Potential value and impact of data mining and machine learning in clinical diagnostics

A novel AI-based diagnostic model for pertussis pneumonia

It is still very difficult to diagnose pertussis based on a doctor's experience. Our aim is to develop a model based on machine learning algorithms combined with biochemical blood tests to diagnose pertussis. A total of 295 patients with pertussis and 295 patients with non-pertussis lower respiratory infections between January 2022 and January 2023, matched for age and gender ratio, were included in our study. Patients underwent a reverse transcription polymerase chain reaction test for pertussis and other viruses. Univariate logistic regression analysis was used to screen for clinical and blood biochemical features associated with pertussis. The optimal features and 3 machine learning algorithms including K-nearest neighbor, support vector machine, and eXtreme Gradient Boosting (XGBoost) were used to develop diagnostic models. Using univariate logistic regression analysis, 18 out of the 27 features were considered optimal features associated with pertussis The XGBoost model was significantly superior to both the support vector machine model (Delong test, P = .01) and the K-nearest neighbor model (Delong test, P = .01), with the area under the receiver operating characteristic curve of 0.96 and an accuracy of 0.923. Our diagnostic model based on blood biochemical test results at admission and XGBoost algorithm can help doctors effectively diagnose pertussis.

Integrating Machine Learning into Statistical Methods in Disease Risk Prediction Modeling: A Systematic Review

Background: Disease prediction models often use statistical methods or machine learning, both with their own corresponding application scenarios, raising the risk of errors when used alone. Integrating machine learning into statistical methods may yield robust prediction models. This systematic review aims to comprehensively assess current development of global disease prediction integration models. Methods: PubMed, EMbase, Web of Science, CNKI, VIP, WanFang, and SinoMed databases were searched to collect studies on prediction models integrating machine learning into statistical methods from database inception to 2023 May 1. Information including basic characteristics of studies, integrating approaches, application scenarios, modeling details, and model performance was extracted. Results: A total of 20 eligible studies in English and 1 in Chinese were included. Five studies concentrated on diagnostic models, while 16 studies concentrated on predicting disease occurrence or prognosis. Integrating strategies of classification models included majority voting, weighted voting, stacking, and model selection (when statistical methods and machine learning disagreed). Regression models adopted strategies including simple statistics, weighted statistics, and stacking. AUROC of integration models surpassed 0.75 and performed better than statistical methods and machine learning in most studies. Stacking was used for situations with >100 predictors and needed relatively larger amount of training data. Conclusion: Research on integrating machine learning into statistical methods in prediction models remains limited, but some studies have exhibited great potential that integration models outperform single models. This study provides insights for the selection of integration methods for different scenarios. Future research could emphasize on the improvement and validation of integrating strategies.

Unlocking Optimal Glycemic Interpretation: Redefining HbA1c Analysis in Female Patients With Diabetes and Iron-Deficiency Anemia Using Machine Learning Algorithms

OBJECTIVE

In response to the nuanced glycemic challenges faced by women with iron deficiency anemia (IDA) associated with diabetes, this study uses advanced machine learning algorithms to redefine hemoglobin (Hb)A1c measurement values. We aimed to improve the accuracy of glycemic interpretation by recognizing the critical interaction between erythrocytes, iron, and glycemic levels in this specific demographic group.

METHODS

This retrospective observational study included 17,526 adult women with HbA1c levels recorded from 2017 to 2022. Samples were classified as diabetic, prediabetic, or non-diabetic based on HbA1c and fasting blood glucose (FBG) levels for distribution analysis without impacting model training. Support Vector Machines, Linear Regression, Random Forest, and K-Nearest Neighbor algorithms as machine learning (ML) methods were used to predict HbA1c levels. Following the training of the model, HbA1c values were predicted for the IDA samples using the trained model.

RESULTS

According to our results, there has been a 0.1 unit change in HbA1c values, which has resulted in a clinical decision change in some patients.

DISCUSSION

Using ML to analyze HbA1c results in women with IDA may unveil distinctions among patients whose HbA1c values hover near critical medical decision thresholds. This intersection of technology and laboratory science holds promise for enhancing precision in medical decision-making processes.

Hepatic inflammation and fibrosis are profiles related to mid-term mortality in biopsy-proven MASLD: A multicenter study in Japan

AIMS

A multi-stakeholder consensus has proposed MASLD (metabolic dysfunction-associated steatotic liver disease). We aimed to investigate the pathological findings related to the mid-term mortality of patients with biopsy-proven MASLD in Japan.

METHODS

We enrolled 1349 patients with biopsy-proven MASLD. The observational period was 8010 person years. We evaluated independent factors associated with mortality in patients with MASLD by Cox regression analysis. We also investigated pathological profiles related to mortality in patients with MASLD using data-mining analysis.

RESULTS

The prevalence of MASH and stage 3/4 fibrosis was observed in 65.6% and 17.4%, respectively. Forty-five patients with MASLD died. Of these, liver-related events were the most common cause at 40% (n = 18), followed by extrahepatic malignancies at 26.7% (n = 12). Grade 2/3 lobular inflammation and stage 3/4 fibrosis had a 1.9-fold and 1.8-fold risk of mortality, respectively. In the decision-tree analysis, the profiles with the worst prognosis were characterised by Grade 2/3 hepatic inflammation, along with advanced ballooning (grade 1/2) and fibrosis (stage 3/4). This profile showed a mortality at 8.3%. Furthermore, the random forest analysis identified that hepatic fibrosis and inflammation were the first and second responsible factors for the mid-term prognosis of patients with MASLD.

CONCLUSIONS

In patients with biopsy-proven MASLD, the prevalence of MASH and advanced fibrosis was approximately 65% and 20%, respectively. The leading cause of mortality was liver-related events. Hepatic inflammation and fibrosis were significant factors influencing mid-term mortality. These findings highlight the importance of targeting inflammation and fibrosis in the management of patients with MASLD.

Novel gene signatures predicting and immune infiltration analysis in Parkinson's disease: based on combining random forest with artificial neural network

BACKGROUND

Parkinson's disease (PD) ranks as the second most prevalent neurodegenerative disorder globally, and its incidence is rapidly rising. The diagnosis of PD relies on clinical characteristics. Although current treatments aim to alleviate symptoms, they do not effectively halt the disease's progression. Early detection and intervention hold immense importance. This study aimed to establish a new PD diagnostic model.

METHODS

Data from a public database were adopted for the construction and validation of a PD diagnostic model with random forest and artificial neural network models. The CIBERSORT platform was applied for the evaluation of immune cell infiltration in PD. Quantitative real-time PCR was performed to verify the accuracy and reliability of the bioinformatics analysis results.

RESULTS

Leveraging existing gene expression data from the Gene Expression Omnibus (GEO) database, we sifted through differentially expressed genes (DEGs) in PD and identified 30 crucial genes through a random forest classifier. Furthermore, we successfully designed a novel PD diagnostic model using an artificial neural network and verified its diagnostic efficacy using publicly available datasets. Our research also suggests that mast cells may play a significant role in the onset and progression of PD.

CONCLUSION

This work developed a new PD diagnostic model with machine learning techniques and suggested the immune cells as a potential target for PD therapy.

Predictive models-assisted diagnosis of AIDS-associated Pneumocystis jirovecii pneumonia in the emergency room, based on clinical, laboratory, and radiological data

We assessed predictive models (PMs) for diagnosing Pneumocystis jirovecii pneumonia (PCP) in AIDS patients seen in the emergency room (ER), aiming to guide empirical treatment decisions. Data from suspected PCP cases among AIDS patients were gathered prospectively at a reference hospital's ER, with diagnoses later confirmed through sputum PCR analysis. We compared clinical, laboratory, and radiological data between PCP and non-PCP groups, using the Boruta algorithm to confirm significant differences. We evaluated ten PMs tailored for various ERs resource levels to diagnose PCP. Four scenarios were created, two based on X-ray findings (diffuse interstitial infiltrate) and two on CT scans ("ground-glass"), incorporating mandatory variables: lactate dehydrogenase, O2sat, C-reactive protein, respiratory rate (> 24 bpm), and dry cough. We also assessed HIV viral load and CD4 cell count. Among the 86 patients in the study, each model considered either 6 or 8 parameters, depending on the scenario. Many models performed well, with accuracy, precision, recall, and AUC scores > 0.8. Notably, nearest neighbor and naïve Bayes excelled (scores > 0.9) in specific scenarios. Surprisingly, HIV viral load and CD4 cell count did not improve model performance. In conclusion, ER-based PMs using readily available data can significantly aid PCP treatment decisions in AIDS patients.

Refining the rheological characteristics of high drug loading ointment via SDS and machine learning

This paper presents an optimized preparation process for external ointment using the Definitive Screening Design (DSD) method. The ointment is a Traditional Chinese Medicine (TCM) formula developed by Professor WYH, a renowned TCM practitioner in Jiangsu Province, China, known for its proven clinical efficacy. In this study, a stepwise regression model was employed to analyze the relationship between key process factors (such as mixing speed and time) and rheological parameters. Machine learning techniques, including Monte Carlo simulation, decision tree analysis, and Gaussian process, were used for parameter optimization. Through rigorous experimentation and verification, we have successfully identified the optimal preparation process for WYH ointment. The optimized parameters included drug ratio of 24.5%, mixing time of 8 min, mixing speed of 1175 rpm, petroleum dosage of 79 g, liquid paraffin dosage of 6.7 g. The final ointment formulation was prepared using method B. This research not only contributes to the optimization of the WYH ointment preparation process but also provides valuable insights and practical guidance for designing the preparation processes of other TCM ointments. This advanced DSD method enhances the screening approach for identifying the best preparation process, thereby improving the scientific rigor and quality of TCM ointment preparation processes.

Construction of a machine learning-based risk prediction model for depression in middle-aged and elderly hypertensive people in China: a longitudinal study

Background

Hypertension is a common chronic disease that can trigger symptoms such as anxiety and depression. Therefore, it is essential to predict their risk of depression. The aim of this study is to find the best prediction model and provide effective intervention strategies for health professionals.

Methods

The study subjects were 2733 middle-aged and older adults who participated in the China Health and Retirement Longitudinal Study (CHARLS) between 2018 and 2020. R software was used for Lasso regression analysis to screen the best predictor variables, and logistic regression, random forest and XGBoost models were constructed. Finally, the prediction efficiency of the three models was compared.

Results

In this study, 18 variables were included, and LASSO regression screened out 10 variables that were important for the establishment of the model. Among the three models, Logistic Regression model showed the best performance in various evaluation indicators.

Conclusion

The prediction model based on machine learning can accurately assess the likelihood of depression in middle-aged and elderly patients with hypertension in the next three years. And by combining Logistic regression and nomograms, we were able to provide a clear interpretation of personalized risk predictions.

Identification of MACC1 as a potential biomarker for pulmonary arterial hypertension based on bioinformatics and machine learning

BACKGROUND

Pulmonary arterial hypertension (PAH) is a life-threatening disease characterized by abnormal early activation of pulmonary arterial smooth muscle cells (PASMCs), yet the underlying mechanisms remain to be elucidated.

METHODS

Normal and PAH gene expression profiles were obtained from the Gene Expression Omnibus (GEO) database and analyzed using gene set enrichment analysis (GSEA) to uncover the underlying mechanisms. Weighted gene co-expression network analysis (WGCNA) and machine learning methods were deployed to further filter hub genes. A number of immune infiltration analysis methods were applied to explore the immune landscape of PAH. Enzyme-linked immunosorbent assay (ELISA) was employed to compare MACC1 levels between PAH and normal subjects. The important role of MACC1 in the progression of PAH was verified through Western blot and real-time qPCR, among others.

RESULTS

39 up-regulated and 7 down-regulated genes were identified by 'limma' and 'RRA' packages. WGCNA and machine learning further narrowed down the list to 4 hub genes, with MACC1 showing strong diagnostic capacity. In vivo and in vitro experiments revealed that MACC1 was highsly associated with malignant features of PASMCs in PAH.

CONCLUSIONS

These findings suggest that targeting MACC1 may offer a promising therapeutic strategy for treating PAH, and further clinical studies are warranted to evaluate its efficacy.

The role of machine learning in advancing diabetic foot: a review

Background

Diabetic foot complications impose a significant strain on healthcare systems worldwide, acting as a principal cause of morbidity and mortality in individuals with diabetes mellitus. While traditional methods in diagnosing and treating these conditions have faced limitations, the emergence of Machine Learning (ML) technologies heralds a new era, offering the promise of revolutionizing diabetic foot care through enhanced precision and tailored treatment strategies.

Objective

This review aims to explore the transformative impact of ML on managing diabetic foot complications, highlighting its potential to advance diagnostic accuracy and therapeutic approaches by leveraging developments in medical imaging, biomarker detection, and clinical biomechanics.

Methods

A meticulous literature search was executed across PubMed, Scopus, and Google Scholar databases to identify pertinent articles published up to March 2024. The search strategy was carefully crafted, employing a combination of keywords such as "Machine Learning," "Diabetic Foot," "Diabetic Foot Ulcers," "Diabetic Foot Care," "Artificial Intelligence," and "Predictive Modeling." This review offers an in-depth analysis of the foundational principles and algorithms that constitute ML, placing a special emphasis on their relevance to the medical sciences, particularly within the specialized domain of diabetic foot pathology. Through the incorporation of illustrative case studies and schematic diagrams, the review endeavors to elucidate the intricate computational methodologies involved.

Results

ML has proven to be invaluable in deriving critical insights from complex datasets, enhancing both the diagnostic precision and therapeutic planning for diabetic foot management. This review highlights the efficacy of ML in clinical decision-making, underscored by comparative analyses of ML algorithms in prognostic assessments and diagnostic applications within diabetic foot care.

Conclusion

The review culminates in a prospective assessment of the trajectory of ML applications in the realm of diabetic foot care. We believe that despite challenges such as computational limitations and ethical considerations, ML remains at the forefront of revolutionizing treatment paradigms for the management of diabetic foot complications that are globally applicable and precision-oriented. This technological evolution heralds unprecedented possibilities for treatment and opportunities for enhancing patient care.

Analyzing Longitudinal Health Screening Data with Feature Ensemble and Machine Learning Techniques: Investigating Diagnostic Risk Factors of Metabolic Syndrome for Chronic Kidney Disease Stages 3a to 3b

Longitudinal data, while often limited, contain valuable insights into features impacting clinical outcomes. To predict the progression of chronic kidney disease (CKD) in patients with metabolic syndrome, particularly those transitioning from stage 3a to 3b, where data are scarce, utilizing feature ensemble techniques can be advantageous. It can effectively identify crucial risk factors, influencing CKD progression, thereby enhancing model performance. Machine learning (ML) methods have gained popularity due to their ability to perform feature selection and handle complex feature interactions more effectively than traditional approaches. However, different ML methods yield varying feature importance information. This study proposes a multiphase hybrid risk factor evaluation scheme to consider the diverse feature information generated by ML methods. The scheme incorporates variable ensemble rules (VERs) to combine feature importance information, thereby aiding in the identification of important features influencing CKD progression and supporting clinical decision making. In the proposed scheme, we employ six ML models-Lasso, RF, MARS, LightGBM, XGBoost, and CatBoost-each renowned for its distinct feature selection mechanisms and widespread usage in clinical studies. By implementing our proposed scheme, thirteen features affecting CKD progression are identified, and a promising AUC score of 0.883 can be achieved when constructing a model with them.

Bibliometric Analysis of Development Trends and Research Hotspots in the Study of Data Mining in Nursing Based on CiteSpace

Backgrounds

With the advent of the big data era, hospital information systems and mobile care systems, among others, generate massive amounts of medical data. Data mining, as a powerful information processing technology, can discover non-obvious information by processing large-scale data and analyzing them in multiple dimensions. How to find the effective information hidden in the database and apply it to nursing clinical practice has received more and more attention from nursing researchers.

Aim

To look over the articles on data mining in nursing, compiled research status, identified hotspots, highlighted research trends, and offer recommendations for how data mining technology might be used in the nursing area going forward.

Methods

Data mining in nursing publications published between 2002 and 2023 were taken from the Web of Science Core Collection. CiteSpace was utilized for reviewing the number of articles, countries/regions, institutions, journals, authors, and keywords.

Results

According to the findings, the pace of data mining in nursing progress is not encouraging. Nursing data mining research is dominated by the United States and China. However, no consistent core group of writers or organizations has emerged in the field of nursing data mining. Studies on data mining in nursing have been increasingly gradually conducted in the 21st century, but the overall number is not large. Institution of Columbia University, journal of Cin-computers Informatics Nursing, author Diana J Wilkie, Muhammad Kamran Lodhi, Yingwei Yao are most influential in nursing data mining research. Nursing data mining researchers are currently focusing on electronic health records, text mining, machine learning, and natural language processing. Future research themes in data mining in nursing most include nursing informatics and clinical care quality enhancement.

Conclusion

Research data shows that data mining gives more perspectives for the growth of the nursing discipline and encourages the discipline's development, but it also introduces a slew of new issues that need researchers to address.

Multiparametric hippocampal signatures for early diagnosis of Alzheimer's disease using 18F-FDG PET/MRI Radiomics

PURPOSE

This study aimed to explore the utility of hippocampal radiomics using multiparametric simultaneous positron emission tomography (PET)/magnetic resonance imaging (MRI) for early diagnosis of Alzheimer's disease (AD).

METHODS

A total of 53 healthy control (HC) participants, 55 patients with amnestic mild cognitive impairment (aMCI), and 51 patients with AD were included in this study. All participants accepted simultaneous PET/MRI scans, including 18F-fluorodeoxyglucose (18F-FDG) PET, 3D arterial spin labeling (ASL), and high-resolution T1-weighted imaging (3D T1WI). Radiomics features were extracted from the hippocampus region on those three modal images. Logistic regression models were trained to classify AD and HC, AD and aMCI, aMCI and HC respectively. The diagnostic performance and radiomics score (Rad-Score) of logistic regression models were evaluated from 5-fold cross-validation.

RESULTS

The hippocampal radiomics features demonstrated favorable diagnostic performance, with the multimodal classifier outperforming the single-modal classifier in the binary classification of HC, aMCI, and AD. Using the multimodal classifier, we achieved an area under the receiver operating characteristic curve (AUC) of 0.98 and accuracy of 96.7% for classifying AD from HC, and an AUC of 0.86 and accuracy of 80.6% for classifying aMCI from HC. The value of Rad-Score differed significantly between the AD and HC (p < 0.001), aMCI and HC (p < 0.001) groups. Decision curve analysis showed superior clinical benefits of multimodal classifiers compared to neuropsychological tests.

CONCLUSION

Multiparametric hippocampal radiomics using PET/MRI aids in the identification of early AD, and may provide a potential biomarker for clinical applications.

Risk of papillary thyroid carcinoma and nodular goiter associated with exposure to semi-volatile organic compounds: A multi-pollutant assessment based on machine learning algorithms

BACKGROUND

Exposure to semi-volatile organic compounds (SVOCs) may link to thyroid nodule risk, but studies of mixed-SVOCs exposure effects are lacking. Traditional analytical methods are inadequate for dealing with mixed exposures, while machine learning (ML) seems to be a good way to fill the gaps in the field of environmental epidemiology research.

OBJECTIVES

Different ML algorithms were used to explore the relationship between mixed-SVOCs exposure and thyroid nodule.

METHODS

A 1:1:1 age- and gender-matched case-control study was conducted in which 96 serum SVOCs were measured in 50 papillary thyroid carcinoma (PTC), 50 nodular goiters (NG), and 50 controls. Different ML techniques such as Random Forest, AdaBoost were selected based on their predictive power, and variables were selected based on their weights in the models. Weighted quantile sum (WQS) regression and Bayesian kernel machine regression (BKMR) were used to assess the mixed effects of the SVOCs exposure on thyroid nodule.

RESULTS

Forty-three of 96 SVOCs with detection rate >80 % were included in the analysis. ML algorithms showed a consistent selection of SVOCs associated with thyroid nodule. Fluazifop-butyl and fenpropathrin are positively associated with PTC and NG in single compound models (all P < 0.05). WQS model shows that exposure to mixed-SVOCs was associated with an increased risk of PTC and NG, with the mixture dominated by fenpropathrin, followed by fluazifop-butyl and propham. In the BKMR model, mixtures showed a significant positive association with thyroid nodule risk at high exposure levels, and fluazifop-butyl showed positive effects associated with PTC and NG.

CONCLUSION

This study confirms the feasibility of ML methods for variable selection in high-dimensional complex data and showed that mixed exposure to SVOCs was associated with increased risk of PTC and NG. The observed association was primarily driven by fluazifop-butyl and fenpropathrin. The findings warranted further investigation.

The relationship between anthropometric indices and the presence of hypertension in an Iranian population sample using data mining algorithms

Hypertension (HTN) is a common chronic condition associated with increased morbidity and mortality. Anthropometric indices of adiposity are known to be associated with a risk of HTN. The aim of this study was to identify the anthropometric indices that best associate with HTN in an Iranian population. 9704 individuals aged 35-65 years were recruited as part of the Mashhad Stroke and Heart Atherosclerotic Disorder (MASHAD) study. Demographic and anthropometric data of all participants were recorded. HTN was defined as a systolic blood pressure (SBP) ≥ 140 mmHg, and/ or a diastolic blood pressure (DBP) ≥ 90 mmHg on two subsequent measurements, or being treated with oral drug therapy for BP. Data mining methods including Logistic Regression (LR), Decision Tree (DT), and Bootstrap Forest (BF) were applied. Of 9704 participants, 3070 had HTN, and 6634 were normotensive. LR showed that body roundness index (BRI), body mass index (BMI) and visceral adiposity index (VAI) were significantly associated with HTN in both genders (P < 0.0001). BRI showed the greatest association with HTN (OR = 1.276, 95%CI = (1.224, 1.330)). For BMI we had OR = 1.063, 95%CI = (1.047, 1.080), for VAI we had OR = 1.029, 95%CI = (1.020, 1.038). An age < 47 years and BRI < 4.04 was associated with a 90% probability of being normotensive. The BF indicated that age, sex and BRI had the most important role in HTN. In summary, among anthropometric indices the most powerful indicator for discriminating hypertensive from normotensive patients was BRI.

Machine learning-based identification of colorectal advanced adenoma using clinical and laboratory data: a phase I exploratory study in accordance with updated World Endoscopy Organization guidelines for noninvasive colorectal cancer screening tests

Objective

The recent World Endoscopy Organization (WEO) guidelines now recognize precursor lesions of colorectal cancer (CRC) as legitimate screening targets. However, an optimal screening method for detecting advanced adenoma (AA), a significant precursor lesion, remains elusive.

Methods

We employed five machine learning methods, using clinical and laboratory data, to develop and validate a diagnostic model for identifying patients with AA (569 AAs vs. 3228 controls with normal colonoscopy). The best-performing model was selected based on sensitivity and specificity assessments. Its performance in recognizing adenoma-carcinoma sequence was evaluated in line with guidelines, and adjustable thresholds were established. For comparison, the Fecal Occult Blood Test (FOBT) was also selected.

Results

The XGBoost model demonstrated superior performance in identifying AA, with a sensitivity of 70.8% and a specificity of 83.4%. It successfully detected 42.7% of non-advanced adenoma (NAA) and 80.1% of CRC. The model-transformed risk assessment scale provided diagnostic performance at different positivity thresholds. Compared to FOBT, the XGBoost model better identified AA and NAA, however, was less effective in CRC.

Conclusion

The XGBoost model, compared to FOBT, offers improved accuracy in identifying AA patients. While it may not meet the recommendations of some organizations, it provides value for individuals who are unable to use FOBT for various reasons.

White blood cell and platelet distribution widths are associated with hypertension: data mining approaches

In this paper, we are going to investigate the association between Hypertension (HTN) and routine hematologic indices in a cohort of Iranian adults. The data were obtained from a total population of 9704 who were aged 35-65 years, a prospective study was designed. The association between hematologic factors and HTN was assessed using logistic regression (LR) analysis and a decision tree (DT) algorithm. A total of 9704 complete datasets were analyzed in this cohort study (N = 3070 with HTN [female 62.47% and male 37.52%], N = 6634 without HTN [female 58.90% and male 41.09%]). Several variables were significantly different between the two groups, including age, smoking status, BMI, diabetes millitus, high sensitivity C-reactive protein (hs-CRP), uric acid, FBS, total cholesterol, HGB, LYM, WBC, PDW, RDW, RBC, sex, PLT, MCV, SBP, DBP, BUN, and HCT (P < 0.05). For unit odds ratio (OR) interpretation, females are more likely to have HTN (OR = 1.837, 95% CI = (1.620, 2.081)). Among the analyzed variables, age and WBC had the most significant associations with HTN OR = 1.087, 95% CI = (1.081, 1.094) and OR = 1.096, 95% CI = (1.061, 1.133), respectively (P-value < 0.05). In the DT model, age, followed by WBC, sex, and PDW, has the most significant impact on the HTN risk. Ninety-eight percent of patients had HTN in the subgroup with older age (≥58), high PDW (≥17.3), and low RDW (<46). Finally, we found that elevated WBC and PDW are the most associated factor with the severity of HTN in the Mashhad general population as well as female gender and older age.

Development and internal validation of diagnostic prediction models using machine-learning algorithms in dogs with hypothyroidism

Introduction

Hypothyroidism can be easily misdiagnosed in dogs, and prediction models can support clinical decision-making, avoiding unnecessary testing and treatment. The aim of this study is to develop and internally validate diagnostic prediction models for hypothyroidism in dogs by applying machine-learning algorithms.

Methods

A single-institutional cross-sectional study was designed searching the electronic database of a Veterinary Teaching Hospital for dogs tested for hypothyroidism. Hypothyroidism was diagnosed based on suggestive clinical signs and thyroid function tests. Dogs were excluded if medical records were incomplete or a definitive diagnosis was lacking. Predictors identified after data processing were dermatological signs, alopecia, lethargy, hematocrit, serum concentrations of cholesterol, creatinine, total thyroxine (tT4), and thyrotropin (cTSH). Four models were created by combining clinical signs and clinicopathological variables expressed as quantitative (models 1 and 2) and qualitative variables (models 3 and 4). Models 2 and 4 included tT4 and cTSH, models 1 and 3 did not. Six different algorithms were applied to each model. Internal validation was performed using a 10-fold cross-validation. Apparent performance was evaluated by calculating the area under the receiver operating characteristic curve (AUROC).

Results

Eighty-two hypothyroid and 233 euthyroid client-owned dogs were included. The best performing algorithms were naive Bayes in model 1 (AUROC = 0.85; 95% confidence interval [CI] = 0.83-0.86) and in model 2 (AUROC = 0.98; 95% CI = 0.97-0.99), logistic regression in model 3 (AUROC = 0.88; 95% CI = 0.86-0.89), and random forest in model 4 (AUROC = 0.99; 95% CI = 0.98-0.99). Positive predictive value was 0.76, 0.84, 0.93, and 0.97 in model 1, 2, 3, and 4, respectively. Negative predictive value was 0.89, 0.89, 0.99, and 0.99 in model 1, 2, 3, and 4, respectively.

Discussion

Machine learning-based prediction models were accurate in predicting and quantifying the likelihood of hypothyroidism in dogs based on internal validation performed in a single-institution, but external validation is required to support the clinical applicability of these models.

Predictive Value of Machine Learning Models in Postoperative Mortality of Older Adults Patients with Hip Fracture: A Systematic Review and Meta-analysis

BACKGROUND

Some researchers have used machine learning to predict mortality in old patients with hip fracture, but its application value lacks an evidence-based basis. Hence, we conducted this meta-analysis to explore the predictive accuracy of machine learning for mortality in old patients with hip fracture.

METHODS

We systematically retrieved PubMed, Cochrane, Embase, and Web of Science for relevant studies published before July 15, 2022. The PROBAST assessment tool was used to assess the risk of bias in the included studies. A random-effects model was used for the meta-analysis of C-index, whereas a bivariate mixed-effects model was used for the meta-analysis of sensitivity and specificity. The meta-analysis was performed on R and Stata.

RESULTS

Eighteen studies were included, involving 8 machine learning models and 398,422 old patients undergoing hip joint surgery, of whom 60,457 died. According to the meta-analysis, the pooled C-index for machine learning models was 0.762 (95% CI: 0.691 ∼ 0.833) in the training set and 0.838 (95% CI: 0.783 ∼ 0.892) in the validation set, which is better than the C-index of the main clinical scale (Nottingham Hip Fracture Score), that is, 0.702 (95% CI: 0.681 ∼ 0.723). Among different machine learning models, ANN and Bayesian belief network had the best predictive performance.

CONCLUSION

Machine learning models are more accurate in predicting mortality in old patients after hip joint surgery than current mainstream clinical scoring systems. Subsequent research could focus on updating clinical scoring systems and improving their predictive performance by relying on machine learning models.

Potential value of gastrointestinal myoelectrical activity in the diagnosis of anxiety-depression disorder: a population-based study

BACKGROUND

Depression and anxiety are frequently coexisted mental illness. The lack of solid objective diagnostic criteria has led to a high rate of suicide. The brain-gut axis bridges the gastrointestinal system with neuropsychiatric disorders. However, it is still not possible to reflect mental disease with gastrointestinal information. The study aimed to explore the auxiliary diagnostic value of gastrointestinal myoelectrical activity in anxiety-depression disorders (ADD) without gastrointestinal disturbance.

METHODS

A natural population cohort from 3 districts in Western China were established. The Patient Health Questionnaire-9 and the Generalized Anxiety Disorder-7 were used to assess ADD. Gastrointestinal myoelectrical activity of ADD were measured by multi-channel cutaneous electrogastroenterogram (EGEG). Then the parameters of EGEG between ADD and healthy controls were analyzed.

RESULTS

The average amplitude and response area of intestinal channel in ADD were significantly lower than those of controls (153.49 ± 78.69 vs. 179.83 ± 103.90, 57.27 ± 29.05 vs. 67.70 ± 38.32), which were shown to be protective factors for ADD (OR = 0.944 and 0.844, respectively). Further, the scale item scores related to the core symptoms of anxiety and depression were also associated with these two channels (p < 0.05), and the gastrointestinal electrical signals of ADD are significantly changed in the elderly compared to the young adults.

CONCLUSIONS

The intestinal myoelectrical activity has a certain auxiliary diagnostic value in psychiatric disorders and is expected to provide objective reference for the diagnosis of anxiety and depression.

Machine Learning and Radiomics of Bone Scintigraphy: Their Role in Predicting Recurrence of Localized or Locally Advanced Prostate Cancer

BACKGROUND

Machine-learning (ML) and radiomics features have been utilized for survival outcome analysis in various cancers. This study aims to investigate the application of ML based on patients' clinical features and radiomics features derived from bone scintigraphy (BS) and to evaluate recurrence-free survival in local or locally advanced prostate cancer (PCa) patients after the initial treatment.

METHODS

A total of 354 patients who met the eligibility criteria were analyzed and used to train the model. Clinical information and radiomics features of BS were obtained. Survival-related clinical features and radiomics features were included in the ML model training. Using the pyradiomics software, 128 radiomics features from each BS image's region of interest, validated by experts, were extracted. Four textural matrices were also calculated: GLCM, NGLDM, GLRLM, and GLSZM. Five training models (Logistic Regression, Naive Bayes, Random Forest, Support Vector Classification, and XGBoost) were applied using K-fold cross-validation. Recurrence was defined as either a rise in PSA levels, radiographic progression, or death. To assess the classifier's effectiveness, the ROC curve area and confusion matrix were employed.

RESULTS

Of the 354 patients, 101 patients were categorized into the recurrence group with more advanced disease status compared to the non-recurrence group. Key clinical features including tumor stage, radical prostatectomy, initial PSA, Gleason Score primary pattern, and radiotherapy were used for model training. Random Forest (RF) was the best-performing model, with a sensitivity of 0.81, specificity of 0.87, and accuracy of 0.85. The ROC curve analysis showed that predictions from RF outperformed predictions from other ML models with a final AUC of 0.94 and a p-value of <0.001. The other models had accuracy ranges from 0.52 to 0.78 and AUC ranges from 0.67 to 0.84.

CONCLUSIONS

The study showed that ML based on clinical features and radiomics features of BS improves the prediction of PCa recurrence after initial treatment. These findings highlight the added value of ML techniques for risk classification in PCa based on clinical features and radiomics features of BS.

Platelet distribution widths and white blood cell are associated with cardiovascular diseases: data mining approaches

OBJECTIVE

To investigate the association between cardiovascular diseases (CVDs) and haematologic factors in a cohort of Iranian adults.

METHOD

For a total population of 9,704 aged 35 to 65, a prospective study was designed. Haematologic factors and demographic characteristics (such as gender, age, and smoking status) were completed for all participants. The association between haematologic factors and CVDs was assessed through logistic regression (LR) analysis, decision tree (DT), and bootstrap forest (BF).

RESULTS

Almost all of the included factors were significantly associated with CVD (p<.001). Among the included factors, were: age, white blood cell (WBC), and platelet distribution width (PDW) had the strongest correlation with the development of CVD. For unit OR interpretation, WBC has been represented as the most remarkable risk factor for CVD (OR: 1.22 (CI 95% (1.18, 1.27))). Also, age is associated with an increase in the odds of CVD + occurrence (OR: 1.12 (CI 95% (1.11, 1.13))). Moreover, males are times more likely to develop CVD than females (OR: 1.39 (CI 95% (1.22, 1.58))). In DT model, age is the best classifier factor in CVD development, followed by WBC and PDW. Furthermore, based on the BF algorithm, the most crucial factors correlated with CVD are age, WBC, PDW, sex, and smoking status.

CONCLUSION

The obtained result from LR, DT, and BF models confirmed that age, WBC, and PDW are the most crucial factors for the development of CVD.

An integrated machine learning predictive scheme for longitudinal laboratory data to evaluate the factors determining renal function changes in patients with different chronic kidney disease stages

Background and objectives

Chronic kidney disease (CKD) is a global health concern. This study aims to identify key factors associated with renal function changes using the proposed machine learning and important variable selection (ML&IVS) scheme on longitudinal laboratory data. The goal is to predict changes in the estimated glomerular filtration rate (eGFR) in a cohort of patients with CKD stages 3-5.

Design

A retrospective cohort study.

Setting and participants

A total of 710 outpatients who presented with stable nondialysis-dependent CKD stages 3-5 at the Shin-Kong Wu Ho-Su Memorial Hospital Medical Center from 2016 to 2021.

Methods

This study analyzed trimonthly laboratory data including 47 indicators. The proposed scheme used stochastic gradient boosting, multivariate adaptive regression splines, random forest, eXtreme gradient boosting, and light gradient boosting machine algorithms to evaluate the important factors for predicting the results of the fourth eGFR examination, especially in patients with CKD stage 3 and those with CKD stages 4-5, with or without diabetes mellitus (DM).

Main outcome measurement

Subsequent eGFR level after three consecutive laboratory data assessments.

Results

Our ML&IVS scheme demonstrated superior predictive capabilities and identified significant factors contributing to renal function changes in various CKD groups. The latest levels of eGFR, blood urea nitrogen (BUN), proteinuria, sodium, and systolic blood pressure as well as mean levels of eGFR, BUN, proteinuria, and triglyceride were the top 10 significantly important factors for predicting the subsequent eGFR level in patients with CKD stages 3-5. In individuals with DM, the latest levels of BUN and proteinuria, mean levels of phosphate and proteinuria, and variations in diastolic blood pressure levels emerged as important factors for predicting the decline of renal function. In individuals without DM, all phosphate patterns and latest albumin levels were found to be key factors in the advanced CKD group. Moreover, proteinuria was identified as an important factor in the CKD stage 3 group without DM and CKD stages 4-5 group with DM.

Conclusion

The proposed scheme highlighted factors associated with renal function changes in different CKD conditions, offering valuable insights to physicians for raising awareness about renal function changes.

Prediction model of ocular metastasis from primary liver cancer: Machine learning-based development and interpretation study

BACKGROUND

Ocular metastasis (OM) is a rare metastatic site of primary liver cancer (PLC). The purpose of this study was to establish a clinical predictive model of OM in PLC patients based on machine learning (ML).

METHODS

We retrospectively collected the clinical data of 1540 PLC patients and divided it into a training set and an internal test set in a 7:3 proportion. PLC patients were divided into OM and non-ocular metastasis (NOM) groups, and univariate logistic regression analysis was performed between the two groups. The variables with univariate logistic analysis p < 0.05 were selected for the ML model. We constructed six ML models, which were internally verified by 10-fold cross-validation. The prediction performance of each ML model was evaluated by receiver operating characteristic curves (ROCs). We also constructed a web calculator based on the optimal performance ML model to personalize the risk probability for OM.

RESULTS

Six variables were selected for the ML model. The extreme gradient boost (XGB) ML model achieved the optimal differential diagnosis ability, with an area under the curve (AUC) = 0.993, accuracy = 0.992, sensitivity = 0.998, and specificity = 0.984. Based on these results, an online web calculator was constructed by using the XGB ML model to help clinicians diagnose and treat the risk probability of OM in PLC patients. Finally, the Shapley additive explanations (SHAP) library was used to obtain the six most important risk factors for OM in PLC patients: CA125, ALP, AFP, TG, CA199, and CEA.

CONCLUSION

We used the XGB model to establish a risk prediction model of OM in PLC patients. The predictive model can help identify PLC patients with a high risk of OM, provide early and personalized diagnosis and treatment, reduce the poor prognosis of OM patients, and improve the quality of life of PLC patients.

Clinical Data based XGBoost Algorithm for infection risk prediction of patients with decompensated cirrhosis: a 10-year (2012-2021) Multicenter Retrospective Case-control study

OBJECTIVES

To appraise effective predictors for infection in patients with decompensated cirrhosis (DC) by using XGBoost algorithm in a retrospective case-control study.

METHODS

Clinical data were retrospectively collected from 6,648 patients with DC admitted to five tertiary hospitals. Indicators with significant differences were determined by univariate analysis and least absolute contraction and selection operator (LASSO) regression. Further multi-tree extreme gradient boosting (XGBoost) machine learning-based model was used to rank importance of features selected from LASSO and subsequently constructed infection risk prediction model with simple-tree XGBoost model. Finally, the simple-tree XGBoost model is compared with the traditional logical regression (LR) model. Performances of models were evaluated by area under the receiver operating characteristic curve (AUROC), sensitivity, and specificity.

RESULTS

Six features, including total bilirubin, blood sodium, albumin, prothrombin activity, white blood cell count, and neutrophils to lymphocytes ratio were selected as predictors for infection in patients with DC. Simple-tree XGBoost model conducted by these features can predict infection risk accurately with an AUROC of 0.971, sensitivity of 0.915, and specificity of 0.900 in training set. The performance of simple-tree XGBoost model is better than that of traditional LR model in training set, internal verification set, and external feature set (P < 0.001).

CONCLUSIONS

The simple-tree XGBoost predictive model developed based on a minimal amount of clinical data available to DC patients with restricted medical resources could help primary healthcare practitioners promptly identify potential infection.

A Novel Predictive Machine Learning Model Integrating Cytokines in Cervical-Vaginal Mucus Increases the Prediction Rate for Preterm Birth

Preterm birth (PB) is a leading cause of perinatal morbidity and mortality. PB prediction is performed by measuring cervical length, with a detection rate of around 70%. Although it is known that a cytokine-mediated inflammatory process is involved in the pathophysiology of PB, none screening method implemented in clinical practice includes cytokine levels as a predictor variable. Here, we quantified cytokines in cervical-vaginal mucus of pregnant women (18-23.6 weeks of gestation) with high or low risk for PB determined by cervical length, also collecting relevant obstetric information. IL-2, IL-6, IFN-γ, IL-4, and IL-10 were significantly higher in the high-risk group, while IL-1ra was lower. Two different models for PB prediction were created using the Random Forest machine-learning algorithm: a full model with 12 clinical variables and cytokine values and the adjusted model, including the most relevant variables-maternal age, IL-2, and cervical length- (detection rate 66 vs. 87%, false positive rate 12 vs. 3.33%, false negative rate 28 vs. 6.66%, and area under the curve 0.722 vs. 0.875, respectively). The adjusted model that incorporate cytokines showed a detection rate eight points higher than the gold standard calculator, which may allow us to identify the risk PB risk more accurately and implement strategies for preventive interventions.

A bidirectional link between metabolic syndrome and elevation in alanine aminotransferase in elderly female: a longitudinal community study

Pre-obesity, as a significant risk factor for the progression of metabolic syndrome (MS), has become a prevalent public health threat globally. In this three-year longitudinal study of pre-obese women at baseline, the goal was to clarify the female-specific bidirectional relationship between the risk of MS and blood alanine aminotransferase. In this manuscript, the MS score was determined using the following equation: MS score = 2*waist/height + fasting glucose/5.6 + TG/1.7 + SBP/130-HDL/1.02 for men and 1.28 for women, which is highly related to the risk of MS. With 2,338 participants, a hierarchical nonlinear model with random effects was utilized to analyze the temporal trends of serum characteristics from 2017 to 2019. A bivariate cross-lagged panel model (CLPM) was employed to estimate the structural relations of frequently measured variables at three different time points to determine the directionality of the relationship between the risk of MS and serum characteristics. MassARRAY Analyzer 4 platforms were used to evaluate and genotype candidate SNPs. In this study, the MS score only rose with age in females; it was positively correlated with serum alanine aminotransferase (ALT) in females; the CLPM revealed that the MS score in 2017 predicted ALT in 2018 (β = 0.066, p < 0.001); and ALT in 2018 predicted an MS score in 2019 (β = 0.037, p < 0.050); both relationships were seen in females. Additionally, the MS score in elderly females with NAFLD was related to the rs295 in the lipoprotein lipase (LPL) gene (p = 0.042). Our work showed that there may be female-specific causal correlations between elevated ALT and risk of MS and that the polymorphism rs295 in LPL may serve as a marker for the prognosis of MS. The genetic roles of rs295 in the LPL gene in the onset of MS and the development of ALT in the elderly Chinese Han population are thus provided by this, offering one potential mechanism.

Development of Data Mining Algorithms for Identifying the Best Anthropometric Predictors for Cardiovascular Disease: MASHAD Cohort Study

INTRODUCTION

Many studies have been published to assess the best anthropometric measurements associated with cardiovascular diseases (CVDs), but controversies still exist.

AIM

Investigating the association between CVDs and anthropometric measurements among Iranian adults.

METHODS

For a total population of 9354 aged 35 to 65, a prospective study was designed. Anthropometric measurements including ABSI (A Body Shape Index), Body Adiposity Index (BAI), Body Mass Index (BMI), Waist to Height Ratio (WHtR), Body Round Index (BRI), HC (Hip Circumference), Demispan, Mid-arm circumference (MAC), Waist-to-hip (WH) and Waist Circumference (WC) were completed. The association between these parameters and CVDs were assessed through logistic regression (LR) and decision tree (DT) models.

RESULTS

During the 6-year follow-up, 4596 individuals (49%) developed CVDs. According to the LR, age, BAI, BMI, Demispan, and BRI, in male and age, WC, BMI, and BAI in female had a significant association with CVDs (p-value < 0.03). Age and BRI for male and age and BMI for female represent the most appropriate estimates for CVDs (OR: 1.07, (95% CI: 1.06, 1.08), 1.36 (1.22, 1.51), 1.14 (1.13, 1.15), and 1.05 (1.02, 1.07), respectively). In the DT for male, those with BRI ≥ 3.87, age ≥ 46 years, and BMI ≥ 35.97 had the highest risk to develop CVDs (90%). Also, in the DT for female, those with age ≥ 54 years and WC ≥ 84 had the highest risk to develop CVDs (71%).

CONCLUSION

BRI and age in male and age and BMI in female had the greatest association with CVDs. Also, BRI and BMI was the strongest indices for this prediction.

Machine Learning for Early Diagnosis of ATTRv Amyloidosis in Non-Endemic Areas: A Multicenter Study from Italy

BACKGROUND

Hereditary transthyretin amyloidosis with polyneuropathy (ATTRv) is an adult-onset multisystemic disease, affecting the peripheral nerves, heart, gastrointestinal tract, eyes, and kidneys. Nowadays, several treatment options are available; thus, avoiding misdiagnosis is crucial to starting therapy in early disease stages. However, clinical diagnosis may be difficult, as the disease may present with unspecific symptoms and signs. We hypothesize that the diagnostic process may benefit from the use of machine learning (ML).

METHODS

397 patients referring to neuromuscular clinics in 4 centers from the south of Italy with neuropathy and at least 1 more red flag, as well as undergoing genetic testing for ATTRv, were considered. Then, only probands were considered for analysis. Hence, a cohort of 184 patients, 93 with positive and 91 (age- and sex-matched) with negative genetics, was considered for the classification task. The XGBoost (XGB) algorithm was trained to classify positive and negative TTR mutation patients. The SHAP method was used as an explainable artificial intelligence algorithm to interpret the model findings.

RESULTS

diabetes, gender, unexplained weight loss, cardiomyopathy, bilateral carpal tunnel syndrome (CTS), ocular symptoms, autonomic symptoms, ataxia, renal dysfunction, lumbar canal stenosis, and history of autoimmunity were used for the model training. The XGB model showed an accuracy of 0.707 ± 0.101, a sensitivity of 0.712 ± 0.147, a specificity of 0.704 ± 0.150, and an AUC-ROC of 0.752 ± 0.107. Using the SHAP explanation, it was confirmed that unexplained weight loss, gastrointestinal symptoms, and cardiomyopathy showed a significant association with the genetic diagnosis of ATTRv, while bilateral CTS, diabetes, autoimmunity, and ocular and renal involvement were associated with a negative genetic test.

CONCLUSIONS

Our data show that ML might potentially be a useful instrument to identify patients with neuropathy that should undergo genetic testing for ATTRv. Unexplained weight loss and cardiomyopathy are relevant red flags in ATTRv in the south of Italy. Further studies are needed to confirm these findings.

High-Resolution and Multidimensional Phenotypes Can Complement Genomics Data to Diagnose Diseases in the Neonatal Population

Advances in genomic medicine have greatly improved our understanding of human diseases. However, phenome is not well understood. High-resolution and multidimensional phenotypes have shed light on the mechanisms underlying neonatal diseases in greater details and have the potential to optimize clinical strategies. In this review, we first highlight the value of analyzing traditional phenotypes using a data science approach in the neonatal population. We then discuss recent research on high-resolution, multidimensional, and structured phenotypes in neonatal critical diseases. Finally, we briefly introduce current technologies available for the analysis of multidimensional data and the value that can be provided by integrating these data into clinical practice. In summary, a time series of multidimensional phenome can improve our understanding of disease mechanisms and diagnostic decision-making, stratify patients, and provide clinicians with optimized strategies for therapeutic intervention; however, the available technologies for collecting multidimensional data and the best platform for connecting multiple modalities should be considered.

Spatial metabolomics reveal mechanisms of dexamethasone against pediatric pneumonia

Currently, drugs are limited to treating pediatric pneumonia in clinical practice. It is urgent to find one new precise prevention and control therapy. The dynamically changing biomarkers during the development of pediatric pneumonia could help diagnose this disease, determine its severity, assess the risk of future events, and guide its treatment. Dexamethasone has been recognized as an effective agent with anti-inflammatory activity. However, its mechanisms against pediatric pneumonia remain unclear. In this study, spatial metabolomics was used to reveal the potential and characteristics of dexamethasone. Specifically, bioinformatics was first applied to find the critical biomarkers of differential expression in pediatric pneumonia. Subsequently, Desorption Electrospray Ionization mass spectrometry imaging-based metabolomics screened the differential metabolites affected by dexamethasone. Then, a gene-metabolite interaction network was built to mark functional correlation pathways for exploring integrated information and core biomarkers related to the pathogenesis and etiology of pediatric pneumonia. Further, these were validated by molecular biology and targeted metabolomics. As a result, genes of Cluster of Differentiation19, Fc fragment of IgG receptor IIb, Cluster of Differentiation 22, B-cell linker, Cluster of Differentiation 79B and metabolites of Triethanolamine, Lysophosphatidylcholine(18:1(9Z)), Phosphatidylcholine(16:0/16:0), phosphatidylethanolamine(O-18:1(1Z)/20:4(5Z,8Z,11Z,14Z)) were identified as the critical biomarkers in pediatric pneumonia. B cell receptor signaling pathway and glycerophospholipid metabolism were integrally analyzed as the main pathways of these biomarkers. The above data were illustrated using a Lipopolysaccharides-induced lung injury juvenile rat model. This work will provide evidence for the precise treatment of pediatric pneumonia.

Machine Learning to Predict the Response to Lenvatinib Combined with Transarterial Chemoembolization for Unresectable Hepatocellular Carcinoma

BACKGROUND

Lenvatinib and transarterial chemoembolization (TACE) are first-line treatments for unresectable hepatocellular carcinoma (HCC), but the objective response rate (ORR) is not satisfactory. We aimed to predict the response to lenvatinib combined with TACE before treatment for unresectable HCC using machine learning (ML) algorithms based on clinical data.

METHODS

Patients with unresectable HCC receiving the combination therapy of lenvatinib combined with TACE from two medical centers were retrospectively collected from January 2020 to December 2021. The response to the combination therapy was evaluated over the following 4-12 weeks. Five types of ML algorithms were applied to develop the predictive models, including classification and regression tree (CART), adaptive boosting (AdaBoost), extreme gradient boosting (XGBoost), random forest (RF), and support vector machine (SVM). The performance of the models was assessed by the receiver operating characteristic (ROC) curve and area under the receiver operating characteristic curve (AUC). The Shapley Additive exPlanation (SHAP) method was applied to explain the model.

RESULTS

A total of 125 unresectable HCC patients were included in the analysis after the inclusion and exclusion criteria, among which 42 (33.6%) patients showed progression disease (PD), 49 (39.2%) showed stable disease (SD), and 34 (27.2%) achieved partial response (PR). The nonresponse group (PD + SD) included 91 patients, while the response group (PR) included 34 patients. The top 40 most important features from all 64 clinical features were selected using the recursive feature elimination (RFE) algorithm to develop the predictive models. The predictive power was satisfactory, with AUCs of 0.74 to 0.91. The SVM model and RF model showed the highest accuracy (86.5%), and the RF model showed the largest AUC (0.91, 95% confidence interval (CI): 0.61-0.95). The SHAP summary plot and decision plot illustrated the impact of the top 40 features on the efficacy of the combination therapy, and the SHAP force plot successfully predicted the efficacy at the individualized level.

CONCLUSIONS

A new predictive model based on clinical data was developed using ML algorithms, which showed favorable performance in predicting the response to lenvatinib combined with TACE for unresectable HCC. Combining ML with SHAP could provide an explicit explanation of the efficacy prediction.

Prediction of type 2 diabetes mellitus using hematological factors based on machine learning approaches: a cohort study analysis

Type 2 Diabetes Mellitus (T2DM) is a significant public health problem globally. The diagnosis and management of diabetes are critical to reduce the diabetes complications including cardiovascular disease and cancer. This study was designed to assess the potential association between T2DM and routinely measured hematological parameters. This study was a subsample of 9000 adults aged 35-65 years recruited as part of Mashhad stroke and heart atherosclerotic disorder (MASHAD) cohort study. Machine learning techniques including logistic regression (LR), decision tree (DT) and bootstrap forest (BF) algorithms were applied to analyze data. All data analyses were performed using SPSS version 22 and SAS JMP Pro version 13 at a significant level of 0.05. Based on the performance indices, the BF model gave high accuracy, precision, specificity, and AUC. Previous studies suggested the positive relationship of triglyceride-glucose (TyG) index with T2DM, so we considered the association of TyG index with hematological factors. We found this association was aligned with their results regarding T2DM, except MCHC. The most effective factors in the BF model were age and WBC (white blood cell). The BF model represented a better performance to predict T2DM. Our model provides valuable information to predict T2DM like age and WBC.

Determining the prevalence of childhood hypertension and its concomitant metabolic abnormalities using data mining methods in the Northeastern region of Hungary

Objective

Identifying hypertension in children and providing treatment for it have a marked impact on the patients' long-term cardiovascular outcomes. The global prevalence of childhood hypertension is increasing, yet its investigation has been rather sporadic in Eastern Europe. Therefore, our goal was to determine the prevalence of childhood hypertension and its concomitant metabolic abnormalities using data mining methods.

Methods

We evaluated data from 3 to 18-year-old children who visited the University of Debrecen Clinical Center's hospital throughout a 15-year study period (n = 92,198; boys/girls: 48/52%).

Results

We identified a total of 3,687 children with hypertension (2,107 boys and 1,580 girls), with a 4% calculated prevalence of hypertension in the whole study population and a higher prevalence in boys (4.7%) as compared to girls (3.2%). Among boys we found an increasing prevalence in consecutive age groups in the study population, but among girls the highest prevalences are identified in the 12-15-year age group. Markedly higher BMI values were found in hypertensive children as compared to non-hypertensives in all age groups. Moreover, significantly higher total cholesterol (4.27 ± 0.95 vs. 4.17 ± 0.88 mmol/L), LDL-C (2.62 ± 0.79 vs. 2.44 ± 0.74 mmol/L) and triglyceride (1.2 (0.85-1.69) vs. 0.94 (0.7-1.33) mmol/L), and lower HDL-C (1.2 ± 0.3 vs. 1.42 ± 0.39 mmol/L) levels were found in hypertensive children. Furthermore, significantly higher serum uric acid levels were found in children with hypertension (299.2 ± 86.1 vs. 259.9 ± 73.3 μmol/L), while glucose levels did not differ significantly.

Conclusion

Our data suggest that the calculated prevalence of childhood hypertension in our region is comparable to data from other European countries and is associated with early metabolic disturbances. Data mining is an effective method for identifying childhood hypertension and its metabolic consequences.

Diagnostic value of soluble biomarkers for parapneumonic pleural effusion

Parapneumonic pleural effusion (PPE) is a common complication in patients with pneumonia. Timely and accurate diagnosis of PPE is of great value for its management. Measurement of biomarkers in circulating and pleural fluid have the advantages of easy accessibility, short turn-around time, objectiveness and low cost and thus have utility for PPE diagnosis and stratification. To date, many biomarkers have been reported to be of value for the management of PPE. Here, we review the values of pleural fluid and circulating biomarkers for the diagnosis and stratification PPE. The biomarkers discussed are C-reactive protein, procalcitonin, presepsin, soluble triggering receptor expressed on myeloid cells 1, lipopolysaccharide-binding protein, inflammatory markers, serum amyloid A, soluble urokinase plasminogen activator receptor, matrix metalloproteinases, pentraxin-3 and cell-free DNA. We found that none of the available biomarkers has adequate performance for diagnosing and stratifying PPE. Therefore, further work is needed to identify and validate novel biomarkers, and their combinations, for the management of PPE.

Development and validation of a machine learning model for differential diagnosis of malignant pleural effusion using routine laboratory data

BACKGROUND

The differential diagnosis of malignant pleural effusion (MPE) and benign pleural effusion (BPE) presents a clinical challenge. In recent years, the use of artificial intelligence (AI) machine learning models for disease diagnosis has increased.

OBJECTIVE

This study aimed to develop and validate a diagnostic model for early differentiation between MPE and BPE based on routine laboratory data.

DESIGN

This was a retrospective observational cohort study.

METHODS

A total of 2352 newly diagnosed patients with pleural effusion (PE), between January 2008 and March 2021, were eventually enrolled. Among them, 1435, 466, and 451 participants were randomly assigned to the training, validation, and testing cohorts in a ratio of 3:1:1. Clinical parameters, including age, sex, and laboratory parameters of PE patients, were abstracted for analysis. Based on 81 candidate laboratory variables, five machine learning models, namely extreme gradient boosting (XGBoost) model, logistic regression (LR) model, random forest (RF) model, support vector machine (SVM) model, and multilayer perceptron (MLP) model were developed. Their respective diagnostic performances for MPE were evaluated by receiver operating characteristic (ROC) curves.

RESULTS

Among the five models, the XGBoost model exhibited the best diagnostic performance for MPE (area under the curve (AUC): 0.903, 0.918, and 0.886 in the training, validation, and testing cohorts, respectively). Additionally, the XGBoost model outperformed carcinoembryonic antigen (CEA) levels in pleural fluid (PF), serum, and the PF/serum ratio (AUC: 0.726, 0.699, and 0.692 in the training cohort; 0.763, 0.695, and 0.731 in the validation cohort; and 0.722, 0.729, and 0.693 in the testing cohort, respectively). Furthermore, compared with CEA, the XGBoost model demonstrated greater diagnostic power and sensitivity in diagnosing lung cancer-induced MPE.

CONCLUSION

The development of a machine learning model utilizing routine laboratory biomarkers significantly enhances the diagnostic capability for distinguishing between MPE and BPE. The XGBoost model emerges as a valuable tool for the diagnosis of MPE.

Pleural fluid biochemical analysis: the past, present and future

Identifying the cause of pleural effusion is challenging for pulmonologists. Imaging, biopsy, microbiology and biochemical analyses are routinely used for diagnosing pleural effusion. Among these diagnostic tools, biochemical analyses are promising because they have the advantages of low cost, minimal invasiveness, observer independence and short turn-around time. Here, we reviewed the past, present and future of pleural fluid biochemical analysis. We reviewed the history of Light’s criteria and its modifications and the current status of biomarkers for heart failure, malignant pleural effusion, tuberculosis pleural effusion and parapneumonic pleural effusion. In addition, we anticipate the future of pleural fluid biochemical analysis, including the utility of machine learning, molecular diagnosis and high-throughput technologies. Clinical Chemistry and Laboratory Medicine (CCLM) should address the topic of pleural fluid biochemical analysis in the future to promote specific knowledge in the laboratory professional community.

Potential prognostic biomarkers of sudden cardiac death discovered by machine learning

OBJECTIVE

Sudden cardiac death (SCD) is a serious public health burden. This study aims to find prognostic biomarkers of SCD using machine learning.

METHODS

The myocardial samples from 21 accidental death and 82 sudden death donors were compared to seek for differential genes. Enriched active genes were found according to the PPI interaction network. GSEA analyzed differences in function and pathway between control and experimental groups. Related diseases caused by active genes are mainly exhibited through DO enrichment. Prognostic biomarkers for SCD are identified via two machine learning algorithms. The CIBERSORT method was used to compare the immune microenvironment changes in patients with SCD.

RESULTS

SCD was mainly associated with heart and kidney diseases caused by atherosclerosis. DEFA1B, BGN, SERPINE1, CCL2 and HBB are considered to be prognostic biomarkers for SCD after machine learning. And immune infiltration plays an important role in the process of SCD.

CONCLUSION

We discovered 5 prognostic biomarkers for SCD. And immune microenvironment changes was also found in SCD. Moreover, atherosclerosis might be an important risk factor for SCD.

Using machine learning to find genes associated with sudden death

Objective

To search for significant biomarkers associated with sudden death (SD).

Methods

Differential genes were screened by comparing the whole blood samples from 15 cases of accidental death (AD) and 88 cases of SD. The protein-protein interaction (PPI) network selects core genes that interact most frequently. Machine learning is applied to find characteristic genes related to SD. The CIBERSORT method was used to explore the immune-microenvironment changes.

Results

A total of 10 core genes (MYL1, TNNC2, TNNT3, TCAP, TNNC1, TPM2, MYL2, TNNI1, ACTA1, CKM) were obtained and they were mainly related to myocarditis, hypertrophic myocarditis and dilated cardiomyopathy (DCM). Characteristic genes of MYL2 and TNNT3 associated with SD were established by machine learning. There was no significant change in the immune-microenvironment before and after SD.

Conclusion

Detecting characteristic genes is helpful to identify patients at high risk of SD and speculate the cause of death.

Deep Learning Model for Predicting Rhythm Outcomes after Radiofrequency Catheter Ablation in Patients with Atrial Fibrillation

Current guidelines on atrial fibrillation (AF) emphasized that radiofrequency catheter ablation (RFCA) should be decided after fully considering its prognosis. However, a robust prediction model reflecting the complex interactions between the features affecting prognosis remains to be developed. In this paper, we propose a deep learning model for predicting the late recurrence after RFCA in patients with AF. Aiming to predict the late recurrence (LR) of AF within 1 year after pulmonary vein isolation, we designed a multimodal model based on the multilayer perceptron architecture. For quantitative evaluation, we conducted 4-fold cross-validation on data from 177 AF patients including 47 LR patients. The proposed model (area under the receiver operating characteristic curve-AUROC, 0.766) outperformed the acute patient physiologic and laboratory evaluation (APPLE) score (AUROC, 0.605), CHA2DS2-VASc score (AUROC, 0.595), linear regression (AUROC, 0.541), logistic regression (AUROC, 0.546), extreme gradient boosting (AUROC, 0.608), and support vector machine (AUROC, 0.638). The proposed model exhibited better performance than clinical indicators (APPLE and CHA2DS2-VASc score) and machine learning techniques (linear regression, logistic regression, extreme gradient boosting, and support vector machine). The model will support clinical decision-making for selecting good responders to the RFCA intervention.

[Characteristics of wideband tympanometry in patients with Ménière's disease based on neutral network]

Objective:To construct a prediction model for Ménière's disease based on neural network and evaluate its prediction ability. Methods:Sixty-four patients with Ménière's disease underwent gadolinium enhanced magnetic resonance imaging of inner ear which showed endolymphatic hydrops. Meanwhile, 40 healthy adults were enrolled as controls. The database of wideband tympanometry of patients and control subjects was analyzed, and the neural network model was established by MATLAB 2021a software. The prediction ability of the model was evaluated by accuracy, positive predictive value, negative predictive value, the Youden index, sensitivity, specificity, receiver operating characteristic curve and area under curve （AUC）. Results:A feedforward network model was built with a single hidden layer to predict Ménière's disease with wideband tympanometry. There were 104 features in the input layer, 13 neuron nodes in the hidden layer and 1 output neuron in the output layer. The accuracy of the model was 83.2%, the positive predictive value was 80.7%, the negative predictive value was 84.3%, the sensitivity was 76.5%, the specificity was 83.7%, the Youden index was 0.602, and the AUC was 0.855. Conclusion:Based on neural network, the prediction model of Ménière's disease with high accuracy was constructed according to the results of wideband tympanometry, which provided reference for the diagnose of Ménière's disease.

Prediction models used in the progression of chronic kidney disease: A scoping review

Objective

To provide a review of prediction models that have been used to measure clinical or pathological progression of chronic kidney disease (CKD).

Design

Scoping review.

Data sources

Medline, EMBASE, CINAHL and Scopus from the year 2011 to 17th February 2022.

Study selection

All English written studies that are published in peer-reviewed journals in any country, that developed at least a statistical or computational model that predicted the risk of CKD progression.

Data extraction

Eligible studies for full text review were assessed on the methods that were used to predict the progression of CKD. The type of information extracted included: the author(s), title of article, year of publication, study dates, study location, number of participants, study design, predicted outcomes, type of prediction model, prediction variables used, validation assessment, limitations and implications.

Results

From 516 studies, 33 were included for full-text review. A qualitative analysis of the articles was compared following the extracted information. The study populations across the studies were heterogenous and data acquired by the studies were sourced from different levels and locations of healthcare systems. 31 studies implemented supervised models, and 2 studies included unsupervised models. Regardless of the model used, the predicted outcome included measurement of risk of progression towards end-stage kidney disease (ESKD) of related definitions, over given time intervals. However, there is a lack of reporting consistency on details of the development of their prediction models.

Conclusions

Researchers are working towards producing an effective model to provide key insights into the progression of CKD. This review found that cox regression modelling was predominantly used among the small number of studies in the review. This made it difficult to perform a comparison between ML algorithms, more so when different validation methods were used in different cohort types. There needs to be increased investment in a more consistent and reproducible approach for future studies looking to develop risk prediction models for CKD progression.

“Big Data” Approaches for Prevention of the Metabolic Syndrome

Metabolic syndrome (MetS) is characterized by the concurrence of multiple metabolic disorders resulting in the increased risk of a variety of diseases related to disrupted metabolism homeostasis. The prevalence of MetS has reached a pandemic level worldwide. In recent years, extensive amount of data have been generated throughout the research targeted or related to the condition with techniques including high-throughput screening and artificial intelligence, and with these “big data”, the prevention of MetS could be pushed to an earlier stage with different data source, data mining tools and analytic tools at different levels. In this review we briefly summarize the recent advances in the study of “big data” applications in the three-level disease prevention for MetS, and illustrate how these technologies could contribute tobetter preventive strategies.

Simple Method to Predict Insulin Resistance in Children Aged 6-12 Years by Using Machine Learning

BACKGROUND

Due to the increasing insulin resistance (IR) in childhood, rates of diabetes and cardiovascular disease may rise in the future and seriously threaten the healthy development of children. Finding an easy way to predict IR in children can help pediatricians to identify these children in time and intervene appropriately, which is particularly important for practitioners in primary health care.

PATIENTS AND METHODS

Seventeen features from 503 children 6-12 years old were collected. We defined IR by HOMA-IR greater than 3.0, thus classifying children with IR and those without IR. Data were preprocessed by multivariate imputation and oversampling to resolve missing values and data imbalances; then, recursive feature elimination was applied to further select features of interest, and 5 machine learning methods-namely, logistic regression (LR), support vector machine (SVM), random forest (RF), extreme gradient boosting (XGBoost), and gradient boosting with categorical features support (CatBoost)-were used for model training. We tested the trained models on an external test set containing information from 133 children, from which performance metrics were extracted and the optimal model was selected.

RESULTS

After feature selection, the numbers of chosen features for the LR, SVM, RF, XGBoost, and CatBoost models were 6, 9, 10, 14, and 6, respectively. Among them, glucose, waist circumference, and age were chosen as predictors by most of the models. Finally, all 5 models achieved good performance on the external test set. Both XGBoost and CatBoost had the same AUC (0.85), which was highest among those of all models. Their accuracy, sensitivity, precision, and F1 scores were also close, but the specificity of XGBoost reached 0.79, which was significantly higher than that of CatBoost, so XGBoost was chosen as the optimal model.

CONCLUSION

The model developed herein has a good predictive ability for IR in children 6-12 years old and can be clinically applied to help pediatricians identify children with IR in a simple and inexpensive way.

A pilot study of the effects of crocin on high-density lipoprotein cholesterol uptake capacity in patients with metabolic syndrome: A randomized clinical trial

A randomized clinical trial high-density lipoprotein (HDL) cholesterol uptake capacity (CUC) is reduced in patients with metabolic syndrome (MetS). We have assessed the effect of crocin supplementation on HDL CUC in patients with MetS. Forty-four subjects with MetS were randomly allocated to one of two groups: one group received placebo and the other group received crocin at a dose of 30 mg (two tablets of 15 mg per day) for 8 weeks. Serum biochemical parameters were measured using an AutoAnalyzer BT3000 (BioTechnica). The modified CUC method is a cell free, simple, and high-throughput assay that used to evaluate HDL CUC of serum samples. The decision tree analysis was undertaken using JMP Pro (SAS) version 13. The mean age of the crocin and placebo groups were 38.97 ± 13.33 and 43.46 ± 12.77 years, respectively. There was a significant increase in serum HDL CUC in the crocin group compared to that of the placebo group in patients with MetS (p-value< 0.05). The decision tree analysis showed that serum HDL functionality was more important variable than HDL-C level in predicting patients with hypertension at baseline (p-value < 0.05). Crocin administration (30 mg for a period of 8 weeks) was found to improve serum HDL CUC in patients with MetS. TRIAL REGISTRATION: IRCT2013080514279N1.

Factors determining the serum 25-hydroxyvitamin D response to vitamin D supplementation: Data mining approach

Vitamin D supplementation has been shown to prevent vitamin D deficiency, but various factors can affect the response to supplementation. Data mining is a statistical method for pulling out information from large databases. We aimed to evaluate the factors influencing serum 25-hydroxyvitamin D levels in response to supplementation of vitamin D using a random forest (RF) model. Data were extracted from the survey of ultraviolet intake by nutritional approach study. Vitamin D levels were measured at baseline and at the end of study to evaluate the responsiveness. We examined the relationship between 76 potential influencing factors on vitamin D response using RF. We found several features that were highly correlated to the serum vitamin D response to supplementation by RF including anthropometric factors (body mass index [BMI], free fat mass [FFM], fat percentage, waist-to-hip ratio [WHR]), liver function tests (serum gamma-glutamyl transferase [GGT], total bilirubin, total protein), hematological parameters (mean corpuscular volume [MCV], mean corpuscular hemoglobin concentration [MCHC], hematocrit), and measurement of insulin sensitivity (homeostatic model assessment of insulin resistance). BMI, total bilirubin, FFM, and GGT were found to have a positive relationship and homeostatic model assessment for insulin resistance, MCV, MCHC, fat percentage, total protein, and WHR were found to have a negative correlation to vitamin D concentration in response to supplementation. The accuracy of RF in predicting the response was 93% compared to logistic regression, for which the accuracy was 40%, in the evaluation of the correlation of the components of the data set to serum vitamin D.