MACHINE LEARNING FOR DETECTION AND DIAGNOSIS OF DISEASE

Designing nanotheranostics with machine learning

The inherent limits of traditional diagnoses and therapies have driven the development and application of emerging nanotechnologies for more effective and safer management of diseases, herein referred to as 'nanotheranostics'. Although many important technological successes have been achieved in this field, widespread adoption of nanotheranostics as a new paradigm is hindered by specific obstacles, including time-consuming synthesis of nanoparticles, incomplete understanding of nano-bio interactions, and challenges regarding chemistry, manufacturing and the controls required for clinical translation and commercialization. As a key branch of artificial intelligence, machine learning (ML) provides a set of tools capable of performing time-consuming and result-perception tasks, thus offering unique opportunities for nanotheranostics. This Review summarizes the progress and challenges in this emerging field of ML-aided nanotheranostics, and discusses the opportunities in developing next-generation nanotheranostics with reliable datasets and advanced ML models to offer better clinical benefits to patients.

Diagnostic value of the Sirtuins family in acute rejection of kidney transplantation assessed on the basis of transcriptomics and animal experiments

BACKGROUND

The Sirtuins (SIRT) family plays a key role in the diagnosis and treatment of many renal diseases, but no studies have been reported in acute rejection of kidney transplantation. The aim of this study was to explore the diagnostic value of SIRT family change characteristics in acute rejection of kidney transplantation.

METHODS

We first explored the SIRT family expression profile in renal tissues using the HPA database; subsequently, we explored the potential biological functions and mechanistic changes during acute rejection of kidney transplantation by GSEA enrichment analysis. The Cibersort algorithm specifies the level of immune cell infiltration and explores the correlation between the SIRT family and immune cells using correlation analysis; Next, we constructed a diagnostic model using "Logistic regression analysis" and "Nomogram model", and evaluated the diagnostic model using calibration curves and ROC curves, and the decision curve (DCA) was used to evaluate the clinical diagnostic value of SIRT family changes; Finally, we constructed a model of acute rejection of rat kidney transplantation, and assessed rat kidney function by detecting the levels of urea nitrogen and creatinine in serum. Meanwhile, the expression level of SIRT family in kidney tissues was initially verified by transcriptome sequencing and RT-PCR.

RESULTS

We found that all seven SIRT family members were located and expressed in renal tissues. The results of enrichment analysis revealed that a large number of immune-related biological functions and pathways are activated during acute rejection of kidney transplantation, the difference was statistically significant (p < 0.05). The Cibersort algorithm revealed significant changes in the level of infiltration of 10 immune cells (p < 0.05), while correlation analysis revealed a strong link between the SIRT family and immune cells (p < 0.05). We constructed a diagnostic model for acute rejection using seven SIRT families, and the ROC curves(AUC = 0.71)and calibration curves proved their good diagnostic value, and the DCA curves also proved the role of SIRT families in clinical decision-making. Next, we again demonstrated the good diagnostic performance of the SIRT family in ABMR and TCMR, respectively(ROC curves:AUC = 0.64，AUC = 0.81). Finally, in a rat model of acute rejection of kidney transplantation, we found that renal function (BUN and creatinine) was significantly impaired in rats in the Allo group compared to rats in the Syn group (P < 0.05). Meanwhile, by transcriptome analysis and RT-PCR assay, we found that, except for SIRT1, the remaining SIRT family members were significantly changed in kidney tissues (P < 0.05).

CONCLUSION

The SIRT family has significant changes during acute rejection in kidney transplantation, and the SIRT family may be able to serve as a potential therapeutic target for alleviating acute rejection in kidney transplantation.

Deep learning-based Alzheimer's disease detection: reproducibility and the effect of modeling choices

Introduction

Machine Learning (ML) has emerged as a promising approach in healthcare, outperforming traditional statistical techniques. However, to establish ML as a reliable tool in clinical practice, adherence to best practices in data handling, and modeling design and assessment is crucial. In this work, we summarize and strictly adhere to such practices to ensure reproducible and reliable ML. Specifically, we focus on Alzheimer's Disease (AD) detection, a challenging problem in healthcare. Additionally, we investigate the impact of modeling choices, including different data augmentation techniques and model complexity, on overall performance.

Methods

We utilize Magnetic Resonance Imaging (MRI) data from the ADNI corpus to address a binary classification problem using 3D Convolutional Neural Networks (CNNs). Data processing and modeling are specifically tailored to address data scarcity and minimize computational overhead. Within this framework, we train 15 predictive models, considering three different data augmentation strategies and five distinct 3D CNN architectures with varying convolutional layers counts. The augmentation strategies involve affine transformations, such as zoom, shift, and rotation, applied either concurrently or separately.

Results

The combined effect of data augmentation and model complexity results in up to 10% variation in prediction accuracy. Notably, when affine transformation are applied separately, the model achieves higher accuracy, regardless the chosen architecture. Across all strategies, the model accuracy exhibits a concave behavior as the number of convolutional layers increases, peaking at an intermediate value. The best model reaches excellent performance both on the internal and additional external testing set.

Discussions

Our work underscores the critical importance of adhering to rigorous experimental practices in the field of ML applied to healthcare. The results clearly demonstrate how data augmentation and model depth-often overlooked factors- can dramatically impact final performance if not thoroughly investigated. This highlights both the necessity of exploring neglected modeling aspects and the need to comprehensively report all modeling choices to ensure reproducibility and facilitate meaningful comparisons across studies.

Electrochemical platform for detecting Escherichia coli bacteria using machine learning methods

We present an electrochemical platform designed to reduce time of Escherichia coli bacteria detection from 24 to 48-h to 30 min. The presented approach is based on a system which includes gallium-indium (eGaIn) alloy to provide conductivity and a hydrogel system to preserve bacteria and their metabolic species during the analysis. The work is dedicated to accurate and fast detection of Escherichia coli bacteria in different environments with the supply of machine learning methods. Electrochemical data obtained during the analysis is processed via multilayer perceptron model to identify i.e. predict bacterial concentration in the samples. The performed approach provides the effectiveness of bacteria identification in the range of 102-109 colony forming units per ml with the average accuracy of 97%. The proposed bioelectrochemical system combined with machine learning model is prospective for food analysis, agriculture, biomedicine.

An integrated machine learning framework for developing and validating a diagnostic model of major depressive disorder based on interstitial cystitis-related genes

BACKGROUND

Major depressive disorder (MDD) and interstitial cystitis (IC) are two highly debilitating conditions that often coexist with reciprocal effect, significantly exacerbating patients' suffering. However, the molecular underpinnings linking these disorders remain poorly understood.

METHODS

Transcriptomic data from GEO datasets including those of MDD and IC patients was systematically analyzed to develop and validate our model. Following removal of batch effect, differentially expressed genes (DEGs) between respective disease and control groups were identified. Shared DEGs of the conditions then underwent functional enrichment analyses. Additionally, immune infiltration analysis was quantified through ssGSEA. A diagnostic model for MDD was constructed by exploring 113 combinations of 12 machine learning algorithms with 10-fold cross-validation on the training sets following by external validation on test sets. Finally, the "Enrichr" platform was utilized to identify potential drugs for MDD.

RESULTS

Totally, 21 key genes closely associated with both MDD and IC were identified, predominantly involved in immune processes based on enrichment analyses. Immune infiltration analysis revealed distinct profiles of immune cell infiltration in MDD and IC compared to healthy controls. From these genes, a robust 11-gene (ABCD2, ATP8B4, TNNT1, AKR1C3, SLC26A8, S100A12, PTX3, FAM3B, ITGA2B, OLFM4, BCL7A) diagnostic signature was constructed, which exhibited superior performance over existing MDD diagnostic models both in training and testing cohorts. Additionally, epigallocatechin gallate and 10 other drugs emerged as potential targets for MDD.

CONCLUSION

Our work developed a diagnostic model for MDD employing a combination of bioinformatic techniques and machine learning methods, focusing on shared genes between MDD and IC.

Transcriptomics-based exploration of shared M1-type macrophage-related biomarker in acute kidney injury after kidney transplantation and acute rejection after kidney transplantation

BACKGROUND

Macrophage type 1 (M1) cells are associated with both acute kidney injury (AKI) during kidney transplantation and acute rejection (AR) after kidney transplantation. Our study explored M1-related biomarkers involved in both AKI and AR and their potential biological functions.

METHODS

Based on the Gene Expression Omnibus (GEO) database, the immune cell infiltration levels and differentially expressed genes were examined in AKI and AR in the kidney transplantation; M1-related genes shared in AKI and AR were identified using weighted gene co-expression analysis (WGCNA) system. Subsequently, protein-protein interaction (PPI) networks and machine learning methods to identify Hub genes and construct diagnostic models. Both AKI model and AR rat models were built to validate the expressions of Hub genes and test the injury phenotype, oxidative stress markers, and inflammatory factors. Finally, the transcription factor (TF)-Hub gene and micro-RNA (miRNA)-Hub gene regulatory networks were constructed based on identified Hub genes.

RESULTS

Out of 2167 differential expression genes (DEGs) in AKI and 2100 DEGs in AR, four M1-related Hub genes were obtained by PPI networks and machine learning methods, namely GBP2, TYROBP, CCR5, and TLR8. The calibration curves in the nomogram diagnostic model for these four Hub genes suggested the same predictive probability as an ideal model for AKI and AR after kidney transplantation (AUC values of the area under the ROC curve were all >0.7). The same observations were confirmed in ischemia reperfusion injury (IRI) and AR rat models by identifying common four Hub genes (GBP2, TYROBP, TLR8, and CCR5). Western blots showed that these four Hub genes were significantly different in rat models of IRI and AR (all p<0.05). Compared with the control group, IRI and AR groups showed aggravated histopathological damage and increased secretion of oxidative stress markers and inflammatory factors in rat kidneys (all p<0.05). Finally, TF-Hub and miRNA-Hub gene regulatory networks were constructed to provide a theoretical basis for the regulation of Hub genes.

CONCLUSION

We identified four macrophage M1-related Hub genes shared among AKI and AR after kidney transplantation. These genes may be considered for diagnosis of AKI and AR after kidney transplantation.

Identification of diagnostic genes and drug prediction in metabolic syndrome-associated rheumatoid arthritis by integrated bioinformatics analysis, machine learning, and molecular docking

Background

Interactions between the immune and metabolic systems may play a crucial role in the pathogenesis of metabolic syndrome-associated rheumatoid arthritis (MetS-RA). The purpose of this study was to discover candidate biomarkers for the diagnosis of RA patients who also had MetS.

Methods

Three RA datasets and one MetS dataset were obtained from the Gene Expression Omnibus (GEO) database. Differential expression analysis, weighted gene co-expression network analysis (WGCNA), and machine learning algorithms including Least Absolute Shrinkage and Selection Operator (LASSO) regression and Random Forest (RF) were employed to identify hub genes in MetS-RA. Enrichment analysis was used to explore underlying common pathways between MetS and RA. Receiver operating characteristic curves were applied to assess the diagnostic performance of nomogram constructed based on hub genes. Protein-protein interaction, Connectivity Map (CMap) analyses, and molecular docking were utilized to predict the potential small molecule compounds for MetS-RA treatment. qRT-PCR was used to verify the expression of hub genes in fibroblast-like synoviocytes (FLS) of MetS-RA. The effects of small molecule compounds on the function of RA-FLS were evaluated by wound-healing assays and angiogenesis experiments. The CIBERSORT algorithm was used to explore immune cell infiltration in MetS and RA.

Results

MetS-RA key genes were mainly enriched in immune cell-related signaling pathways and immune-related processes. Two hub genes (TYK2 and TRAF2) were selected as candidate biomarkers for developing nomogram with ideal diagnostic performance through machine learning and proved to have a high diagnostic value (area under the curve, TYK2, 0.92; TRAF2, 0.90). qRT-PCR results showed that the expression of TYK2 and TRAF2 in MetS-RA-FLS was significantly higher than that in non-MetS-RA-FLS (nMetS-RA-FLS). The combination of CMap analysis and molecular docking predicted camptothecin (CPT) as a potential drug for MetS-RA treatment. In vitro validation, CPT was observed to suppress the cell migration capacity and angiogenesis capacity of MetS-RA-FLS. Immune cell infiltration results revealed immune dysregulation in MetS and RA.

Conclusion

Two hub genes were identified in MetS-RA, a nomogram for the diagnosis of RA and MetS was established based on them, and a potential therapeutic small molecule compound for MetS-RA was predicted, which offered a novel research perspective for future serum-based diagnosis and therapeutic intervention of MetS-RA.

Detection of lung cancer through SERS analysis of serum

Early detection and postoperative assessment are crucial for improving overall survival among lung cancer patients. Here, we report a non-invasive technique that integrates Raman spectroscopy with machine learning for the detection of lung cancer. The study encompassed 88 postoperative lung cancer patients, 73 non-surgical lung cancer patients, and 68 healthy subjects. The primary aim was to explore variations in serum metabolism across these cohorts. Comparative analysis of average Raman spectra was conducted, while principal component analysis was employed for data visualization. Subsequently, the augmented dataset was used to train convolutional neural networks (CNN) and Resnet models, leading to the development of a diagnostic framework. The CNN model exhibited superior performance, as verified by the receiver operating characteristic curve. Notably, postoperative patients demonstrated an increased likelihood of recurrence, emphasizing the crucial need for continuous postoperative monitoring. In summary, the integration of Raman spectroscopy with CNN-based classification shows potential for early detection and postoperative assessment of lung cancer.

Development of a machine learning-based model to predict prognosis of alpha-fetoprotein-positive hepatocellular carcinoma

BACKGROUND

Patients with alpha-fetoprotein (AFP)-positive hepatocellular carcinoma (HCC) have aggressive biological behavior and poor prognosis. Therefore, survival time is one of the greatest concerns for patients with AFP-positive HCC. This study aimed to demonstrate the utilization of six machine learning (ML)-based prognostic models to predict overall survival of patients with AFP-positive HCC.

METHODS

Data on patients with AFP-positive HCC were extracted from the Surveillance, Epidemiology, and End Results database. Six ML algorithms (extreme gradient boosting [XGBoost], logistic regression [LR], support vector machine [SVM], random forest [RF], K-nearest neighbor [KNN], and decision tree [ID3]) were used to develop the prognostic models of patients with AFP-positive HCC at one year, three years, and five years. Area under the receiver operating characteristic curve (AUC), confusion matrix, calibration curves, and decision curve analysis (DCA) were used to evaluate the model.

RESULTS

A total of 2,038 patients with AFP-positive HCC were included for analysis. The 1-, 3-, and 5-year overall survival rates were 60.7%, 28.9%, and 14.3%, respectively. Seventeen features regarding demographics and clinicopathology were included in six ML algorithms to generate a prognostic model. The XGBoost model showed the best performance in predicting survival at 1-year (train set: AUC = 0.771; test set: AUC = 0.782), 3-year (train set: AUC = 0.763; test set: AUC = 0.749) and 5-year (train set: AUC = 0.807; test set: AUC = 0.740). Furthermore, for 1-, 3-, and 5-year survival prediction, the accuracy in the training and test sets was 0.709 and 0.726, 0.721 and 0.726, and 0.778 and 0.784 for the XGBoost model, respectively. Calibration curves and DCA exhibited good predictive performance as well.

CONCLUSIONS

The XGBoost model exhibited good predictive performance, which may provide physicians with an effective tool for early medical intervention and improve the survival of patients.

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.

Identification of CXC Chemokine Receptor 2 (CXCR2) as a Novel Eosinophils-Independent Diagnostic Biomarker of Pediatric Eosinophilic Esophagitis by Integrated Bioinformatic and Machine-Learning Analysis

Background

Eosinophilic esophagitis (EoE) is a complex allergic condition frequently accompanied by various atopic comorbidities in children, which significantly affects their life qualities. Therefore, this study aimed to evaluate pivotal molecular markers that may facilitate the diagnosis of EoE in pediatric patients.

Methods

Three available EoE-associated gene expression datasets in children: GSE184182, GSE 197702, GSE55794, along with GSE173895 were downloaded from the GEO database. Differentially expressed genes (DEGs) identified by "limma" were intersected with key module genes identified by weighted gene co-expression network analysis (WGCNA), and the shared genes went through functional enrichment analysis. The protein-protein interaction (PPI) network and the machine learning algorithms: least absolute shrinkage and selection operator (LASSO), random forest (RF), and XGBoost were used to reveal candidate diagnostic markers for EoE. The receiver operating characteristic (ROC) curve showed the efficacy of differential diagnosis of this marker, along with online databases predicting its molecular regulatory network. Finally, we performed gene set enrichment analysis (GSEA) and assessed immune cell infiltration of EoE/control samples by using the CIBERSORT algorithm. The correlations between the key diagnostic biomarker and immune cells were also investigated.

Results

The intersection of 936 DEGs and 1446 key module genes in EoE generated 567 genes, which were primarily enriched in immune regulation. Following the construction of the PPI network and filtration by machine learning, CXCR2 served as a potential diagnostic biomarker of pediatric EoE with a perfect diagnostic efficacy (AUC = ~1.00) in regional tissue/peripheral whole blood samples. Multiple infiltrated immune cells were observed to participate in disrupting the homeostasis of esophageal epithelium to varying degrees.

Conclusion

The immune-correlated CXCR2 gene was proved to be a promising diagnostic indicator for EoE, and dysregulated regulatory T cells (Tregs)/neutrophils might play a crucial role in the pathogenesis of EoE in children.

Machine learning in onco-pharmacogenomics: a path to precision medicine with many challenges

Over the past two decades, Next-Generation Sequencing (NGS) has revolutionized the approach to cancer research. Applications of NGS include the identification of tumor specific alterations that can influence tumor pathobiology and also impact diagnosis, prognosis and therapeutic options. Pharmacogenomics (PGx) studies the role of inheritance of individual genetic patterns in drug response and has taken advantage of NGS technology as it provides access to high-throughput data that can, however, be difficult to manage. Machine learning (ML) has recently been used in the life sciences to discover hidden patterns from complex NGS data and to solve various PGx problems. In this review, we provide a comprehensive overview of the NGS approaches that can be employed and the different PGx studies implicating the use of NGS data. We also provide an excursus of the ML algorithms that can exert a role as fundamental strategies in the PGx field to improve personalized medicine in cancer.

Special Issue "Machine Learning Methods for Biomedical Data Analysis"

Machine learning is an effective method for developing automatic algorithms for analysing sophisticated biomedical data [...].

Protecting patient safety and privacy in the era of artificial intelligence

The promise of artificial intelligence (AI) in health care has propelled a significant uptrend in the number of clinical trials in AI and global market spending in this novel technology. In vascular surgery, this technology has the ability to diagnose disease, predict disease outcomes, and assist with image-guided surgery. As we enter an era of rapid change, it is critical to evaluate the ethical concerns of AI, particularly as it may impact patient safety and privacy. This is particularly important to discuss in the early stages of AI, as technology frequently outpaces the policies and ethical guidelines regulating it. Issues at the forefront include patient privacy and confidentiality, protection of patient autonomy and informed consent, accuracy and applicability of this technology, and propagation of health care disparities. Vascular surgeons should be equipped to work with AI, as well as discuss its novel risks to patient safety and privacy.

Risk factors and prognostic nomogram for patients with second primary cancers after lung cancer using classical statistics and machine learning

Previous studies have revealed an increased risk of secondary primary cancers (SPC) after lung cancer. The prognostic prediction models for SPC patients after lung cancer are particularly needed to guide screening. Therefore, we study retrospectively analyzed the Surveillance, Epidemiology, and End Results (SEER) database using classical statistics and machine learning to explore the risk factors and construct a novel overall survival (OS) prediction nomogram for patients with SPC after lung cancer. Data of patients with SPC after lung cancer, covering 2000 to 2016, were gathered from the SEER database. The incidence of SPC after lung cancer was calculated by Standardized incidence ratios (SIRs). Cox proportional hazards regression, machine learning (ML), Kaplan-Meier (KM) methods, and log-rank tests were conducted to identify the important prognostic factors for predicting OS. These significant prognostic factors were used for the development of an OS prediction nomogram. Totally, 10,487 SPC samples were randomly divided into training and validation cohorts (model construction and internal validation) from the SEER database. In the random forest (RF) and extreme gradient boosting (XGBoost) feature importance ranking models, age was the most important variable which was also reflected in the nomogram. And, the models that combined machine learning with cox proportional hazards had a better predictive performance than the model that only used cox proportional hazards (AUC = 0.762 in RF, AUC = 0.737 in XGBoost, AUC = 0.722 in COX). Calibration curves and decision curve analysis (DCA) curves also revealed that our nomogram has excellent clinical utility. The web-based dynamic nomogram calculator was accessible on https://httseer.shinyapps.io/DynNomapp/ . The prognosis characteristics of SPC following lung cancer were systematically reviewed. The dynamic nomogram we constructed can provide survival predictions to assist clinicians in making individualized decisions.

Prediction performance of the machine learning model in predicting mortality risk in patients with traumatic brain injuries: a systematic review and meta-analysis

PURPOSE

With the in-depth application of machine learning(ML) in clinical practice, it has been used to predict the mortality risk in patients with traumatic brain injuries(TBI). However, there are disputes over its predictive accuracy. Therefore, we implemented this systematic review and meta-analysis, to explore the predictive value of ML for TBI.

METHODOLOGY

We systematically retrieved literature published in PubMed, Embase.com, Cochrane, and Web of Science as of November 27, 2022. The prediction model risk of bias(ROB) assessment tool (PROBAST) was used to assess the ROB of models and the applicability of reviewed questions. The random-effects model was adopted for the meta-analysis of the C-index and accuracy of ML models, and a bivariate mixed-effects model for the meta-analysis of the sensitivity and specificity.

RESULT

A total of 47 papers were eligible, including 156 model, with 122 newly developed ML models and 34 clinically recommended mature tools. There were 98 ML models predicting the in-hospital mortality in patients with TBI; the pooled C-index, sensitivity, and specificity were 0.86 (95% CI: 0.84, 0.87), 0.79 (95% CI: 0.75, 0.82), and 0.89 (95% CI: 0.86, 0.92), respectively. There were 24 ML models predicting the out-of-hospital mortality; the pooled C-index, sensitivity, and specificity were 0.83 (95% CI: 0.81, 0.85), 0.74 (95% CI: 0.67, 0.81), and 0.75 (95% CI: 0.66, 0.82), respectively. According to multivariate analysis, GCS score, age, CT classification, pupil size/light reflex, glucose, and systolic blood pressure (SBP) exerted the greatest impact on the model performance.

CONCLUSION

According to the systematic review and meta-analysis, ML models are relatively accurate in predicting the mortality of TBI. A single model often outperforms traditional scoring tools, but the pooled accuracy of models is close to that of traditional scoring tools. The key factors related to model performance include the accepted clinical variables of TBI and the use of CT imaging.

Novel models by machine learning to predict prognosis of breast cancer brain metastases

BACKGROUND

Breast cancer brain metastases (BCBM) are the most fatal, with limited survival in all breast cancer distant metastases. These patients are deemed to be incurable. Thus, survival time is their foremost concern. However, there is a lack of accurate prediction models in the clinic. What's more, primary surgery for BCBM patients is still controversial.

METHODS

The data used for analysis in this study was obtained from the SEER database (2010-2019). We made a COX regression analysis to identify prognostic factors of BCBM patients. Through cross-validation, we constructed XGBoost models to predict survival in patients with BCBM. Meanwhile, a BCBM cohort from our hospital was used to validate our models. We also investigated the prognosis of patients treated with surgery or not, using propensity score matching and K-M survival analysis. Our results were further validated by subgroup COX analysis in patients with different molecular subtypes.

RESULTS

The XGBoost models we created had high precision and correctness, and they were the most accurate models to predict the survival of BCBM patients (6-month AUC = 0.824, 1-year AUC = 0.813, 2-year AUC = 0.800 and 3-year survival AUC = 0.803). Moreover, the models still exhibited good performance in an externally independent dataset (6-month: AUC = 0.820; 1-year: AUC = 0.732; 2-year: AUC = 0.795; 3-year: AUC = 0.936). Then we used Shiny-Web tool to make our models be easily used from website. Interestingly, we found that the BCBM patients with an annual income of over USD$70,000 had better BCSS (HR = 0.523, 95%CI 0.273-0.999, P < 0.05) than those with less than USD$40,000. The results showed that in all distant metastasis sites, only lung metastasis was an independent poor prognostic factor for patients with BCBM (OS: HR = 1.606, 95%CI 1.157-2.230, P < 0.01; BCSS: HR = 1.698, 95%CI 1.219-2.365, P < 0.01), while bone, liver, distant lymph nodes and other metastases were not. We also found that surgical treatment significantly improved both OS and BCSS in BCBM patients with the HER2 + molecular subtypes and was beneficial to OS of the HR-/HER2- subtype. In contrast, surgery could not help BCBM patients with HR + /HER2- subtype improve their prognosis (OS: HR = 0.887, 95%CI 0.608-1.293, P = 0.510; BCSS: HR = 0.909, 95%CI 0.604-1.368, P = 0.630).

CONCLUSION

We analyzed the clinical features of BCBM patients and constructed 4 machine-learning prognostic models to predict their survival. Our validation results indicate that these models should be highly reproducible in patients with BCBM. We also identified potential prognostic factors for BCBM patients and suggested that primary surgery might improve the survival of BCBM patients with HER2 + and triple-negative subtypes.

Screening of immune-related secretory proteins linking chronic kidney disease with calcific aortic valve disease based on comprehensive bioinformatics analysis and machine learning

BACKGROUND

Chronic kidney disease (CKD) is one of the most significant cardiovascular risk factors, playing vital roles in various cardiovascular diseases such as calcific aortic valve disease (CAVD). We aim to explore the CKD-associated genes potentially involving CAVD pathogenesis, and to discover candidate biomarkers for the diagnosis of CKD with CAVD.

METHODS

Three CAVD, one CKD-PBMC and one CKD-Kidney datasets of expression profiles were obtained from the GEO database. Firstly, to detect CAVD key genes and CKD-associated secretory proteins, differentially expressed analysis and WGCNA were carried out. Protein-protein interaction (PPI), functional enrichment and cMAP analyses were employed to reveal CKD-related pathogenic genes and underlying mechanisms in CKD-related CAVD as well as the potential drugs for CAVD treatment. Then, machine learning algorithms including LASSO regression and random forest were adopted for screening candidate biomarkers and constructing diagnostic nomogram for predicting CKD-related CAVD. Moreover, ROC curve, calibration curve and decision curve analyses were applied to evaluate the diagnostic performance of nomogram. Finally, the CIBERSORT algorithm was used to explore immune cell infiltration in CAVD.

RESULTS

The integrated CAVD dataset identified 124 CAVD key genes by intersecting differential expression and WGCNA analyses. Totally 983 CKD-associated secretory proteins were screened by differential expression analysis of CKD-PBMC/Kidney datasets. PPI analysis identified two key modules containing 76 nodes, regarded as CKD-related pathogenic genes in CAVD, which were mostly enriched in inflammatory and immune regulation by enrichment analysis. The cMAP analysis exposed metyrapone as a more potential drug for CAVD treatment. 17 genes were overlapped between CAVD key genes and CKD-associated secretory proteins, and two hub genes were chosen as candidate biomarkers for developing nomogram with ideal diagnostic performance through machine learning. Furthermore, SLPI/MMP9 expression patterns were confirmed in our external cohort and the nomogram could serve as novel diagnosis models for distinguishing CAVD. Finally, immune cell infiltration results uncovered immune dysregulation in CAVD, and SLPI/MMP9 were significantly associated with invasive immune cells.

CONCLUSIONS

We revealed the inflammatory-immune pathways underlying CKD-related CAVD, and developed SLPI/MMP9-based CAVD diagnostic nomogram, which offered novel insights into future serum-based diagnosis and therapeutic intervention of CKD with CAVD.

Prognostic Models Using Machine Learning Algorithms and Treatment Outcomes of Occult Breast Cancer Patients

BACKGROUND

Occult breast cancer (OBC) is an uncommon malignant tumor and the prognosis and treatment of OBC remain controversial. Currently, there exists no accurate prognostic clinical model for OBC, and the treatment outcomes of chemotherapy and surgery in its different molecular subtypes are still unknown.

METHODS

The SEER database provided the data used for this study's analysis (2010-2019). To identify the prognostic variables for patients with ODC, we conducted Cox regression analysis and constructed prognostic models using six machine learning algorithms to predict overall survival (OS) of OBC patients. A series of validation methods, including calibration curve and area under the curve (AUC value) of receiver operating characteristic curve (ROC) were employed to validate the accuracy and reliability of the logistic regression (LR) models. The effectiveness of clinical application of the predictive models was validated using decision curve analysis (DCA). We also investigated the role of chemotherapy and surgery in OBC patients with different molecular subtypes, with the help of K-M survival analysis as well as propensity score matching, and these results were further validated by subgroup Cox analysis.

RESULTS

The LR models performed best, with high precision and applicability, and they were proved to predict the OS of OBC patients in the most accurate manner (test set: 1-year AUC = 0.851, 3-year AUC = 0.790 and 5-year survival AUC = 0.824). Interestingly, we found that the N1 and N2 stage OBC patients had more favorable prognosis than N0 stage patients, but the N3 stage was similar to the N0 stage (OS: N0 vs. N1, HR = 0.6602, 95%CI 0.4568-0.9542, p < 0.05; N0 vs. N2, HR = 0.4716, 95%CI 0.2351-0.9464, p < 0.05; N0 vs. N3, HR = 0.96, 95%CI 0.6176-1.5844, p = 0.96). Patients aged >80 and distant metastases were also independent prognostic factors for OBC. In terms of treatment, our multivariate Cox regression analysis discovered that surgery and radiotherapy were both independent protective variables for OBC patients, but chemotherapy was not. We also found that chemotherapy significantly improved both OS and breast cancer-specific survival (BCSS) only in the HR-/HER2+ molecular subtype (OS: HR = 0.15, 95%CI 0.037-0.57, p < 0.01; BCSS: HR = 0.027, 95%CI 0.027-0.81, p < 0.05). However, surgery could help only the HR-/HER2+ and HR+/HER2- subtypes improve prognosis.

CONCLUSIONS

We analyzed the clinical features and prognostic factors of OBC patients; meanwhile, machine learning prognostic models with high precision and applicability were constructed to predict their overall survival. The treatment results in different molecular subtypes suggested that primary surgery might improve the survival of HR+/HER2- and HR-/HER2+ subtypes, however, only the HR-/HER2+ subtype could benefit from chemotherapy. The necessity of surgery and chemotherapy needs to be carefully considered for OBC patients with other subtypes.

Multifeature Fusion Classification Method for Adaptive Endoscopic Ultrasonography Tumor Image

Endoscopic ultrasonography (EUS) has been found to be of great advantage in the diagnosis of digestive tract submucosal tumors. However, EUS-based diagnosis is limited by variability in subjective interpretation on the part of doctors. Tumor classification of ultrasound images with the computer-aided diagnosis system can significantly improve the diagnostic efficiency and accuracy of doctors. In this study, we proposed a multifeature fusion classification method for adaptive EUS tumor images. First, for different ultrasound tumor images, we selected the region of interest based on prior information to facilitate the estimation in the subsequent works. Second, we proposed a method based on image gray histogram feature extraction with principal component analysis dimensionality reduction, which learns the gray distribution of different tumor images effectively. Third, we fused the reduced grayscale features with the improved local binary pattern features and gray-level co-occurrence matrix features, and then used the multiclassification support vector machine. Finally, in the experiment, we selected the 431 ultrasound images of 109 patients in the hospital and compared the experimental effects of different features and different classifiers. The results revealed that the proposed method performed best, with the highest accuracy of 96.18% and an area under the curve of 99%. It is evident that the method proposed in this study can efficiently contribute to the classification of EUS tumor images.

Rapid diagnosis of membranous nephropathy based on serum and urine Raman spectroscopy combined with deep learning methods

Membranous nephropathy is the main cause of nephrotic syndrome, which has an insidious onset and may progress to end-stage renal disease with a high mortality rate, such as renal failure and uremia. At present, the diagnosis of membranous nephropathy mainly relies on the clinical manifestations of patients and pathological examination of kidney biopsy, which are expensive, time-consuming, and have certain chance and other disadvantages. Therefore, there is an urgent need to find a rapid, accurate and non-invasive diagnostic technique for the diagnosis of membranous nephropathy. In this study, Raman spectra of serum and urine were combined with deep learning methods to diagnose membranous nephropathy. After baseline correction and smoothing of the data, Gaussian white noise of different decibels was added to the training set for data amplification, and the amplified data were imported into ResNet, AlexNet and GoogleNet models to obtain the evaluation results of the models for membranous nephropathy. The experimental results showed that the three deep learning models achieved an accuracy of 1 for the classification of serum data of patients with membranous nephropathy and control group, and the discrimination of urine data was above 0.85, among which AlexNet was the best classification model for both samples. The above experimental results illustrate the great potential of serum- and urine-based Raman spectroscopy combined with deep learning methods for rapid and accurate identification of patients with membranous nephropathy.

Identification of clinical heterogeneity and construction of a novel subtype predictive model in patients with ankylosing spondylitis: An unsupervised machine learning study

BACKGROUND

Accurate classification of patients with ankylosing spondylitis (AS) is the premise of precision medicine so as to perform different medical interventions for different patient types. AS pathology is closely related to the changes in the immune microenvironment. In this study, we used unsupervised machine learning (UML) to classify patients with AS based on clinical characteristics. We then constructed a novel subtype predictive model for AS based on the clinical classification, after which we investigated the difference in the immune microenvironment to unravel the AS pathogenesis.

METHODS

Overall, 196 patients with AS were enrolled. UML was used to cluster AS patients by similar clinical characteristics. Functional ability, disease status, and grading of radiologic features were assessed to verify the accuracy and heterogeneity of UML clustering. Least Absolute Shrinkage and Selection Operator (LASSO) regression and Random Forest algorithm were used to screen and identify predictive factors for the novel subtype of AS. Logistic regression was also performed to construct a predictive model of this novel subtype. Datasets were downloaded from the Gene Expression Omnibus database to assess immune cell infiltration, and the results were validated using data of routine blood tests from 3671 AS patients and 5720 non-AS patients. The differential expression of Fat Mass and Obesity-Associated Protein (FTO), an m⁶A regulator, between AS patients and healthy control subjects was confirmed using immunohistochemistry.

RESULTS

UML clustering identified two clusters. The clinical characteristics of the two clusters were significantly heterogeneous. For the novel subtype of AS identified in UML clustering, a predictive model was built using three predictive factors, namely, C-reactive protein (CRP), absolute value of neutrophils (NEU), and absolute value of monocytes (MONO). The area under the curve of the predictive model was 0.983. Heterogeneity in the neutrophil and monocyte counts in AS was verified through immune cell infiltration analysis. Data from routine blood tests revealed that NEU and MONO were significantly higher in AS patients than in non-AS patients (p < 0.001). FTO expression was negatively correlated with both NEU and MONO. Immunohistochemistry analysis confirmed the downregulated expression of FTO.

CONCLUSIONS

UML provides an explicable and remarkable classification of a heterogeneous cohort of AS patients. A novel subtype of AS was identified in UML clustering. CRP, NEU, and MONO were the independent predictive factors for the novel subtype of AS. FTO expression was correlated with immune cell infiltration in AS patients.

Multi-region machine learning-based novel ensemble approaches for predicting COVID-19 pandemic in Africa

Coronavirus disease 2019 (COVID-19) has produced a global pandemic, which has devastating effects on health, economy and social interactions. Despite the less contraction and spread of COVID-19 in Africa compared to some other continents in the world, Africa remains amongst the most vulnerable regions due to less technology and unequipped or poor health system. Recent happenings showed that COVID-19 may stay for years owing to the discoveries of new variants (such as Omicron) and new wave of infections in several countries. Therefore, accurate prediction of new cases is vital to make informed decisions and in evaluating the measures that should be implemented. Studies on COVID-19 prediction are limited in Africa despite the risks and dangers that the virus possessed. Hence, this study was performed to predict daily COVID-19 cases in 10 African countries spread across the north, south, east, west and central Africa considering countries with few and large number of daily COVID-19 cases. Machine learning (ML) models due to their nonlinearity and accurate prediction capabilities were employed for this purpose, including artificial neural network (ANN), adaptive neuro-fuzzy inference system (ANFIS), support vector machine (SVM) and conventional multiple linear regression (MLR) models. As any other natural process, the COVID-19 pandemic may contain both linear and nonlinear aspects. In such circumstances, neither nonlinear (ML) nor linear (MLR) models could be sufficient; hence, combining both ML and MLR models may produce better accuracy. Consequently, to improve the prediction efficiency of the ML models, novel ensemble approaches including ANN-E and SVM-E were employed. The advantage of using ensemble approaches is that they provide collective benefits of all the standalone models, thereby reducing their weaknesses and enhancing their prediction capabilities. The obtained results showed that ANFIS led to better prediction performance with MAD = 0.0106, MSE = 0.0003, RMSE = 0.0185 and R2 = 0.9059 in the validation step. The results of the proposed ensemble approaches demonstrated very high improvements in predicting the COVID-19 pandemic in Africa with MAD = 0.0073, MSE = 0.0002, RMSE = 0.0155 and R2 = 0.9616. The ANN-E improved the standalone models performance in the validation step up to 10%, 14%, 42%, 6%, 83%, 11%, 7%, 5%, 7% and 31% for Morocco, Sudan, Namibia, South Africa, Uganda, Rwanda, Nigeria, Senegal, Gabon and Cameroon, respectively. This study results offer a solid foundation in the application of ensemble approaches for predicting COVID-19 pandemic across all regions and countries in the world.

Exploring risk factors for cervical lymph node metastasis in papillary thyroid microcarcinoma: construction of a novel population-based predictive model

BACKGROUND

Machine learning was a highly effective tool in model construction. We aim to establish a machine learning-based predictive model for predicting the cervical lymph node metastasis (LNM) in papillary thyroid microcarcinoma (PTMC).

METHODS

We obtained data on PTMC from the SEER database, including 10 demographic and clinicopathological characteristics. Univariate and multivariate logistic regression (LR) analyses were applied to screen the risk factors for cervical LNM in PTMC. Risk factors with P < 0.05 in multivariate LR analysis were used as modeling variables. Five different machine learning (ML) algorithms including extreme gradient boosting (XGBoost), random forest (RF), adaptive boosting (AdaBoost), gaussian naive bayes (GNB) and multi-layer perceptron (MLP) and traditional regression analysis were used to construct the prediction model. Finally, the area under the receiver operating characteristic (AUROC) curve was used to compare the model performance.

RESULTS

Through univariate and multivariate LR analysis, we screened out 9 independent risk factors most closely associated with cervical LNM in PTMC, including age, sex, race, marital status, region, histology, tumor size, and extrathyroidal extension (ETE) and multifocality. We used these risk factors to build an ML prediction model, in which the AUROC value of the XGBoost algorithm was higher than the other 4 ML algorithms and was the best ML model. We optimized the XGBoost algorithm through 10-fold cross-validation, and its best performance on the training set (AUROC: 0.809, 95%CI 0.800-0.818) was better than traditional LR analysis (AUROC: 0.780, 95%CI 0.772-0.787).

CONCLUSIONS

ML algorithms have good predictive performance, especially the XGBoost algorithm. With the continuous development of artificial intelligence, ML algorithms have broad prospects in clinical prognosis prediction.

Automatic recognition of schizophrenia from facial videos using 3D convolutional neural network

Schizophrenia affects patients and their families and society because of chronic impairments in cognition, behavior, and emotion. However, its clinical diagnosis mainly depends on the clinicians' knowledge of the patients' symptoms. Other auxiliary diagnostic methods such as MRI and EEG are cumbersome and time-consuming. Recently, the convolutional neural network (CNN) has been applied to the auxiliary diagnosis of psychiatry. Hence, in this study, a method based on deep learning and facial videos is proposed for the rapid detection of schizophrenia. Herein, 125 videos from 125 schizophrenic patients and 75 videos from 75 healthy controls based on emotional stimulation tasks were obtained. The video preprocessing included the experiment clips extraction, face detection, facial region cropping, resizing to 500 × 500 pixel size, and uniform sampling of 100 frames. The preprocessed facial videos were used to train the Resnet18_3D. We utilized ten-fold cross-validation, and held-out testing set to evaluate the model with the accuracy, the precision, the sensitivity, the specificity, the balanced accuracy, and the AUC. The Resnet18_3D trained on Film_order achieved the best performance with accuracy, sensitivity, specificity, balanced accuracy, and AUC of 89.00%, 96.80%, 76.00%, 86.40% and 0.9397. The neural network model indeed recognizes healthy controls and schizophrenic patients through the changes in the area of the face. The results show that facial video under emotional stimulation can be used to classify schizophrenic patients and help clinicians with diagnosis in the clinical environment. Among the different types of stimuli, the video stimuli with fixed emotional order showed the best classification performance.

A digital physician peer to automatically detect erroneous prescriptions in radiotherapy

Appropriate dosing of radiation is crucial to patient safety in radiotherapy. Current quality assurance depends heavily on a physician peer-review process, which includes a review of the treatment plan's dose and fractionation. Potentially, physicians may not identify errors during this manual peer review due to time constraints and caseload. A novel prescription anomaly detection algorithm is designed that utilizes historical data from the past to predict anomalous cases. Such a tool can serve as an electronic peer who will assist the peer-review process providing extra safety to the patients. In our primary model, we create two dissimilarity metrics, R and F. R defining how far a new patient's prescription is from historical prescriptions. F represents how far away a patient's feature set is from that of the group with an identical or similar prescription. We flag prescription if either metric is greater than specific optimized cut-off values. We use thoracic cancer patients (n = 2504) as an example and extracted seven features. Our testing set f1 score is between 73%-94% for different treatment technique groups. We also independently validate our results by conducting a mock peer review with three thoracic specialists. Our model has a lower type II error rate compared to the manual peer-review by physicians.

Applying Machine Learning Algorithms for the Classification of Mink Infected with Aleutian Disease Using Different Data Sources

Simple Summary

Aleutian disease (AD) is a major infectious disease found in mink farms, and it causes financial losses to the mink industry. Controlling AD often requires a counterimmunoelectrophoresis (CIEP) method, which is relatively expensive for mink farmers. Therefore, predicting AD infected mink without using CIEP records will be important for controlling AD in mink farms. In the current study, we applied nine machine learning algorithms to classify AD-infected mink. We indicated that the random forest could be used to classify AD-infected mink (accuracy of 0.962) accurately. This result could be used for implementing machine learning in controlling AD in the mink farms.

Abstract

American mink (Neogale vison) is one of the major sources of fur for the fur industries worldwide, whereas Aleutian disease (AD) is causing severe financial losses to the mink industry. A counterimmunoelectrophoresis (CIEP) method is commonly employed in a test-and-remove strategy and has been considered a gold standard for AD tests. Although machine learning is widely used in livestock species, little has been implemented in the mink industry. Therefore, predicting AD without using CIEP records will be important for controlling AD in mink farms. This research presented the assessments of the CIEP classification using machine learning algorithms. The Aleutian disease was tested on 1157 individuals using CIEP in an AD-positive mink farm (Nova Scotia, Canada). The comprehensive data collection of 33 different features was used for the classification of AD-infected mink. The specificity, sensitivity, accuracy, and F1 measure of nine machine learning algorithms were evaluated for the classification of AD-infected mink. The nine models were artificial neural networks, decision tree, extreme gradient boosting, gradient boosting method, K-nearest neighbors, linear discriminant analysis, support vector machines, naive bayes, and random forest. Among the 33 tested features, the Aleutian mink disease virus capsid protein-based enzyme-linked immunosorbent assay was found to be the most important feature for classifying AD-infected mink. Overall, random forest was the best-performing algorithm for the current dataset with a mean sensitivity of 0.938 ± 0.003, specificity of 0.986 ± 0.005, accuracy of 0.962 ± 0.002, and F1 value of 0.961 ± 0.088, and across tenfold of the cross-validation. Our work demonstrated that it is possible to use the random forest algorithm to classify AD-infected mink accurately. It is recommended that further model tests in other farms need to be performed and the genomic information needs to be used to optimize the model for implementing machine learning methods for AD detection.

Machine learning-based clustering in cervical spondylotic myelopathy patients to identify heterogeneous clinical characteristics

Background

Anterior cervical decompression and fusion can effectively treat cervical spondylotic myelopathy (CSM). Accurately classifying patients with CSM who have undergone anterior cervical decompression and fusion is the premise of precision medicine. In this study, we used machine learning algorithms to classify patients and compare the postoperative efficacy of each classification.

Methods

A total of 616 patients with cervical spondylotic myelopathy who underwent anterior cervical decompression and fusion were enrolled. Unsupervised machine learning algorithms (UMLAs) were used to cluster subjects according to similar clinical characteristics. Then, the results of clustering were visualized. The surgical outcomes were used to verify the accuracy of machine learning clustering.

Results

We identified two clusters in these patients who had significantly different baseline clinical characteristics, preoperative complications, the severity of neurological symptoms, and the range of decompression required for surgery. UMLA divided the CSM patients into two clusters according to the severity of their illness. The repose to surgical treatment between the clusters was significantly different.

Conclusions

Our results showed that UMLA could be used to rationally classify a heterogeneous cohort of CSM patients effectively, and thus, it might be used as the basis for a data-driven platform for identifying the cluster of patients who can respond to a particular treatment method.

Trustworthy AI: A Computational Perspective

In the past few decades, artificial intelligence (AI) technology has experienced swift developments, changing everyone’s daily life and profoundly altering the course of human society. The intention behind developing AI was and is to benefit humans by reducing labor, increasing everyday conveniences, and promoting social good. However, recent research and AI applications indicate that AI can cause unintentional harm to humans by, for example, making unreliable decisions in safety-critical scenarios or undermining fairness by inadvertently discriminating against a group or groups. Consequently, trustworthy AI has recently garnered increased attention regarding the need to avoid the adverse effects that AI could bring to people, so people can fully trust and live in harmony with AI technologies.

A tremendous amount of research on trustworthy AI has been conducted and witnessed in recent years. In this survey, we present a comprehensive appraisal of trustworthy AI from a computational perspective to help readers understand the latest technologies for achieving trustworthy AI. Trustworthy AI is a large and complex subject, involving various dimensions. In this work, we focus on six of the most crucial dimensions in achieving trustworthy AI: (i) Safety & Robustness, (ii) Nondiscrimination & Fairness, (iii) Explainability, (iv) Privacy, (v) Accountability & Auditability, and (vi) Environmental Well-being. For each dimension, we review the recent related technologies according to a taxonomy and summarize their applications in real-world systems. We also discuss the accordant and conflicting interactions among different dimensions and discuss potential aspects for trustworthy AI to investigate in the future.

Discrimination of subjective cognitive decline from healthy control based on glucose-oxygen metabolism network coupling features and machine learning

BACKGROUND

Our previous studies have proved that preclinical Alzheimer's disease (AD) which including subjective cognitive decline (SCD) stage, can be distinguished from normal control (NC) by glucose-oxygen metabolism coupling at the voxel level, but whether the coupling at the network level worked has not been studied. Therefore, this study aimed to explore the coupling relationship between brain glucose metabolic connectivity network and oxygen functional connectivity network, and whether its feasibility as a biomarker to discriminate SCD from healthy control (HC).

METHODS

Resting-state functional magnetic resonance imaging (rs-fMRI) and glucose positron emission tomography (PET) based on hybrid PET/MRI scans were used to investigate metabolism-oxygen metabolism coupling in 56 SCD individuals and 54 HCs. Network coupling features were selected by logistic regression-recursive feature elimination (LR-RFE), and then a linear support vector machine (SVM) was used to distinguish SCD and HC by using 5-fold cross-validation.

RESULTS

The classification average accuracy of network coupling had reached 76.36% with a standard deviation of 9.85% (with a sensitivity of 77.82%±15.13% and a specificity of 75.30%±15.15%). After receiver operating characteristic (ROC) analysis, the average area under curve (AUC) of network coupling was 0.788 (95% confidence interval [Formula: see text]).

CONCLUSION

This study provided a new perspective for exploring network coupling. The proposed classification method highlighted the potential clinical application by combing glucose-oxygen metabolism coupling and machine learning in identifying SCD.

Development of Various Diabetes Prediction Models Using Machine Learning Techniques

BACKGROUND

There are many models for predicting diabetes mellitus (DM), but their clinical implication remains vague. Therefore, we aimed to create various DM prediction models using easily accessible health screening test parameters.

METHODS

Two sets of variables were used to develop eight DM prediction models. One set comprised 62 easily accessible examination results of commonly used variables from a tertiary university hospital. The second set comprised 27 of the 62 variables included in the national routine health checkups. Gradient boosting and random forest algorithms were used to develop the models. Internal validation was performed using the stratified 10-fold cross-validation method.

RESULTS

The area under the receiver operating characteristic curve (ROC-AUC) for the 62-variable DM model making 12-month predictions for subjects without diabetes was the largest (0.928) among those of the eight DM prediction models. The ROC-AUC dropped by more than 0.04 when training with the simplified 27-variable set but still showed fairly good performance with ROC-AUCs between 0.842 and 0.880. The accuracy was up to 11.5% higher (from 0.807 to 0.714) when fasting glucose was included.

CONCLUSION

We created easily applicable diabetes prediction models that deliver good performance using parameters commonly assessed during tertiary university hospital and national routine health checkups. We plan to perform prospective external validation, hoping that the developed DM prediction models will be widely used in clinical practice.

Improved Analysis of COVID-19 Influenced Pneumonia from the Chest X-Rays Using Fine-Tuned Residual Networks

COVID-19 has remained a threat to world life despite a recent reduction in cases. There is still a possibility that the virus will evolve and become more contagious. If such a situation occurs, the resulting calamity will be worse than in the past if we act irresponsibly. COVID-19 must be widely screened and recognized early to avert a global epidemic. Positive individuals should be quarantined immediately, as this is the only effective way to prevent a global tragedy that has occurred previously. No positive case should go unrecognized. However, current COVID-19 detection procedures require a significant amount of time during human examination based on genetic and imaging techniques. Apart from RT-PCR and antigen-based tests, CXR and CT imaging techniques aid in the rapid and cost-effective identification of COVID. However, discriminating between diseased and normal X-rays is a time-consuming and challenging task requiring an expert's skill. In such a case, the only solution was an automatic diagnosis strategy for identifying COVID-19 instances from chest X-ray images. This article utilized a deep convolutional neural network, ResNet, which has been demonstrated to be the most effective for image classification. The present model is trained using pretrained ResNet on ImageNet weights. The versions of ResNet34, ResNet50, and ResNet101 were implemented and validated against the dataset. With a more extensive network, the accuracy appeared to improve. Nonetheless, our objective was to balance accuracy and training time on a larger dataset. By comparing the prediction outcomes of the three models, we concluded that ResNet34 is a more likely candidate for COVID-19 detection from chest X-rays. The highest accuracy level reached 98.34%, which was higher than the accuracy achieved by other state-of-the-art approaches examined in earlier studies. Subsequent analysis indicated that the incorrect predictions occurred with approximately 100% certainty. This uncovered a severe weakness in CNN, particularly in the medical area, where critical decisions are made. However, this can be addressed further in a future study by developing a modified model to incorporate uncertainty into the predictions, allowing medical personnel to manually review the incorrect predictions.

The Validity of Machine Learning Procedures in Orthodontics: What Is Still Missing?

Artificial intelligence (AI) models and procedures hold remarkable predictive efficiency in the medical domain through their ability to discover hidden, non-obvious clinical patterns in data. However, due to the sparsity, noise, and time-dependency of medical data, AI procedures are raising unprecedented issues related to the mismatch between doctors’ mentalreasoning and the statistical answers provided by algorithms. Electronic systems can reproduce or even amplify noise hidden in the data, especially when the diagnosis of the subjects in the training data set is inaccurate or incomplete. In this paper we describe the conditions that need to be met for AI instruments to be truly useful in the orthodontic domain. We report some examples of computational procedures that are capable of extracting orthodontic knowledge through ever deeper patient representation. To have confidence in these procedures, orthodontic practitioners should recognize the benefits, shortcomings, and unintended consequences of AI models, as algorithms that learn from human decisions likewise learn mistakes and biases.

Establishment of a Combined Diagnostic Model of Abdominal Aortic Aneurysm with Random Forest and Artificial Neural Network

Objectives. Abdominal aortic aneurysm (AAA), a disease with high mortality, is limited by the current diagnostic methods in the early screening. This study aimed to screen novel and significant biomarkers and construct a diagnostic model for AAA by using a novel machine learning method, i.e., an ensemble of the random forest (RF) algorithm and artificial neural network (ANN). Methods and Results. Through a search of the Gene Expression Omnibus (GEO) database, two large-sample gene expression datasets (GSE57691 and GSE47472) were downloaded and preprocessed. Differentially expressed genes (DEGs) in GSE57691 were identified by R software, followed by Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment using the Database for Annotation, Visualization, and Integrated Discovery (DAVID). Essential metabolic pathways related to positive regulation of cell death and NAD binding were found. Then, RF was used to identify key genes from the DEGs, and an AAA diagnostic model was established by ANN. A transcription factor (TF) regulatory network of key genes related to angiogenesis and endothelial migration was constructed. Finally, a validation dataset was used to validate the model and the area under the receiver operating characteristic curve (AUC) value was high. Conclusion. Potential AAA-associated gene biomarkers were identified by RF, and a novel early diagnostic model of AAA was established by ANN. The AUC indicated that the diagnostic model had a highly satisfactory diagnostic performance. In conclusion, this study will provide a promising theoretical basis for further clinical and experimental studies.

iApp: An Autonomous Inspection, Auscultation, Percussion, and Palpation Platform

Disease symptoms often contain features that are not routinely recognized by patients but can be identified through indirect inspection or diagnosis by medical professionals. Telemedicine requires sufficient information for aiding doctors' diagnosis, and it has been primarily achieved by clinical decision support systems (CDSSs) utilizing visual information. However, additional medical diagnostic tools are needed for improving CDSSs. Moreover, since the COVID-19 pandemic, telemedicine has garnered increasing attention, and basic diagnostic tools (e.g., classical examination) have become the most important components of a comprehensive framework. This study proposes a conceptual system, iApp, that can collect and analyze quantified data based on an automatically performed inspection, auscultation, percussion, and palpation. The proposed iApp system consists of an auscultation sensor, camera for inspection, and custom-built hardware for automatic percussion and palpation. Experiments were designed to categorize the eight abdominal divisions of healthy subjects based on the system multi-modal data. A deep multi-modal learning model, yielding a single prediction from multi-modal inputs, was designed for learning distinctive features in eight abdominal divisions. The model's performance was evaluated in terms of the classification accuracy, sensitivity, positive predictive value, and F-measure, using epoch-wise and subject-wise methods. The results demonstrate that the iApp system can successfully categorize abdominal divisions, with the test accuracy of 89.46%. Through an automatic examination of the iApp system, this proof-of-concept study demonstrates a sophisticated classification by extracting distinct features of different abdominal divisions where different organs are located. In the future, we intend to capture the distinct features between normal and abnormal tissues while securing patient data and demonstrate the feasibility of a fully telediagnostic system that can support abnormality diagnosis.

A hybrid machine learning approach of fuzzy-rough-k-nearest neighbor, latent semantic analysis, and ranker search for efficient disease diagnosis

Machine learning approaches have a valuable contribution in improving competency in automated decision systems. Several machine learning approaches have been developed in the past studies in individual disease diagnosis prediction. The present study aims to develop a hybrid machine learning approach for diagnosis predictions of multiple diseases based on the combination of efficient feature generation, selection, and classification methods. Specifically, the combination of latent semantic analysis, ranker search, and fuzzy-rough-k-nearest neighbor has been proposed and validated in the diagnosis prediction of the primary tumor, post-operative, breast cancer, lymphography, audiology, fertility, immunotherapy, and COVID-19, etc. The performance of the proposed approach is compared with single and other hybrid machine learning approaches in terms of accuracy, analysis time, precision, recall, F-measure, the area under ROC, and the Kappa coefficient. The proposed hybrid approach performs better than single and other hybrid approaches in the diagnosis prediction of each of the selected diseases. Precisely, the suggested approach achieved the maximum recognition accuracy of 99.12%of the primary tumor, 96.45%of breast cancer Wisconsin, 94.44%of cryotherapy, 93.81%of audiology, and significant improvement in the classification accuracy and other evaluation metrics in the recognition of the rest of the selected diseases. Besides, it handles the missing values in the dataset effectively.

Machine learning in vascular surgery: a systematic review and critical appraisal

Machine learning (ML) is a rapidly advancing field with increasing utility in health care. We conducted a systematic review and critical appraisal of ML applications in vascular surgery. MEDLINE, Embase, and Cochrane CENTRAL were searched from inception to March 1, 2021. Study screening, data extraction, and quality assessment were performed by two independent reviewers, with a third author resolving discrepancies. All original studies reporting ML applications in vascular surgery were included. Publication trends, disease conditions, methodologies, and outcomes were summarized. Critical appraisal was conducted using the PROBAST risk-of-bias and TRIPOD reporting adherence tools. We included 212 studies from a pool of 2235 unique articles. ML techniques were used for diagnosis, prognosis, and image segmentation in carotid stenosis, aortic aneurysm/dissection, peripheral artery disease, diabetic foot ulcer, venous disease, and renal artery stenosis. The number of publications on ML in vascular surgery increased from 1 (1991-1996) to 118 (2016-2021). Most studies were retrospective and single center, with no randomized controlled trials. The median area under the receiver operating characteristic curve (AUROC) was 0.88 (range 0.61-1.00), with 79.5% [62/78] studies reporting AUROC ≥ 0.80. Out of 22 studies comparing ML techniques to existing prediction tools, clinicians, or traditional regression models, 20 performed better and 2 performed similarly. Overall, 94.8% (201/212) studies had high risk-of-bias and adherence to reporting standards was poor with a rate of 41.4%. Despite improvements over time, study quality and reporting remain inadequate. Future studies should consider standardized tools such as PROBAST and TRIPOD to improve study quality and clinical applicability.

Colorimetric and Electrochemical Screening for Early Detection of Diabetes Mellitus and Diabetic Retinopathy-Application of Sensor Arrays and Machine Learning

In this review, a selection of works on the sensing of biomarkers related to diabetes mellitus (DM) and diabetic retinopathy (DR) are presented, with the scope of helping and encouraging researchers to design sensor-array machine-learning (ML)-supported devices for robust, fast, and cost-effective early detection of these devastating diseases. First, we highlight the social relevance of developing systematic screening programs for such diseases and how sensor-arrays and ML approaches could ease their early diagnosis. Then, we present diverse works related to the colorimetric and electrochemical sensing of biomarkers related to DM and DR with non-invasive sampling (e.g., urine, saliva, breath, tears, and sweat samples), with a special mention to some already-existing sensor arrays and ML approaches. We finally highlight the great potential of the latter approaches for the fast and reliable early diagnosis of DM and DR.

Combined Structural MR and Diffusion Tensor Imaging Classify the Presence of Alzheimer's Disease With the Same Performance as MR Combined With Amyloid Positron Emission Tomography: A Data Integration Approach

Background: In recent years, classification frameworks using imaging data have shown that multimodal classification methods perform favorably over the use of a single imaging modality for the diagnosis of Alzheimer's Disease. The currently used clinical approach often emphasizes the use of qualitative MRI and/or PET data for clinical diagnosis. Based on the hypothesis that classification of isolated imaging modalities is not predictive of their respective value in combined approaches, we investigate whether the combination of T1 Weighted MRI and diffusion tensor imaging (DTI) can yield an equivalent performance as the combination of quantitative structural MRI (sMRI) with amyloid-PET. Methods: We parcellated the brain into regions of interest (ROI) following different anatomical labeling atlases. For each region of interest different metrics were extracted from the different imaging modalities (sMRI, PiB-PET, and DTI) to be used as features. Thereafter, the feature sets were reduced using an embedded-based feature selection method. The final reduced sets were then used as input in support vector machine (SVM) classifiers. Three different base classifiers were created, one for each imaging modality, and validated using internal (n = 41) and external data from the ADNI initiative (n = 330 for sMRI, n = 148 for DTI and n = 55 for PiB-PET) sources. Finally, the classifiers were ensembled using a weighted method in order to evaluate the performance of different combinations. Results: For the base classifiers the following performance levels were found: sMRI-based classifier (accuracy, 92%; specificity, 97% and sensitivity, 87%), PiB-PET (accuracy, 91%; specificity, 89%; and sensitivity, 92%) and the lowest performance was attained with DTI (accuracy, 80%; specificity, 76%; and sensitivity, 82%). From the multimodal approaches, when integrating two modalities, the following results were observed: sMRI+PiB-PET (accuracy, 98%; specificity, 98%; and sensitivity, 99%), sMRI+DTI (accuracy, 97%; specificity, 99%; and sensitivity, 94%) and PiB-PET+DTI (accuracy, 91%; specificity, 90%; and sensitivity, 93%). Finally, the combination of all imaging modalities yielded an accuracy of 98%, specificity of 97% and sensitivity of 99%. Conclusion: Although DTI in isolation shows relatively poor performance, when combined with structural MR, it showed a surprising classification performance which was comparable to MR combined with amyloid PET. These results are consistent with the notion that white matter changes are also important in Alzheimer's Disease.

Nonwearable Sensor-Based In-Home Assessment of Subtle Daily Behavioral Changes as a Candidate Biomarker for Mild Cognitive Impairment

Patients show subtle changes in daily behavioral patterns, revealed by traditional assessments (e.g., performance- or questionnaire-based assessments) even in the early stage of Alzheimer's disease (AD; i.e., the mild cognitive impairment (MCI) stage). An increase in studies on the assessment of daily behavioral changes in patients with MCI and AD using digital technologies (e.g., wearable and nonwearable sensor-based assessment) has been noted in recent years. In addition, more objective, quantitative, and realistic evidence of altered daily behavioral patterns in patients with MCI and AD has been provided by digital technologies rather than traditional assessments. Therefore, this study hypothesized that the assessment of daily behavioral changes with digital technologies can replace or assist traditional assessment methods for early MCI and AD detection. In this review, we focused on research using nonwearable sensor-based in-home assessment. Previous studies on the assessment of behavioral changes in MCI and AD using traditional performance- or questionnaire-based assessments are first described. Next, an overview of previous studies on the assessment of behavioral changes in MCI and AD using nonwearable sensor-based in-home assessment is provided. Finally, the usefulness and problems of nonwearable sensor-based in-home assessment for early MCI and AD detection are discussed. In conclusion, this review stresses that subtle changes in daily behavioral patterns detected by nonwearable sensor-based in-home assessment can be early MCI and AD biomarkers.

Complexity and data mining in dental research: A network medicine perspective on interceptive orthodontics

Procedures and models of computerized data analysis are becoming researchers' and practitioners' thinking partners by transforming the reasoning underlying biomedicine. Complexity theory, Network analysis and Artificial Intelligence are already approaching this discipline, intending to provide support for patient's diagnosis, prognosis and treatments. At the same time, due to the sparsity, noisiness and time-dependency of medical data, such procedures are raising many unprecedented problems related to the mismatch between the human mind's reasoning and the outputs of computational models. Thanks to these computational, non-anthropocentric models, a patient's clinical situation can be elucidated in the orthodontic discipline, and the growth outcome can be approximated. However, to have confidence in these procedures, orthodontists should be warned of the related benefits and risks. Here we want to present how these innovative approaches can derive better patients' characterization, also offering a different point of view about patient's classification, prognosis and treatment.