Predicting the efficiency of arsenic immobilization in soils by biochar using machine learning.

Arsenic (As) pollution in soils is a pervasive environmental issue. Biochar immobilization offers a promising solution for addressing soil As contamination. The efficiency of biochar in immobilizing As in soils primarily hinges on the characteristics of both the soil and the biochar. However, the influence of a specific property on As immobilization varies among different studies, and the development and application of arsenic passivation materials based on biochar often rely on empirical knowledge. To enhance immobilization efficiency and reduce labor and time costs, a machine learning (ML) model was employed to predict As immobilization efficiency before biochar application. In this study, we collected a dataset comprising 182 data points on As immobilization efficiency from 17 publications to construct three ML models. The results demonstrated that the random forest (RF) model outperformed gradient boost regression tree and support vector regression models in predictive performance. Relative importance analysis and partial dependence plots based on the RF model were conducted to identify the most crucial factors influencing As immobilization. These findings highlighted the significant roles of biochar application time and biochar pH in As immobilization efficiency in soils. Furthermore, the study revealed that Fe-modified biochar exhibited a substantial improvement in As immobilization. These insights can facilitate targeted biochar property design and optimization of biochar application conditions to enhance As immobilization efficiency.

Rapid detection of colored and colorless macro- and micro-plastics in complex environment via near-infrared spectroscopy and machine learning.

To better understand the migration behavior of plastic fragments in the environment, development of rapid non-destructive methods for in-situ identification and characterization of plastic fragments is necessary. However, most of the studies had focused only on colored plastic fragments, ignoring colorless plastic fragments and the effects of different environmental media (backgrounds), thus underestimating their abundance. To address this issue, the present study used near-infrared spectroscopy to compare the identification of colored and colorless plastic fragments based on partial least squares-discriminant analysis (PLS-DA), extreme gradient boost, support vector machine and random forest classifier. The effects of polymer color, type, thickness, and background on the plastic fragments classification were evaluated. PLS-DA presented the best and most stable outcome, with higher robustness and lower misclassification rate. All models frequently misinterpreted colorless plastic fragments and its background when the fragment thickness was less than 0.1mm. A two-stage modeling method, which first distinguishes the plastic types and then identifies colorless plastic fragments that had been misclassified as background, was proposed. The method presented an accuracy higher than 99% in different backgrounds. In summary, this study developed a novel method for rapid and synchronous identification of colored and colorless plastic fragments under complex environmental backgrounds.

Neurostructural subgroup in 4291 individuals with schizophrenia identified using the subtype and stage inference algorithm.

Machine learning can be used to define subtypes of psychiatric conditions based on shared biological foundations of mental disorders. Here we analyzed cross-sectional brain images from 4,222 individuals with schizophrenia and 7038 healthy subjects pooled across 41 international cohorts from the ENIGMA, non-ENIGMA cohorts and public datasets. Using the Subtype and Stage Inference (SuStaIn) algorithm, we identify two distinct neurostructural subgroups by mapping the spatial and temporal 'trajectory' of gray matter change in schizophrenia. Subgroup 1 was characterized by an early cortical-predominant loss with enlarged striatum, whereas subgroup 2 displayed an early subcortical-predominant loss in the hippocampus, striatum and other subcortical regions. We confirmed the reproducibility of the two neurostructural subtypes across various sample sites, including Europe, North America and East Asia. This imaging-based taxonomy holds the potential to identify individuals with shared neurobiological attributes, thereby suggesting the viability of redefining existing disorder constructs based on biological factors.

Affordable and real-time antimicrobial resistance prediction from multimodal electronic health records.

The spread of antimicrobial resistance (AMR) leads to challenging complications and losses of human lives plus medical resources, with a high expectancy of deterioration in the future if the problem is not controlled. From a machine learning perspective, data-driven models could aid clinicians and microbiologists by anticipating the resistance beforehand. Our study serves as the first attempt to harness deep learning (DL) techniques and the multimodal data available in electronic health records (EHR) for predicting AMR. In this work, we utilize and preprocess the MIMIC-IV database extensively to produce separate structured input sources for time-invariant and time-series data customized to the AMR task. Then, a multimodality fusion approach merges the two modalities with clinical notes to determine resistance based on an antibiotic or a pathogen. To efficiently predict AMR, our approach builds the foundation for deploying multimodal DL techniques in clinical practice, leveraging the existing patient data.

At-admission prediction of mortality and pulmonary embolism in an international cohort of hospitalised patients with COVID-19 using statistical and machine learning methods.

By September 2022, more than 600 million cases of SARS-CoV-2 infection have been reported globally, resulting in over 6.5 million deaths. COVID-19 mortality risk estimators are often, however, developed with small unrepresentative samples and with methodological limitations. It is highly important to develop predictive tools for pulmonary embolism (PE) in COVID-19 patients as one of the most severe preventable complications of COVID-19. Early recognition can help provide life-saving targeted anti-coagulation therapy right at admission. Using a dataset of more than 800,000 COVID-19 patients from an international cohort, we propose a cost-sensitive gradient-boosted machine learning model that predicts occurrence of PE and death at admission. Logistic regression, Cox proportional hazards models, and Shapley values were used to identify key predictors for PE and death. Our prediction model had a test AUROC of 75.9% and 74.2%, and sensitivities of 67.5% and 72.7% for PE and all-cause mortality respectively on a highly diverse and held-out test set. The PE prediction model was also evaluated on patients in UK and Spain separately with test results of 74.5% AUROC, 63.5% sensitivity and 78.9% AUROC, 95.7% sensitivity. Age, sex, region of admission, comorbidities (chronic cardiac and pulmonary disease, dementia, diabetes, hypertension, cancer, obesity, smoking), and symptoms (any, confusion, chest pain, fatigue, headache, fever, muscle or joint pain, shortness of breath) were the most important clinical predictors at admission. Age, overall presence of symptoms, shortness of breath, and hypertension were found to be key predictors for PE using our extreme gradient boosted model. This analysis based on the, until now, largest global dataset for this set of problems can inform hospital prioritisation policy and guide long term clinical research and decision-making for COVID-19 patients globally. Our machine learning model developed from an international cohort can serve to better regulate hospital risk prioritisation of at-risk patients.

Construction of a clinical prediction model for complicated appendicitis based on machine learning techniques.

Acute appendicitis is a typical surgical emergency worldwide and one of the common causes of surgical acute abdomen in the elderly. Accurately diagnosing and differentiating acute appendicitis can assist clinicians in formulating a scientific and reasonable treatment plan and providing high-quality medical services for the elderly. In this study, we validated and analyzed the different performances of various machine learning models based on the analysis of clinical data, so as to construct a simple, fast, and accurate estimation method for the diagnosis of early acute appendicitis. The dataset of this paper was obtained from the medical data of elderly patients with acute appendicitis attending the First Affiliated Hospital of Anhui University of Chinese Medicine from January 2012 to January 2022, including 196 males (60.87%) and 126 females (39.13%), including 103 (31.99%) patients with complicated appendicitis and 219 (68.01%) patients with uncomplicated appendicitis. By comparing and analyzing the prediction results of the models implemented by nine different machine learning techniques (LR, CART, RF, SVM, Bayes, KNN, NN, FDA, and GBM), we found that the GBM algorithm gave the optimal results and that sensitivity, specificity, PPV, NPV, precision, recall, F1 and brier are 0.9167, 0.9739, 0.9429, 0.9613, 0.9429, 0.9167, 0.9296, and 0.05649, respectively. The GBM model prediction results are interpreted using the SHAP technology framework. Calibration and Decision curve analysis also show that the machine learning model proposed in this paper has some clinical and economic benefits. Finally, we developed the Shiny application for complicated appendicitis diagnosis to assist clinicians in quickly and effectively recognizing patients with complicated appendicitis (CA) and uncomplicated appendicitis (UA), and to formulate a more reasonable and scientific clinical plan for acute appendicitis patient population promptly.

Leveraging AI and Machine Learning in Six-Sigma Documentation for Pharmaceutical Quality Assurance.

The pharmaceutical industry must maintain stringent quality assurance standards to ensure product safety and regulatory compliance. A key component of the well-known Six Sigma methodology for process improvement and quality control is precise and comprehensive documentation. However, there are a number of significant issues with traditional documentation procedures, including as slowness, human error, and difficulties with regulatory standards. This review research looks at innovative ways to employ machine learning (ML) and artificial intelligence (AI) to enhance Six Sigma documentation processes in the pharmaceutical sector. AI and ML provide cutting-edge technologies that have the potential to drastically alter documentation processes by automating data entry, collection, and analysis. Natural language processing (NLP) and computer vision technologies have the potential to significantly reduce human error rates and increase the efficacy of documentation processes. By applying machine learning algorithms to support real-time data analysis, predictive analytics, and proactive quality management, pharmaceutical organizations may be able to identify potential quality issues early on and take proactive efforts to address them. Combining AI and ML improves documentation accuracy and reliability while also strengthening compliance with stringent regulatory criteria. The primary barriers and limitations to the current state of Six Sigma documentation in the pharmaceutical industry are identified in this study. It examines the fundamentals of AI and ML with an emphasis on their specific applications in quality assurance and potential benefits for Six Sigma processes. The report includes extensive case studies that highlight notable developments and explain how AI/ML enhanced documentation is used in the real world.

Gut microbiota-based machine-learning signature for the diagnosis of alcohol-associated and metabolic dysfunction-associated steatotic liver disease.

Alcoholic-associated liver disease (ALD) and metabolic dysfunction-associated steatotic liver disease (MASLD) show a high prevalence rate worldwide. As gut microbiota represents current state of ALD and MASLD via gut-liver axis, typical characteristics of gut microbiota can be used as a potential diagnostic marker in ALD and MASLD. Machine learning (ML) algorithms improve diagnostic performance in various diseases. Using gut microbiota-based ML algorithms, we evaluated the diagnostic index for ALD and MASLD. Fecal 16S rRNA sequencing data of 263 ALD (control, elevated liver enzyme [ELE], cirrhosis, and hepatocellular carcinoma [HCC]) and 201 MASLD (control and ELE) subjects were collected. For external validation, 126 ALD and 84 MASLD subjects were recruited. Four supervised ML algorithms (support vector machine, random forest, multilevel perceptron, and convolutional neural network) were used for classification with 20, 40, 60, and 80 features, in which three nonsupervised ML algorithms (independent component analysis, principal component analysis, linear discriminant analysis, and random projection) were used for feature reduction. A total of 52 combinations of ML algorithms for each pair of subgroups were performed with 60 hyperparameter variations and Stratified ShuffleSplit tenfold cross validation. The ML models of the convolutional neural network combined with principal component analysis achieved areas under the receiver operating characteristic curve (AUCs) > 0.90. In ALD, the diagnostic AUC values of the ML strategy (vs. control) were 0.94, 0.97, and 0.96 for ELE, cirrhosis, and liver cancer, respectively. The AUC value (vs. control) for MASLD (ELE) was 0.93. In the external validation, the AUC values of ALD and MASLD (vs control) were > 0.90 and 0.88, respectively. The gut microbiota-based ML strategy can be used for the diagnosis of ALD and MASLD.ClinicalTrials.gov NCT04339725.

Decoding depression: a comprehensive multi-cohort exploration of blood DNA methylation using machine learning and deep learning approaches.

The causes of depression are complex, and the current diagnosis methods rely solely on psychiatric evaluations with no incorporation of laboratory biomarkers in clinical practices. We investigated the stability of blood DNA methylation depression signatures in six different populations using six public and two domestic cohorts (n = 1942) conducting mega-analysis and meta-analysis of the individual studies. We evaluated 12 machine learning and deep learning strategies for depression classification both in cross-validation (CV) and in hold-out tests using merged data from 8 separate batches, constructing models with both biased and unbiased feature selection. We found 1987 CpG sites related to depression in both mega- and meta-analysis at the nominal level, and the associated genes were nominally related to axon guidance and immune pathways based on enrichment analysis and eQTM data. Random forest classifiers achieved the highest performance (AUC 0.73 and 0.76) in CV and hold-out tests respectively on the batch-level processed data. In contrast, the methylation showed low predictive power (all AUCs < 0.57) for all classifiers in CV and no predictive power in hold-out tests when used with harmonized data. All models achieved significantly better performance (>14% gain in AUCs) with pre-selected features (selection bias), with some of the models (joint autoencoder-classifier) reaching AUCs of up to 0.91 in the final testing regardless of data preparation. Different algorithmic feature selection approaches may outperform limma, however, random forest models perform well regardless of the strategy. The results provide an overview over potential future biomarkers for depression and highlight many important methodological aspects for DNA methylation-based depression profiling including the use of machine learning strategies.

Deciphering the tumour microenvironment of clear cell renal cell carcinoma: Prognostic insights from programmed death genes using machine learning.

Clear cell renal cell carcinoma (ccRCC), a prevalent kidney cancer form characterised by its invasiveness and heterogeneity, presents challenges in late-stage prognosis and treatment outcomes. Programmed cell death mechanisms, crucial in eliminating cancer cells, offer substantial insights into malignant tumour diagnosis, treatment and prognosis. This study aims to provide a model based on 15 types of Programmed Cell Death-Related Genes (PCDRGs) for evaluating immune microenvironment and prognosis in ccRCC patients. ccRCC patients from the TCGA and arrayexpress cohorts were grouped based on PCDRGs. A combination model using Lasso and SuperPC was constructed to identify prognostic gene features. The arrayexpress cohort validated the model, confirming its robustness. Immune microenvironment analysis, facilitated by PCDRGs, employed various methods, including CIBERSORT. Drug sensitivity analysis guided clinical treatment decisions. Single-cell data enabled Programmed Cell Death-Related scoring, subsequent pseudo-temporal and cell-cell communication analyses. A PCDRGs signature was established using TCGA-KIRC data. External validation in the arrayexpress cohort underscored the model's superiority over traditional clinical features. Furthermore, our single-cell analysis unveiled the roles of PCDRG-based single-cell subgroups in ccRCC, both in pseudo-temporal progression and intercellular communication. Finally, we performed CCK-8 assay and other experiments to investigate csf2. In conclusion, these findings reveal that csf2 inhibit the growth, infiltration and movement of cells associated with renal clear cell carcinoma. This study introduces a PCDRGs prognostic model benefiting ccRCC patients while shedding light on the pivotal role of programmed cell death genes in shaping the immune microenvironment of ccRCC patients.

Explainable machine learning on baseline MRI predicts multiple sclerosis trajectory descriptors.

Multiple sclerosis (MS) is a multifaceted neurological condition characterized by challenges in timely diagnosis and personalized patient management. The application of Artificial Intelligence (AI) to MS holds promises for early detection, accurate diagnosis, and predictive modeling. The objectives of this study are: 1) to propose new MS trajectory descriptors that could be employed in Machine Learning (ML) regressors and classifiers to predict patient evolution; 2) to explore the contribution of ML models in discerning MS trajectory descriptors using only baseline Magnetic Resonance Imaging (MRI) studies. This study involved 446 MS patients who had a baseline MRI, at least two measurements of Expanded Disability Status Scale (EDSS), and a 1-year follow-up. Patients were divided into two groups: 1) for model development and 2) for evaluation. Three descriptors: ß1, ß2, and EDSS(t), were related to baseline MRI parameters using regression and classification XGBoost models. Shapley Additive Explanations (SHAP) analysis enhanced model transparency by identifying influential features. The results of this study demonstrate the potential of AI in predicting MS progression using the proposed patient trajectories and baseline MRI scans, outperforming classic Multiple Linear Regression (MLR) methods. In conclusion, MS trajectory descriptors are crucial; incorporating AI analysis into MRI assessments presents promising opportunities to advance predictive capabilities. SHAP analysis enhances model interpretation, revealing feature importance for clinical decisions.

An improved mountain gazelle optimizer based on chaotic map and spiral disturbance for medical feature selection.

Feature selection is an important solution for dealing with high-dimensional data in the fields of machine learning and data mining. In this paper, we present an improved mountain gazelle optimizer (IMGO) based on the newly proposed mountain gazelle optimizer (MGO) and design a binary version of IMGO (BIMGO) to solve the feature selection problem for medical data. First, the gazelle population is initialized using iterative chaotic map with infinite collapses (ICMIC) mapping, which increases the diversity of the population. Second, a nonlinear control factor is introduced to balance the exploration and exploitation components of the algorithm. Individuals in the population are perturbed using a spiral perturbation mechanism to enhance the local search capability of the algorithm. Finally, a neighborhood search strategy is used for the optimal individuals to enhance the exploitation and convergence capabilities of the algorithm. The superior ability of the IMGO algorithm to solve continuous problems is demonstrated on 23 benchmark datasets. Then, BIMGO is evaluated on 16 medical datasets of different dimensions and compared with 8 well-known metaheuristic algorithms. The experimental results indicate that BIMGO outperforms the competing algorithms in terms of the fitness value, number of selected features and sensitivity. In addition, the statistical results of the experiments demonstrate the significantly superior ability of BIMGO to select the most effective features in medical datasets.

Derivation and validation of the first web-based nomogram to predict the spontaneous pregnancy after reproductive surgery using machine learning models.

Objective: Infertility remains a significant global burden over the years. Reproductive surgery is an effective strategy for infertile women. Early prediction of spontaneous pregnancy after reproductive surgery is of high interest for the patients seeking the infertility treatment. However, there are no high-quality models and clinical applicable tools to predict the probability of natural conception after reproductive surgery. Methods: The eligible data involving 1013 patients who operated for infertility between June 2016 and June 2021 in Yantai Yuhuangding Hospital in China, were randomly divided into training and internal testing cohorts. 195 subjects from the Linyi People's Hospital in China were considered for external validation. Both univariate combining with multivariate logistic regression and the least absolute shrinkage and selection operator (LASSO) algorithm were performed to identify independent predictors. Multiple common machine learning algorithms, namely logistic regression, decision tree, random forest, support vector machine, k-nearest neighbor, and extreme gradient boosting, were employed to construct the predictive models. The optimal model was verified by evaluating the model performance in both the internal and external validation datasets. Results: Six clinical indicators, including female age, infertility type, duration of infertility, intraoperative diagnosis, ovulation monitoring, and anti-Müllerian hormone (AMH) level, were screened out. Based on the logistic regression model's superior clinical predictive value, as indicated by the area under the receiver operating characteristic curve (AUC) in both the internal (0.870) and external (0.880) validation sets, we ultimately selected it as the optimal model. Consequently, we utilized it to generate a web-based nomogram for predicting the probability of spontaneous pregnancy after reproductive surgery. Furthermore, the calibration curve, Hosmer-Lemeshow (H-L) test, the decision curve analysis (DCA) and clinical impact curve analysis (CIC) demonstrated that the model has superior calibration degree, clinical net benefit and generalization ability, which were confirmed by both internal and external validations. Conclusion: Overall, our developed first nomogram with online operation provides an early and accurate prediction for the probability of natural conception after reproductive surgery, which helps clinicians and infertile couples make sensible decision of choosing the mode of subsequent conception, natural or IVF, to further improve the clinical practices of infertility treatment.

A novel prognostic signature related to programmed cell death in osteosarcoma.

Background: Osteosarcoma primarily affects children and adolescents, with current clinical treatments often resulting in poor prognosis. There has been growing evidence linking programmed cell death (PCD) to the occurrence and progression of tumors. This study aims to enhance the accuracy of OS prognosis assessment by identifying PCD-related prognostic risk genes, constructing a PCD-based OS prognostic risk model, and characterizing the function of genes within this model. Method: We retrieved osteosarcoma patient samples from TARGET and GEO databases, and manually curated literature to summarize 15 forms of programmed cell death. We collated 1621 PCD genes from literature sources as well as databases such as KEGG and GSEA. To construct our model, we integrated ten machine learning methods including Enet, Ridge, RSF, CoxBoost, plsRcox, survivalSVM, Lasso, SuperPC, StepCox, and GBM. The optimal model was chosen based on the average C-index, and named Osteosarcoma Programmed Cell Death Score (OS-PCDS). To validate the predictive performance of our model across different datasets, we employed three independent GEO validation sets. Moreover, we assessed mRNA and protein expression levels of the genes included in our model, and investigated their impact on proliferation, migration, and apoptosis of osteosarcoma cells by gene knockdown experiments. Result: In our extensive analysis, we identified 30 prognostic risk genes associated with programmed cell death (PCD) in osteosarcoma (OS). To assess the predictive power of these genes, we computed the C-index for various combinations. The model that employed the random survival forest (RSF) algorithm demonstrated superior predictive performance, significantly outperforming traditional approaches. This optimal model included five key genes: MTM1, MLH1, CLTCL1, EDIL3, and SQLE. To validate the relevance of these genes, we analyzed their mRNA and protein expression levels, revealing significant disparities between osteosarcoma cells and normal tissue cells. Specifically, the expression levels of these genes were markedly altered in OS cells, suggesting their critical role in tumor progression. Further functional validation was performed through gene knockdown experiments in U2OS cells. Knockdown of three of these genes-CLTCL1, EDIL3, and SQLE-resulted in substantial changes in proliferation rate, migration capacity, and apoptosis rate of osteosarcoma cells. These findings underscore the pivotal roles of these genes in the pathophysiology of osteosarcoma and highlight their potential as therapeutic targets. Conclusion: The five genes constituting the OS-PCDS model-CLTCL1, MTM1, MLH1, EDIL3, and SQLE-were found to significantly impact the proliferation, migration, and apoptosis of osteosarcoma cells, highlighting their potential as key prognostic markers and therapeutic targets. OS-PCDS enables accurate evaluation of the prognosis in patients with osteosarcoma.

Glycosylphosphatidylinositol anchor biosynthesis pathway-based biomarker identification with machine learning for prognosis and T cell exhaustion status prediction in breast cancer.

As the primary component of anti-tumor immunity, T cells are prone to exhaustion and dysfunction in the tumor microenvironment (TME). A thorough understanding of T cell exhaustion (TEX) in the TME is crucial for effectively addressing TEX in clinical settings and promoting the efficacy of immune checkpoint blockade therapies. In eukaryotes, numerous cell surface proteins are tethered to the plasma membrane via Glycosylphosphatidylinositol (GPI) anchors, which play a crucial role in facilitating the proper translocation of membrane proteins. However, the available evidence is insufficient to support any additional functional involvement of GPI anchors. Here, we investigate the signature of GPI-anchor biosynthesis in the TME of breast cancer (BC)patients, particularly its correlation with TEX. GPI-anchor biosynthesis should be considered as a prognostic risk factor for BC. Patients with high GPI-anchor biosynthesis showed more severe TEX. And the levels of GPI-anchor biosynthesis in exhausted CD8 T cells was higher than normal CD8 T cells, which was not observed between malignant epithelial cells and normal mammary epithelial cells. In addition, we also found that GPI -anchor biosynthesis related genes can be used to diagnose TEX status and predict prognosis in BC patients, both the TEX diagnostic model and the prognostic model showed good AUC values. Finally, we confirmed our findings in cells and clinical samples. Knockdown of PIGU gene expression significantly reduced the proliferation rate of MDA-MB-231 and MCF-7 cell lines. Immunofluorescence results from clinical samples showed reduced aggregation of CD8 T cells in tissues with high expression of GPAA1 and PIGU.

A Multimorbidity Analysis of Hospitalized Patients With COVID-19 in Northwest Italy: Longitudinal Study Using Evolutionary Machine Learning and Health Administrative Data.

BACKGROUND: Multimorbidity is a significant public health concern, characterized by the coexistence and interaction of multiple preexisting medical conditions. This complex condition has been associated with an increased risk of COVID-19. Individuals with multimorbidity who contract COVID-19 often face a significant reduction in life expectancy. The postpandemic period has also highlighted an increase in frailty, emphasizing the importance of integrating existing multimorbidity details into epidemiological risk assessments. Managing clinical data that include medical histories presents significant challenges, particularly due to the sparsity of data arising from the rarity of multimorbidity conditions. Also, the complex enumeration of combinatorial multimorbidity features introduces challenges associated with combinatorial explosions. OBJECTIVE: This study aims to assess the severity of COVID-19 in individuals with multiple medical conditions, considering their demographic characteristics such as age and sex. We propose an evolutionary machine learning model designed to handle sparsity, analyzing preexisting multimorbidity profiles of patients hospitalized with COVID-19 based on their medical history. Our objective is to identify the optimal set of multimorbidity feature combinations strongly associated with COVID-19 severity. We also apply the Apriori algorithm to these evolutionarily derived predictive feature combinations to identify those with high support. METHODS: We used data from 3 administrative sources in Piedmont, Italy, involving 12,793 individuals aged 45-74 years who tested positive for COVID-19 between February and May 2020. From their 5-year pre-COVID-19 medical histories, we extracted multimorbidity features, including drug prescriptions, disease diagnoses, sex, and age. Focusing on COVID-19 hospitalization, we segmented the data into 4 cohorts based on age and sex. Addressing data imbalance through random resampling, we compared various machine learning algorithms to identify the optimal classification model for our evolutionary approach. Using 5-fold cross-validation, we evaluated each model's performance. Our evolutionary algorithm, utilizing a deep learning classifier, generated prediction-based fitness scores to pinpoint multimorbidity combinations associated with COVID-19 hospitalization risk. Eventually, the Apriori algorithm was applied to identify frequent combinations with high support. RESULTS: We identified multimorbidity predictors associated with COVID-19 hospitalization, indicating more severe COVID-19 outcomes. Frequently occurring morbidity features in the final evolved combinations were age>53, R03BA (glucocorticoid inhalants), and N03AX (other antiepileptics) in cohort 1; A10BA (biguanide or metformin) and N02BE (anilides) in cohort 2; N02AX (other opioids) and M04AA (preparations inhibiting uric acid production) in cohort 3; and G04CA (Alpha-adrenoreceptor antagonists) in cohort 4. CONCLUSIONS: When combined with other multimorbidity features, even less prevalent medical conditions show associations with the outcome. This study provides insights beyond COVID-19, demonstrating how repurposed administrative data can be adapted and contribute to enhanced risk assessment for vulnerable populations.

Integrating machine learning algorithms and multiple immunohistochemistry validation to unveil novel diagnostic markers based on costimulatory molecules for predicting immune microenvironment status in triple-negative breast cancer.

Introduction: Costimulatory molecules are putative novel targets or potential additions to current available immunotherapy, but their expression patterns and clinical value in triple-negative breast cancer (TNBC) are to be clarified. Methods: The gene expression profiles datasets of TNBC patients were obtained from The Cancer Genome Atlas and the Gene Expression Omnibus databases. Diagnostic biomarkers for stratifying individualized tumor immune microenvironment (TIME) were identified using the Least Absolute Shrinkage and Selection Operator (LASSO) and Support Vector Machine-Recursive Feature Elimination (SVM-RFE) algorithms. Additionally, we explored their associations with response to immunotherapy via the multiplex immunohistochemistry (mIHC). Results: A total of 60 costimulatory molecule genes (CMGs) were obtained, and we determined two different TIME subclasses ("hot" and "cold") through the K-means clustering method. The "hot" tumors presented a higher infiltration of activated immune cells, i.e., CD4 memory-activated T cells, resting NK cells, M1 macrophages, and CD8 T cells, thereby enriched in the B cell and T cell receptor signaling pathways. LASSO and SVM-RFE algorithms identified three CMGs (CD86, TNFRSF17 and TNFRSF1B) as diagnostic biomarkers. Following, a novel diagnostic nomogram was constructed for predicting individualized TIME status and was validated with good predictive accuracy in TCGA, GSE76250 and GSE58812 databases. Further mIHC conformed that TNBC patients with high CD86, TNFRSF17 and TNFRSF1B levels tended to respond to immunotherapy. Conclusion: This study supplemented evidence about the value of CMGs in TNBC. In addition, CD86, TNFRSF17 and TNFRSF1B were found as potential biomarkers, significantly promoting TNBC patient selection for immunotherapeutic guidance.

SLO-Net: Enhancing Multiple Sclerosis Diagnosis Beyond Optical Coherence Tomography Using Infrared Reflectance Scanning Laser Ophthalmoscopy Images.

Purpose: Several machine learning studies have used optical coherence tomography (OCT) for multiple sclerosis (MS) classification with promising outcomes. Infrared reflectance scanning laser ophthalmoscopy (IR-SLO) captures high-resolution fundus images, commonly combined with OCT for fixed B-scan positions. However, no machine learning research has utilized IR-SLO images for automated MS diagnosis. Methods: This study utilized a dataset comprised of IR-SLO images and OCT data from Isfahan, Iran, encompassing 32 MS and 70 healthy individuals. A number of convolutional neural networks (CNNs)-namely, VGG-16, VGG-19, ResNet-50, ResNet-101, and a custom architecture-were trained with both IR-SLO images and OCT thickness maps as two separate input datasets. The highest performing models for each modality were then integrated to create a bimodal model that receives the combination of OCT thickness maps and IR-SLO images. Subject-wise data splitting was employed to prevent data leakage among training, validation, and testing sets. Results: Overall, images of the 102 patients from the internal dataset were divided into test, validation, and training subsets. Subsequently, we employed a bootstrapping approach on the training data through iterative sampling with replacement. The performance of the proposed bimodal model was evaluated on the internal test dataset, demonstrating an accuracy of 92.40% ± 4.1% (95% confidence interval [CI], 83.61-98.08), sensitivity of 95.43% ± 5.75% (95% CI, 83.71-100.0), specificity of 92.82% ± 3.72% (95% CI, 81.15-96.77), area under the receiver operating characteristic (AUROC) curve of 96.99% ± 2.99% (95% CI, 86.11-99.78), and area under the precision-recall curve (AUPRC) of 97.27% ± 2.94% (95% CI, 86.83-99.83). Furthermore, to assess the model generalization ability, we examined its performance on an external test dataset following the same bootstrap methodology, achieving promising results, with accuracy of 85.43% ± 0.08% (95% CI, 71.43-100.0), sensitivity of 97.33% ± 0.06% (95% CI, 83.33-100.0), specificity of 84.6% ± 0.10% (95% CI, 71.43-100.0), AUROC curve of 99.67% ± 0.02% (95% CI, 95.63-100.0), and AUPRC of 99.65% ± 0.02% (95% CI, 94.90-100.0). Conclusions: Incorporating both modalities improves the performance of automated diagnosis of MS, showcasing the potential of utilizing IR-SLO as a complementary tool alongside OCT. Translational Relevance: Should the results of our proposed bimodal model be validated in future work with larger and more diverse datasets, diagnosis of MS based on both OCT and IR-SLO can be reliably integrated into routine clinical practice.