative viscosity, easily interpreted. Enhanced by confocal imaging studies of paper microfluidics, our approach provides insights into the mechanisms dissecting coagulation components, achieving high classification precision, recall, and F1-scores. The inverse relationship between RBC wicked diffusion distance and enoxaparin concentration paves the way for machine learning to inform real-time dose prescription adjustments, aligning with individual patient profiles to optimize therapeutic outcomes. This study not only demonstrates the potential of leveraging surplus RATs for coagulation management but also exemplifies a cost-effective, rapid, and smart strategy to enhance clinical decision-making in the post-pandemic era.Graphical

Enhancing agricultural wireless sensor network security through integrated machine learning approaches

AbstractWireless sensor network (WSN) works with a collection of multiple sensor nodes to fetch the data from the deployed environment to fulfill the application whether it is agricultural monitoring, industrial monitoring, etc. The agricultural region can be monitored by deploying sensor nodes to multiple verticals where continuous human presence is not feasible. These devices are equipped with limited resources and are easily vulnerable to various cyber‐attacks. The attacker can hack the sensor nodes to steal critical information from WSN devices. The cluster heads in the WSN play a vital role in the process of routing data packets and attackers launch malicious codes through sender nodes to hack or damage the cluster heads to shut down the entire deployed network of agricultural regions. This research paper proposes a framework to improve the security of WSNs by providing a shield to the cluster heads of the network using machine learning techniques. The experimental study of the paper includes the comparative analysis of three machine learning techniques decision tree classifier, Gaussian Naïve Bayes, and random forest classifier for predicting WSN attacks like flooding, gray hole, blackhole, and TDMA that are deployed to support the proposed WSN security framework on the attack dataset. The random forest classifier achieves an accuracy of 98%, Precision of 97.6%, Recall of 97.6%, and F1 score of 97.8% which is the maximum among the deployed machine learning techniques.

An exploration of descriptive machine learning approaches for antimicrobial resistance: Multidrug resistance patterns in Salmonella enterica

Salmonellosis is one of the most common foodborne diseases worldwide, with the ability to infect humans and animals. Antimicrobial resistance (AMR) and, particularly, multidrug resistance (MDR) among Salmonella enterica poses a risk to human health. Antimicrobial use (AMU) regulations in livestock have been implemented to reduce AMR and MDR in foodborne pathogens. In this study, we used an integrated machine learning approach to investigate Salmonella AMR and MDR patterns before and after the implementation of AMU restrictions in agriculture in the United States. For this purpose, Salmonella isolates from cattle in the National Antimicrobial Resistance Monitoring System (NARMS) dataset were analysed using three descriptive models consisting of hierarchical clustering, network analysis, and association rule mining. The analysis showed the impact of the United States’ 2012 extra-label cephalosporin regulations on AMR trends and revealed a distinctive MDR pattern in the Dublin serotype. The results also indicated that each descriptive model provides insights on a specific aspect of resistance patterns and, therefore, combining these approaches make it possible to gain a deeper understanding of AMR.

Enhancing Breast Cancer Subtype Classification through GediNET: Integrating Disease- Disease Association Data with a Grouping-Scoring-Modeling Approach

The development of sequencing technology and the increase in biological data repositories have allowed for a more thorough understanding of the complex molecular aspects of diseases like cancer. This paper evaluates GediNET, an integrative machine learning approach that employs a Grouping-Scoring-Modeling (GSM) approach to classify different molecular subtypes of breast cancer using the BRCA LumAB_Her2Basal dataset against different feature selection approaches and machine learning classifiers. GediNET distinguishes itself from traditional feature selection methods by analyzing groups of genes to identify relevant disease-disease associations and potential biomarkers. The results of our study show that GediNET performs better than traditional approaches in terms of accuracy and Area Under the Curve (AUC) metrics. This demonstrates that GediNET is effective in understanding the genetic intricacies of breast cancer. This approach improves the identification of molecular subtypes and promotes the development of targeted medicines and customized medicine.

CGSX Ensemble: An Integrative Machine Learning and Deep Learning Approach for Improved Diabetic Retinopathy Classification

This research proposes an integrated approach for automated diabetic retinopathy (DR) diagnosis, leveraging a combination of machine learning and deep learning techniques to extract features and perform classification tasks effectively. Through preprocessing of retinal images to enhance features and mitigate noise, two distinct methodologies are employed: machine learning feature extraction, targeting texture features like Gray-Level Co-occurrence Matrix (GLCM) and Gray-Level Run Length Matrix (GLRLM), and deep learning feature extraction, utilizing pre-trained convolutional neural networks (CNNs) such as VGG, ResNet, or Inception. Following feature extraction, various classifiers, including Support Vector Machines (SVM), Random Forests, and Gradient Boosting Machines, are trained on the extracted features for DR classification. Alternatively, deep learning classifiers like CNNs or recurrent neural networks (RNNs) may be trained directly on the extracted features or on raw images. This comprehensive framework shows promising potential to improve the accuracy and efficiency of diabetic retinopathy (DR) diagnosis, enabling timely intervention and management of this vision-threatening condition.

Enhancing Patient Flow in Emergency Departments: A Machine Learning and Simulation-Based Resource Scheduling Approach

The efficient scheduling of resources within emergency departments (EDs) is crucial to minimizing patient length of stay (LoS) times and maximizing the utilization of limited resources. Reducing patient wait times can enhance the operation of emergency departments and improve patient satisfaction and the quality of medical care. This study develops a simulation model using Discrete Event Simulation (DES) methodology, examining six resource scheduling policies that consider different combinations of general and senior physicians. By leveraging six scheduling policies and machine learning techniques, this model dynamically identifies the most effective scheduling policy, based on a comprehensive dataset of ED visits in South Korea. The ED simulation achieves an accuracy rate of 90% and demonstrates that our proposed integrated machine learning approach reduces average length of stay (LoS) to approximately 322.91 min, compared to 327.10 min under traditional methods. This study underscores the potential of integrating DES and machine learning to enhance resource management in EDs.

Assessing and predicting green gentrification susceptibility using an integrated machine learning approach

ABSTRACT Greenery initiatives, such as green infrastructures (GIs), create sustainable and climate-resilient environments. However, they can also have unintended consequences, such as displacement and gentrification in low-income areas. This paper proposes an integrated machine learning (ML) approach that combines both unsupervised and supervised ML algorithms. First, 35 indicators that contribute to green gentrification were identified and categorised into 7 categories: social, economic, demographic, housing, household, amenities, and GIs. Second, data was collected for all census tracts in New York City. Third, the green gentrification susceptibility was modelled into 6 levels using k-means clustering analysis, which is an unsupervised ML model. Fourth, the Technique for Order of Preference by Similarity to the Ideal Solution (TOPSIS) was used to map the census tracts to their green gentrification susceptibility level. Finally, different supervised ML algorithms were trained and tested to predict the green gentrification susceptibility. The results showed that the artificial neural network (ANN) model is the most accurate in classifying and predicting the green gentrification susceptibility with an overall accuracy of 96%. Moreover, the outcomes showed that the Normal Difference Vegetation Index (NDVI), the proximity to GIs, the GIs frequency, and the total area of GIs were identified as the most important indicators to predict green gentrification susceptibility. Ultimately, the proposed approach allows practitioners and researchers to perform micro-level (i.e. on the census-tracts level) predictions and inferences about green gentrification susceptibility. This allows more focused and targeted mitigation actions to be designed and implemented in the most affected communities, thus promoting environmental justice.

Advancing intrauterine adhesion severity prediction: Integrative machine learning approach with hysteroscopic cold knife system, clinical characteristics and hematological parameters

Intrauterine Adhesion (IUA) constitute a significant determinant impacting female fertility, potentially leading to infertility, miscarriage, menstrual irregularities, and placental complications. The precise assessment of the severity of IUA is pivotal for the customization of personalized treatment plans, aimed at enhancing the success rate of treatments and mitigating reproductive health risks. This study proposes bTLSMA-SVM-FS, a novel feature selection machine learning model that integrates an enhanced slime mould algorithm (SMA), termed TLSMA, with support vector machines (SVM), aiming to develop a predictive model for assessing the severity of IUA. Initially, a series of optimization comparative experiments were conducted on the TLSMA using the CEC 2017 benchmark functions. By comparing it with eleven meta-heuristic algorithms as well as eleven SOTA algorithms, the experimental outcomes corroborated the superior performance of the TLSMA. Subsequently, the developed bTLSMA-SVM-FS model was employed to conduct a thorough analysis of the clinical features of 107 IUA patients from Wenzhou People's Hospital, comprising 61 cases of moderate IUA and 46 cases of severe IUA. The evaluation results of the model demonstrated exceptional performance in predicting the severity of IUA, achieving an accuracy of 86.700 % and a specificity of 87.609 %. Moreover, the model successfully identified critical factors influencing the prediction of IUA severity, including the preoperative Chinese IUA score, production times, thrombin time, preoperative endometrial thickness, and menstruation. The identification of these key factors not only further validated the efficacy of the proposed model but also provided vital scientific evidence for a deeper understanding of the pathogenesis of IUA and the enhancement of targeted treatment strategies.

Performance predication of a solar assisted desiccant air conditioning system using radial basis function neural network: An integrated machine learning approach

In this Paper solar desiccant air conditioning system integrated with cross flow Maisotsenko cycle (M-cycle) indirect evaporative cooler is used to investigate the performance of whole system in different range of parameters. Solar evacuated tube electric heater is used to supply the regeneration temperature to the desiccant wheel, whereas, Desiccant Wheel (DW) and M-cycle is used to handle latent load and sensible load separately. Major contribution of this research is to predict system level performance parameters of a Solar Assisted Desiccant Air Conditioning (Sol-DAC) system using Radial Basis Function Neural Network (RBF-NN) under real transient experimental inlet conditions. Nine parameters are mainly considered as input parameters to train the RBF-NN model, which are, supply Air temperature at the process side of desiccant wheel, supply air humidity ratio at process side of the desiccant wheel, outlet temperature from the desiccant wheel at process side, outlet humidity ratio from the desiccant wheel at process side, regeneration temperature at regeneration side of the DW, outlet temperature from the heat recovery wheel at process side, outlet humidity ratio out from the Heat Recovery Wheel (HRW) at process side, temperature before heat recovery wheel regeneration side of the system, humidity ratio before heat recovery wheel regeneration side of the system. Four parameters are considered as the output of the RBF-NN model, namely: output temperature, output humidity, Cooling Capacity (CC), and Coefficient of Performance (COP). The results of the RBF-NN model shows that the best Mean Squared Error (MSE) and Regression coefficient (R) for outlet temperature prediction are 0.00998279 and 0.99832 when regeneration temperature is 70 °C and inlet humidity at 18 g/kg. Best MSE and R for predication of outlet humidity are 0.0102932 and 0.99485 when the regeneration temperature is 70 °C and inlet humidity at 16 g/kg. Best MSE and R for predication of COP are 0.0106691 and 0.9981 when the regeneration temperature is 70 °C and inlet humidity 12 g/kg. Best MSE and R for predication of CC are 0.0144943 and 0.99711 when the regeneration temperature is 70 °C and inlet humidity 14 g/kg. Experimental and predicted performance parameters were in close agreement and showed minimal deviation. Investigations of predicted results revealed that trained RBF-NN model was capable of predicting the trend of output result under the varying input condition.

Autonomous optimization of process parameters and in-situ anomaly detection in aerosol jet printing by an integrated machine learning approach

Aerosol jet printing (AJP) is a newly-developed technology for printed electronics in additive manufacturing (AM) industry. Compared to conventional printed electronics, AJP technology has the capability of fabricating high-resolution patterns (∼ 10 µm) with multiple materials and high degrees of freedom. However, insufficiently optimizing the printing process, including reliance on traditional trial-and-error methods and theoretical models, may lead to a decline in production quality, consequently limiting the widespread adoption of AJP technology. In this paper, an integrated machine learning approach is developed for autonomous optimization of process parameters and in-situ anomaly detection in AJP. In the developed approach, a coaxial camera is incorporated with a stepwise data-driven modeling approach for optimal operating window identification and process model development of AJP, thereby systematically optimizing the main process parameters prior to printing. Subsequently, a convolutional neural network (CNN) model is developed and integrated with an alignment camera for in-situ process monitoring and printing status evaluation, enabling the anomaly detection of the AJP process. Afterwards, leveraging process similarity through transfer learning allows rapid calibration of detected system drifts or variations in process stability, thereby enhancing the product consistency of AJP. Generally, by utilizing state-of-the-art machine learning techniques, the autonomous optimization of process parameters and anomaly detection in AJP achieve accuracies of 95.3% and 92.7%, respectively. This indicates the effectiveness and systematic nature of the developed approach compared to traditional trial-and-error methods and theoretical models. Furthermore, due to its data-driven characteristics, this approach has the potential to extend its applicability to other noncontact ink writing techniques, facilitating further research for process optimization in AM.

An Integrated Machine Learning Approach for Congestive Heart Failure Prediction.

Congestive heart failure (CHF) is one of the primary sources of mortality and morbidity among the global population. Over 26 million individuals globally are affected by heart disease, and its prevalence is rising by 2% yearly. With advances in healthcare technologies, if we predict CHF in the early stages, one of the leading global mortality factors can be reduced. Therefore, the main objective of this study is to use machine learning applications to enhance the diagnosis of CHF and to reduce the cost of diagnosis by employing minimum features to forecast the possibility of a CHF occurring. We employ a deep neural network (DNN) classifier for CHF classification and compare the performance of DNN with various machine learning classifiers. In this research, we use a very challenging dataset, called the Cardiovascular Health Study (CHS) dataset, and a unique pre-processing technique by integrating C4.5 and K-nearest neighbor (KNN). While the C4.5 technique is used to find significant features and remove the outlier data from the dataset, the KNN algorithm is employed for missing data imputation. For classification, we compare six state-of-the-art machine learning (ML) algorithms (KNN, logistic regression (LR), naive Bayes (NB), random forest (RF), support vector machine (SVM), and decision tree (DT)) with DNN. To evaluate the performance, we use seven statistical measurements (i.e., accuracy, specificity, sensitivity, F1-score, precision, Matthew's correlation coefficient, and false positive rate). Overall, our results reflect our proposed integrated approach, which outperformed other machine learning algorithms in terms of CHF prediction, reducing patient expenses by reducing the number of medical tests. The proposed model obtained 97.03% F1-score, 95.30% accuracy, 96.49% sensitivity, and 97.58% precision.

Computational identification and experimental verification of a novel signature based on SARS-CoV-2-related genes for predicting prognosis, immune microenvironment and therapeutic strategies in lung adenocarcinoma patients.

Early research indicates that cancer patients are more vulnerable to adverse outcomes and mortality when infected with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Nonetheless, the specific attributes of SARS-CoV-2 in lung Adenocarcinoma (LUAD) have not been extensively and methodically examined. We acquired 322 SARS-CoV-2 infection-related genes (CRGs) from the Human Protein Atlas database. Using an integrative machine learning approach with 10 algorithms, we developed a SARS-CoV-2 score (Cov-2S) signature across The Cancer Genome Atlas and datasets GSE72094, GSE68465, and GSE31210. Comprehensive multi-omics analysis, including assessments of genetic mutations and copy number variations, was conducted to deepen our understanding of the prognosis signature. We also analyzed the response of different Cov-2S subgroups to immunotherapy and identified targeted drugs for these subgroups, advancing personalized medicine strategies. The expression of Cov-2S genes was confirmed through qRT-PCR, with GGH emerging as a critical gene for further functional studies to elucidate its role in LUAD. Out of 34 differentially expressed CRGs identified, 16 correlated with overall survival. We utilized 10 machine learning algorithms, creating 101 combinations, and selected the RFS as the optimal algorithm for constructing a Cov-2S based on the average C-index across four cohorts. This was achieved after integrating several essential clinicopathological features and 58 established signatures. We observed significant differences in biological functions and immune cell statuses within the tumor microenvironments of high and low Cov-2S groups. Notably, patients with a lower Cov-2S showed enhanced sensitivity to immunotherapy. We also identified five potential drugs targeting Cov-2S. In vitro experiments revealed a significant upregulation of GGH in LUAD, and its knockdown markedly inhibited tumor cell proliferation, migration, and invasion. Our research has pioneered the development of a consensus Cov-2S signature by employing an innovative approach with 10 machine learning algorithms for LUAD. Cov-2S reliably forecasts the prognosis, mirrors the tumor's local immune condition, and supports clinical decision-making in tumor therapies.

An Integrated Machine Learning Framework Identifies Prognostic Gene Pair Biomarkers Associated with Programmed Cell Death Modalities in Clear Cell Renal Cell Carcinoma.

Clear cell renal cell carcinoma (ccRCC) is a common and lethal urological malignancy for which there are no effective personalized therapeutic strategies. Programmed cell death (PCD) patterns have emerged as critical determinants of clinical prognosis and immunotherapy responses. However, the actual clinical relevance of PCD processes in ccRCC is still poorly understood. We screened for PCD-related gene pairs through single-sample gene set enrichment analysis (ssGSEA), consensus cluster analysis, and univariate Cox regression analysis. A novel machine learning framework incorporating 12 algorithms and 113 unique combinations were used to develop the cell death-related gene pair score (CDRGPS). Additionally, a radiomic score (Rad_Score) derived from computed tomography (CT) image features was used to classify the CDRGPS status as high or low. Finally, we conclusively verified the function of PRSS23 in ccRCC. The CDRGPS was developed through an integrated machine learning approach that leveraged 113 algorithm combinations. CDRGPS represents an independent prognostic biomarker for overall survival and demonstrated consistent performance between training and external validation cohorts. Moreover, CDRGPS showed better prognostic accuracy compared to seven previously published cell death-related signatures. In addition, patients classified as high-risk by CDRGPS exhibited increased responsiveness to tyrosine kinase inhibitors (TKIs), mammalian Target of Rapamycin (mTOR) inhibitors, and immunotherapy. The Rad_Score demonstrated excellent discrimination for predicting high versus low CDRGPS status, with an area under the curve (AUC) value of 0.813 in the Cancer Imaging Archive (TCIA) database. PRSS23 was identified as a significant factor in the metastasis and immune response of ccRCC, thereby validating experimental in vitro results. CDRGPS is a robust and non-invasive tool that has the potential to improve clinical outcomes and enable personalized medicine in ccRCC patients.

Integrated Machine Learning Approach for the Early Prediction of Pressure Ulcers in Spinal Cord Injury Patients.

(1) Background: Pressure ulcers (PUs) substantially impact the quality of life of spinal cord injury (SCI) patients and require prompt intervention. This study used machine learning (ML) techniques to develop advanced predictive models for the occurrence of PUs in patients with SCI. (2) Methods: By analyzing the medical records of 539 patients with SCI, we observed a 35% incidence of PUs during hospitalization. Our analysis included 139 variables, including baseline characteristics, neurological status (International Standards for Neurological Classification of Spinal Cord Injury [ISNCSCI]), functional ability (Korean version of the Modified Barthel Index [K-MBI] and Functional Independence Measure [FIM]), and laboratory data. We used a variety of ML methods-a graph neural network (GNN), a deep neural network (DNN), a linear support vector machine (SVM_linear), a support vector machine with radial basis function kernel (SVM_RBF), K-nearest neighbors (KNN), a random forest (RF), and logistic regression (LR)-focusing on an integrative analysis of laboratory, neurological, and functional data. (3) Results: The SVM_linear algorithm using these composite data showed superior predictive ability (area under the receiver operating characteristic curve (AUC) = 0.904, accuracy = 0.944), as demonstrated by a 5-fold cross-validation. The critical discriminators of PU development were identified based on limb functional status and laboratory markers of inflammation. External validation highlighted the challenges of model generalization and provided a direction for future research. (4) Conclusions: Our study highlights the importance of a comprehensive, multidimensional data approach for the effective prediction of PUs in patients with SCI, especially in the acute and subacute phases. The proposed ML models show potential for the early detection and prevention of PUs, thus contributing substantially to improving patient care in clinical settings.

Refining breast cancer biomarker discovery and drug targeting through an advanced data-driven approach

Breast cancer remains a major public health challenge worldwide. The identification of accurate biomarkers is critical for the early detection and effective treatment of breast cancer. This study utilizes an integrative machine learning approach to analyze breast cancer gene expression data for superior biomarker and drug target discovery. Gene expression datasets, obtained from the GEO database, were merged post-preprocessing. From the merged dataset, differential expression analysis between breast cancer and normal samples revealed 164 differentially expressed genes. Meanwhile, a separate gene expression dataset revealed 350 differentially expressed genes. Additionally, the BGWO_SA_Ens algorithm, integrating binary grey wolf optimization and simulated annealing with an ensemble classifier, was employed on gene expression datasets to identify predictive genes including TOP2A, AKR1C3, EZH2, MMP1, EDNRB, S100B, and SPP1. From over 10,000 genes, BGWO_SA_Ens identified 1404 in the merged dataset (F1 score: 0.981, PR-AUC: 0.998, ROC-AUC: 0.995) and 1710 in the GSE45827 dataset (F1 score: 0.965, PR-AUC: 0.986, ROC-AUC: 0.972). The intersection of DEGs and BGWO_SA_Ens selected genes revealed 35 superior genes that were consistently significant across methods. Enrichment analyses uncovered the involvement of these superior genes in key pathways such as AMPK, Adipocytokine, and PPAR signaling. Protein-protein interaction network analysis highlighted subnetworks and central nodes. Finally, a drug-gene interaction investigation revealed connections between superior genes and anticancer drugs. Collectively, the machine learning workflow identified a robust gene signature for breast cancer, illuminated their biological roles, interactions and therapeutic associations, and underscored the potential of computational approaches in biomarker discovery and precision oncology.

Machine Learning in the Parkinson’s disease smartwatch (PADS) dataset

The utilisation of smart devices, such as smartwatches and smartphones, in the field of movement disorders research has gained significant attention. However, the absence of a comprehensive dataset with movement data and clinical annotations, encompassing a wide range of movement disorders including Parkinson’s disease (PD) and its differential diagnoses (DD), presents a significant gap. The availability of such a dataset is crucial for the development of reliable machine learning (ML) models on smart devices, enabling the detection of diseases and monitoring of treatment efficacy in a home-based setting. We conducted a three-year cross-sectional study at a large tertiary care hospital. A multi-modal smartphone app integrated electronic questionnaires and smartwatch measures during an interactive assessment designed by neurologists to provoke subtle changes in movement pathologies. We captured over 5000 clinical assessment steps from 504 participants, including PD, DD, and healthy controls (HC). After age-matching, an integrative ML approach combining classical signal processing and advanced deep learning techniques was implemented and cross-validated. The models achieved an average balanced accuracy of 91.16% in the classification PD vs. HC, while PD vs. DD scored 72.42%. The numbers suggest promising performance while distinguishing similar disorders remains challenging. The extensive annotations, including details on demographics, medical history, symptoms, and movement steps, provide a comprehensive database to ML techniques and encourage further investigations into phenotypical biomarkers related to movement disorders.

PerSEveML: a web-based tool to identify persistent biomarker structure for rare events using an integrative machine learning approach.

Omics data sets often pose a computational challenge due to their high dimensionality, large size, and non-linear structures. Analyzing these data sets becomes especially daunting in the presence of rare events. Machine learning (ML) methods have gained traction for analyzing rare events, yet there has been limited exploration of bioinformatics tools that integrate ML techniques to comprehend the underlying biology. Expanding upon our previously developed computational framework of an integrative machine learning approach, we introduce PerSEveML, an interactive web-based tool that uses crowd-sourced intelligence to predict rare events and determine feature selection structures. PerSEveML provides a comprehensive overview of the integrative approach through evaluation metrics that help users understand the contribution of individual ML methods to the prediction process. Additionally, PerSEveML calculates entropy and rank scores, which visually organize input features into a persistent structure of selected, unselected, and fluctuating categories that help researchers uncover meaningful hypotheses regarding the underlying biology. We have evaluated PerSEveML on three diverse biologically complex data sets with extremely rare events from small to large scale and have demonstrated its ability to generate valid hypotheses. PerSEveML is available at https://biostats-shinyr.kumc.edu/PerSEveML/ and https://github.com/sreejatadutta/PerSEveML.

Rapid wildfire damage estimation using integrated object-based classification with auto-generated training samples from Sentinel-2 imagery on Google Earth Engine

Satellite data are essential during wildfires for understanding its adverse effects and improving the effectiveness of rapid disaster management. However, existing techniques used for damage assessments are inaccurate and lack automation. In this study, we propose an integrated machine learning approach with auto-generated training samples for a rapid wildfire disaster response framework using Sentinel-2 imagery at 10 m resolution from Google Earth Engine (GEE). First, training samples of burned areas were obtained by utilizing textural data based on features that had changed because of the wildfire, and samples of unburned areas were obtained using the normalized difference vegetation index (NDVI). The images were categorized as burned and unburned images using the object-based image analysis (OBIA) classification method. Finally, using the classified maps, burn severity maps and estimated pixel counts for each severity class were generated and compared. The proposed method was implemented to put out a wildfire that broke out in Uljin, Gyeongsangbuk-do, South Korea in March 2022 and the transferability of the model was evaluated in Gangneung, Gangwon-do, South Korea. The study findings indicate that the random forest (RF) classifier acquired the greatest overall accuracy (OA) of 97.6 % in Uljin; additionally, the model transferability performed well in Gangneung with an OA of 93.8 %. The RF also generated the fewest pixels of the unchanged class when the burn severity map was evaluated. Overall, our study proposes a quick and automated approach for estimating wildfire damage that could be used for immediate mitigation actions.

Identification of Plausible Candidates in Prostate Cancer Using Integrated Machine Learning Approaches.

Currently, prostate-specific antigen (PSA) is commonly used as a prostate cancer (PCa) biomarker. PSA is linked to some factors that frequently lead to erroneous positive results or even needless biopsies of elderly people. In this pilot study, we undermined the potential genes and mutations from several databases and checked whether or not any putative prognostic biomarkers are central to the annotation. The aim of the study was to develop a risk prediction model that could help in clinical decision-making. An extensive literature review was conducted, and clinical parameters for related comorbidities, such as diabetes, obesity, as well as PCa, were collected. Such parameters were chosen with the understanding that variations in their threshold values could hasten the complicated process of carcinogenesis, more particularly PCa. The gathered data was converted to semi-binary data (-1, -0.5, 0, 0.5, and 1), on which machine learning (ML) methods were applied. First, we cross-checked various publicly available datasets, some published RNA-seq datasets, and our whole-exome sequencing data to find common role players in PCa, diabetes, and obesity. To narrow down their common interacting partners, interactome networks were analysed using GeneMANIA and visualised using Cytoscape, and later cBioportal was used (to compare expression level based on Z scored values) wherein various types of mutation w.r.t their expression and mRNA expression (RNA seq FPKM) plots are available. The GEPIA 2 tool was used to compare the expression of resulting similarities between the normal tissue and TCGA databases of PCa. Later, top-ranking genes were chosen to demonstrate striking clustering coefficients using the Cytoscape-cytoHubba module, and GEPIA 2 was applied again to ascertain survival plots. Comparing various publicly available datasets, it was found that BLM is a frequent player in all three diseases, whereas comparing publicly available datasets, GWAS datasets, and published sequencing findings, SPFTPC and PPIMB were found to be the most common. With the assistance of GeneMANIA, TMPO and FOXP1 were found as common interacting partners, and they were also seen participating with BLM. A probabilistic machine learning model was achieved to identify key candidates between diabetes, obesity, and PCa. This, we believe, would herald precision scale modeling for easy prognosis.

Identifying dynamical persistent biomarker structures for rare events using modern integrative machine learning approach.

The evolution of omics and computational competency has accelerated discoveries of the underlying biological processes in an unprecedented way. High throughput methodologies, such as flow cytometry, can reveal deeper insights into cell processes, thereby allowing opportunities for scientific discoveries related to health and diseases. However, working with cytometry data often imposes complex computational challenges due to high-dimensionality, large size, and nonlinearity of the data structure. In addition, cytometry data frequently exhibit diverse patterns across biomarkers and suffer from substantial class imbalances which can further complicate the problem. The existing methods of cytometry data analysis either predict cell population or perform feature selection. Through this study, we propose a "wisdom of the crowd" approach to simultaneously predict rare cell populations and perform feature selection by integrating a pool of modern machine learning (ML) algorithms. Given that our approach integrates superior performing ML models across different normalization techniques based on entropy and rank, our method can detect diverse patterns existing across the model features. Furthermore, the method identifies a dynamic biomarker structure that divides the features into persistently selected, unselected, and fluctuating assemblies indicating the role of each biomarker in rare cell prediction, which can subsequently aid in studies of disease progression.