Diagnostic Prediction Model Research Articles

Identification of Diagnostic Markers for Breast Cancer Based on Differential Gene Expression and Pathway Network.

Background: Breast cancer is the second largest cancer in the world, the incidence of breast cancer continues to rise worldwide, and women’s health is seriously threatened. Therefore, it is very important to explore the characteristic changes of breast cancer from the gene level, including the screening of differentially expressed genes and the identification of diagnostic markers. Methods: The gene expression profiles of breast cancer were obtained from the TCGA database. The edgeR R software package was used to screen the differentially expressed genes between breast cancer patients and normal samples. The function and pathway enrichment analysis of these genes revealed significant enrichment of functions and pathways. Next, download these pathways from KEGG website, extract the gene interaction relations, construct the KEGG pathway gene interaction network. The potential diagnostic markers of breast cancer were obtained by combining the differentially expressed genes with the key genes in the network. Finally, these markers were used to construct the diagnostic prediction model of breast cancer, and the predictive ability of the model and the diagnostic ability of the markers were verified by internal and external data. Results: 1060 differentially expressed genes were identified between breast cancer patients and normal controls. Enrichment analysis revealed 28 significantly enriched pathways (p < 0.05). They were downloaded from KEGG website, and the gene interaction relations were extracted to construct the gene interaction network of KEGG pathway, which contained 1277 nodes and 7345 edges. The key nodes with a degree greater than 30 were extracted from the network, containing 154 genes. These 154 key genes shared 23 genes with differentially expressed genes, which serve as potential diagnostic markers for breast cancer. The 23 genes were used as features to construct the SVM classification model, and the model had good predictive ability in both the training dataset and the validation dataset (AUC = 0.960 and 0.907, respectively). Conclusion: This study showed that the difference of gene expression level is important for the diagnosis of breast cancer, and identified 23 breast cancer diagnostic markers, which provides valuable information for clinical diagnosis and basic treatment experiments.

A Dutch Prediction Tool to Assess the Risk of Incidental Gallbladder Cancers in Cholecystectomies for Benign Gallstone Disease

Introduction: A selective histopathologic (Sel-HP) policy of gallbladder specimens following cholecystectomy is increasingly implemented in low incidence countries. However, the fear of missing incidental gallbladder cancer (iGBC) persists and thereby restrict further implementation of a Sel-HP policy. This study aims to develop a diagnostic prediction model for selecting gallbladders in need of histopathologic examination after cholecystectomy. Method: A registration based retrospective cohort study of nine hospitals in the Southern part of the Netherlands between 2004 and 2014 was used. Five potential clinical predictors for gallbladder cancer (GBC) were selected based on previous literature and expert opinion. The prediction model was internally validated by bootstrapping. Discriminative capacity and accuracy were tested by assessing the area under the receiver operating characteristic curve (AUC) Nagelkerke Pseudo-R2 and Brier score. Finally, a clinical 4-step protocol was formulated using radiological, clinical indications correlated with high susceptibility for GBC, the clinical prediction model and finally the macroscopic examination of the gallbladder. Results: The prediction model was developed using a cohort of 22025 gallbladders containing 75 GBC cases. A model was developed including the following variables: age, sex, urgency, type of surgery and indication for surgery. The AUC was 90.3% (95% confidence interval 86.2% – 94.4%), indicating a good discriminative ability of the model. The calibration plot indicated good calibration of the predicted probabilities. Conclusions: We developed a clinical prediction model for selecting gallbladder specimens for histopathologic examination after cholecystectomy for ruling out GBC. This protocol will support surgeons in selecting appropriate gallbladders specimens for selective pathologic policy.

A Novel Diagnostic Prediction Model to Distinguish Tuberculous Meningitis and Cryptococcal Meningitis

A Novel Diagnostic Prediction Model to Distinguish Tuberculous Meningitis and Cryptococcal Meningitis

Introduction to Clinical Prediction Models.

Clinical prediction models include a diagnostic prediction model to estimate the probability of an individual currently having a disease (e.g., pulmonary embolism) and a prognostic prediction model to estimate the probability of an individual developing a specific health outcome over a specific time period (e.g., myocardial infarction and stroke in 10 years). Clinical prediction models can be developed by applying traditional regression models (e.g., logistic and Cox regression models) or emerging machine learning models to real-world data, such as electronic health records and administrative claims data. For derivation, researchers select candidate variables based on a literature review and clinical knowledge, and predictor variables in the final model based on pre-defined criteria (e.g., thresholds for the size of relative risk and p-values) or strategies such as the stepwise regression and the least absolute shrinkage and selection operator (LASSO) regression. For validation, the clinical prediction model's performance is evaluated in terms of goodness of fit (e.g., R2), discrimination (e.g., area under the receiver operating characteristic curve or c-statistics), and calibration (e.g., calibration plot and Hosmer-Lemeshow test). Performance of a new variable added to an existing clinical prediction model is evaluated in terms of reclassification (e.g., net reclassification improvement and integrated discrimination improvement). The model should be validated using the original data to examine internal validity through methods such as resampling (e.g., cross-validation and bootstrapping) and using other participants' data to examine external validity. For successful implementation of a clinical prediction model in actual clinical practice, presentation methods such as paper-based (nomogram) or web-based calculator and an easy-to-use risk score should be considered.

Development of earth observational diagnostic drought prediction model for regional error calibration: A case study on agricultural drought in Kyrgyzstan

ABSTRACT Drought is a natural disaster that occurs globally and is a main trigger of secondary environmental and socio-economic damages, such as food insecurity, land degradation, and sand-dust storms. As climate change is being accelerated by human activities and environmental changes, both the severity and uncertainties of drought are increasing. In this study, a diagnostic drought prediction model (DDPM) was developed to reduce the uncertainties caused by environmental diversity at the regional level in Kyrgyzstan, by predicting drought with meteorological forecasts and satellite image diagnosis. The DDPM starts with applying a prognostic drought prediction model (PDPM) to 1) estimate future agricultural drought by explaining its relationship with the standardized precipitation index (SPI), an accumulated precipitation anomaly, and 2) compensate for regional variances, which were not reflected sufficiently in the PDPM, by taking advantage of preciseness in the time-series vegetation condition index (VCI), a satellite-based index representing land surface conditions. Comparing the prediction results with the monitored VCI from June to August, it was found that the DDPM outperformed the PDPM, which exploits only meteorological data, in both spatiotemporal and spatial accuracy. In particular, for June to August, respectively, the results of the DDPM (coefficient of determination [R 2] = 0.27, 0.36, and 0.4; root mean squared error [RMSE] = 0.16, 0.13, and 0.13) were more effective in explaining the spatial details of drought severity on a regional scale than those of the PDPM (R 2 = 0.09, 0.10, and 0.11; RMSE = 0.17, 0.15, and 0.16). The DDPM revealed the possibility of advanced drought assessment by integrating the earth observation big data comprising meteorological and satellite data. In particular, the advantage of data fusion is expected to be maximized in areas with high land surface heterogeneity or sparse weather stations by providing observational feedback to the PDPM. This research is anticipated to support policymakers and technical officials in establishing effective policies, action plans, and disaster early warning systems to reduce disaster risk and prevent environmental and socio-economic damage.

484. Identification of Early Features to Differentiate Hospitalized Children Admitted for Suspected MIS-C from Alternative Diagnoses

BackgroundMulti-system inflammatory syndrome in children (MIS-C) is a rare consequence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). MIS-C shares features with common infectious and inflammatory syndromes and differentiation early in the course is difficult. Identification of early features specific to MIS-C may lead to faster diagnosis and treatment. We aimed to determine clinical, laboratory, and cardiac features distinguishing MIS-C patients within the first 24 hours of admission to the hospital from those who present with similar features but ultimately diagnosed with an alternative etiology.MethodsWe performed retrospective chart reviews of children (0-20 years) who were admitted to Vanderbilt Children’s Hospital and evaluated under our institutional MIS-C algorithm between June 10, 2020-April 8, 2021. Subjects were identified by review of infectious disease (ID) consults during the study period as all children with possible MIS-C require an ID consult per our institutional algorithm. Clinical, lab, and cardiac characteristics were compared between children with and without MIS-C. The diagnosis of MIS-C was determined by the treating team and available consultants. P-values were calculated using two-sample t-tests allowing unequal variances for continuous and Pearson’s chi-squared test for categorical variables, alpha set at < 0.05.ResultsThere were 128 children admitted with concern for MIS-C. Of these, 45 (35.2%) were diagnosed with MIS-C and 83 (64.8%) were not. Patients with MIS-C had significantly higher rates of SARS-CoV-2 exposure, hypotension, conjunctival injection, abdominal pain, and abnormal cardiac exam (Table 1). Laboratory evaluation showed that patients with MIS-C had lower platelet count, lymphocyte count and sodium level, with higher c-reactive protein, fibrinogen, B-type natriuretic peptide, and neutrophil percentage (Table 2). Patients with MIS-C also had lower ejection fraction and were more likely to have abnormal electrocardiogram. ConclusionWe identified early features that differed between patients with MIS-C from those without. Development of a diagnostic prediction model based on these early distinguishing features is currently in progress. Disclosures Natasha B. Halasa, MD, MPH, Genentech (Other Financial or Material Support, I receive an honorarium for lectures - it’s a education grant, supported by genetech)Quidel (Grant/Research Support, Other Financial or Material Support, Donation of supplies/kits)Sanofi (Grant/Research Support, Other Financial or Material Support, HAI/NAI testing) Natasha B. Halasa, MD, MPH, Genentech (Individual(s) Involved: Self): I receive an honorarium for lectures - it’s a education grant, supported by genetech, Other Financial or Material Support, Other Financial or Material Support; Sanofi (Individual(s) Involved: Self): Grant/Research Support, Research Grant or Support James A. Connelly, MD, Horizon Therapeutics (Advisor or Review Panel member)X4 Pharmaceuticals (Advisor or Review Panel member)

Artificial Intelligence for Computer Vision in Surgery: A Call for Developing Reporting Guidelines.

Artificial Intelligence for Computer Vision in Surgery: A Call for Developing Reporting Guidelines.

Risk of bias in studies on prediction models developed using supervised machine learning techniques: systematic review

ObjectiveTo assess the methodological quality of studies on prediction models developed using machine learning techniques across all medical specialties.DesignSystematic review.Data sourcesPubMed from 1 January 2018 to 31 December 2019.Eligibility criteriaArticles...

Neutrophils and Lymphocytes Can Help Distinguish Asymptomatic COVID-19 From Moderate COVID-19.

IntroductionAsymptomatic coronavirus disease 2019 (COVID-19) and moderate COVID-19 may be the most common COVID-19 cases. This study was designed to develop a diagnostic model for patients with asymptomatic and moderate COVID-19 based on demographic, clinical, and laboratory variables.MethodsThis retrospective study divided the subjects into 2 groups: asymptomatic COVID-19 (without symptoms, n = 15) and moderate COVID-19 (with symptoms, n = 57). Demographic characteristics, clinical data, routine blood tests, other laboratory tests, and inpatient data were collected and analyzed to compare patients with asymptomatic COVID-19 and moderate COVID-19.ResultsComparison of the asymptomatic COVID-19 group with the moderate COVID-19 group yielded the following results: the patients were younger (P = 0.045); the cluster of differentiation (CD)8+ (cytotoxic) T cell level was higher (P = 0.017); the C-reactive protein (CRP) level was lower (P = 0.001); the white blood cell (WBC, P < 0.001), neutrophil (NEU, P = 0.036), lymphocyte (LYM, P = 0.009), and eosinophil (EOS, P = 0.036) counts were higher; and the serum iron level (P = 0.049) was higher in the asymptomatic COVID-19 group. The multivariate analysis showed that the NEU count (odds ratio [OR] = 2.007, 95% confidence interval (CI): 1.162 - 3.715, P = 0.014) and LYM count (OR = 9.380, 95% CI: 2.382 - 36.934, P = 0.001) were independent factors for the presence of clinical symptoms after COVID-19 infection. The NEU count and LYM count were diagnostic predictors of asymptomatic COVID-19. This diagnostic prediction model showed high discriminatory power, consistency, and net clinical benefits.ConclusionsThe proposed model can distinguish asymptomatic COVID-19 from moderate COVID-19, thereby helping clinicians identify and distinguish patients with potential asymptomatic COVID-19 from those with moderate COVID-19.

Investigating the association of CD36 gene polymorphisms (rs1761667 and rs1527483) with T2DM and dyslipidemia: Statistical analysis, machine learning based prediction, and meta-analysis.

CD36 (cluster of differentiation 36) is a membrane protein involved in lipid metabolism and has been linked to pathological conditions associated with metabolic disorders, such as diabetes and dyslipidemia. A case-control study was conducted and included 177 patients with type-2 diabetes mellitus (T2DM) and 173 control subjects to study the involvement of CD36 gene rs1761667 (G>A) and rs1527483 (C>T) polymorphisms in the pathogenesis of T2DM and dyslipidemia among Jordanian population. Lipid profile, blood sugar, gender and age were measured and recorded. Also, genotyping analysis for both polymorphisms was performed. Following statistical analysis, 10 different neural networks and machine learning (ML) tools were used to predict subjects with diabetes or dyslipidemia. Towards further understanding of the role of CD36 protein and gene in T2DM and dyslipidemia, a protein-protein interaction network and meta-analysis were carried out. For both polymorphisms, the genotypic frequencies were not significantly different between the two groups (p > 0.05). On the other hand, some ML tools like multilayer perceptron gave high prediction accuracy (≥ 0.75) and Cohen’s kappa (κ) (≥ 0.5). Interestingly, in K-star tool, the accuracy and Cohen’s κ values were enhanced by including the genotyping results as inputs (0.73 and 0.46, respectively, compared to 0.67 and 0.34 without including them). This study confirmed, for the first time, that there is no association between CD36 polymorphisms and T2DM or dyslipidemia among Jordanian population. Prediction of T2DM and dyslipidemia, using these extensive ML tools and based on such input data, is a promising approach for developing diagnostic and prognostic prediction models for a wide spectrum of diseases, especially based on large medical databases.

A Clinical Score to Predict "Corrected Thrombolysis in Myocardial Infarction Frame Count" in Patients With ST-Segment Elevation Myocardial Infarction Undergoing Primary Percutaneous Coronary Intervention.

Corrected thrombolysis in myocardial infarction frame count (cTFC) is an objective, simple, and reproducible method to assess coronary blood flow which is a surrogate for cardiovascular outcomes. It is important to learn which factors are associated with cTFC. The goal of this study was to determine predictive models for epicardial blood flow assessed by cTFC and develop a diagnostic predictive model that indicates the individualized assessment of epicardial blood flow prior to primary percutaneous coronary intervention. This is a retrospective study including 3205 patients with ST-segment elevation myocardial infarction who underwent pPCI. The primary outcome was cTFC. Multivariable linear regression analysis was performed. Subsequently, a nomogram was developed to predict cTFC according to the candidate predictors. Median age was 58; the number of male patients was 2381 (74.3%). Median value of cTFC was 22 and interquartile range (IQR): 16.5-28.0). Age, diabetes mellitus (DM), total ischemic time, systolic blood pressure (SBP), heart rate (HR), and history of statin use remained in both full and reduced models. Our model may potentially allow clinicians to identify patients at high risk for impaired epicardial perfusion.

Using Combinations of Both Clinical and Radiographic Parameters to Develop a Diagnostic Prediction Model Demonstrated an Excellent Performance in Early Detection of Patients with Blount's Disease.

Early identification of pathological causes for pediatric genu varum (bowlegs) is crucial for preventing a progressive, irreversible knee deformity of the child. This study aims to develop and validate a diagnostic clinical prediction algorithm for assisting physicians in distinguishing an early stage of Blount’s disease from the physiologic bowlegs to provide an early treatment that could prevent the progressive, irreversible deformity. The diagnostic prediction model for differentiating an early stage of Blount’s disease from the physiologic bowlegs was developed under a retrospective case-control study from 2000 to 2017. Stepwise backward elimination of multivariable logistic regression modeling was used to derive a diagnostic model. A total of 158 limbs from 79 patients were included. Of those, 84 limbs (53.2%) were diagnosed as Blount’s disease. The final model that included age, BMI, MDA, and MMB showed excellent performance (area under the receiver operating characteristic (AuROC) curve: 0.85, 95% confidence interval 0.79 to 0.91) with good calibration. The proposed diagnostic prediction model for discriminating an early stage of Blount’s disease from physiologic bowlegs showed high discriminative ability with minimal optimism.

Bacteria-Derived Extracellular Vesicles in Urine as a Novel Biomarker for Gastric Cancer: Integration of Liquid Biopsy and Metagenome Analysis.

Simple SummaryGastric cancer shows an improved prognosis when diagnosed in its early stage. However, non-invasive diagnostic markers for gastric cancer known to date have poor clinical efficacies. Many studies have shown that gastric cancer patients have distinct microbial changes compared to normal subjects. In the present study, we performed metagenome analysis using body fluid samples (gastric juice, blood, and urine) to investigate the distinct microbial composition using bacteria-derived EVs from gastric cancer patients. We could build diagnostic prediction models for gastric cancer with the metagenomic data and analyzed the accuracy of models. Although further validation is required to apply these findings to real clinical practice yet, our study showed the possibility of gastric cancer diagnosis with the integration of liquid biopsy and metagenome analysis.Early detection is crucial for improving the prognosis of gastric cancer, but there are no non-invasive markers for the early diagnosis of gastric cancer in real clinical settings. Recently, bacteria-derived extracellular vesicles (EVs) emerged as new biomarker resources. We aimed to evaluate the microbial composition in gastric cancer using bacteria-derived EVs and to build a diagnostic prediction model for gastric cancer with the metagenome data. Stool, urine, and serum samples were prospectively collected from 453 subjects (gastric cancer, 181; control, 272). EV portions were extracted from the samples for metagenome analysis. Differences in microbial diversity and composition were analyzed with 16S rRNA gene profiling, using the next-generation sequencing method. Biomarkers were selected using logistic regression models based on relative abundances at the genus level. The microbial composition of healthy groups and gastric cancer patient groups was significantly different in all sample types. The compositional differences of various bacteria, based on relative abundances, were identified at the genus level. Among the diagnostic prediction models for gastric cancer, the urine-based model showed the highest performance when compared to that of stool or serum. We suggest that bacteria-derived EVs in urine can be used as novel metagenomic markers for the non-invasive diagnosis of gastric cancer by integrating the liquid biopsy method and metagenome analysis.

A Novel Diagnostic Predictive Model for Idiopathic Short Stature in Children.

ObjectiveIdiopathic short stature (ISS), an endocrine-related disease, is difficult to diagnose. Previous studies have shown that many children with some inflammation-related diseases often have short stature, but whether inflammation is the underlying mechanism of ISS has not been studied. Here, we attempt to explore the role of inflammation in the occurrence and development of ISS and to demonstrate an available clinical diagnostic model of ISS.MethodsFrozen serum samples were collected from ISS patients (n = 4) and control individuals (n = 4). Isobaric tags for relative and absolute quantitation (iTRAQ) combined with LC-MS/MS analysis were applied to quantitative proteomics analysis. To assess clusters of potentially interacting proteins, functional enrichment (GO and KEGG) and protein-protein interaction network analyses were performed, and the crucial proteins were detected by Molecular Complex Detection (MCODE). Furthermore, serum levels of two selected proteins were measured by ELISA between ISS patients (n = 80) and controls (n = 80). In addition, experiments in vitro were used to further explore the effects of crucial proteins on endochondral ossification.ResultsA total of 437 proteins were quantified, and 84 DEPs (60 upregulated and 24 downregulated) were identified between patients with ISS and controls. Functional enrichment analysis showed that the DEPs were primarily enriched in blood microparticle, acute inflammatory response, protein activation cascade, collagen-containing extracellular matrix, platelet degranulation, etc. According to the results of top 10 fold change DEPs and MCODE analysis, C1QA and C1QB were selected to further experiment. The expression levels of C1QA and C1QB were validated in serum samples. Based on the logistic regression analysis and ROC curve analysis, we constructed a novel diagnostic model by serum levels of C1QA and C1QB with a specificity of 91.2% and a sensitivity of 75% (AUC = 0.900, p <0.001). Finally, the western blotting analysis confirmed the expression levels of OCN, OPN, RUNX2, and Collagen X were downregulated in chondrocytes, and the outcome of Collagen II was upregulated.ConclusionOur study is the first to demonstrate the significant role of inflammation in the development of ISS. In addition, we identify C1QA and C1QB as novel serum biomarkers for the diagnosis of ISS.

Enhanced CT-Based Radiomics to Predict Micropapillary Pattern Within Lung Invasive Adenocarcinoma.

ObjectiveWe aimed to investigate whether enhanced CT-based radiomics can predict micropapillary pattern (MPP) of lung invasive adenocarcinoma (IAC) in the pre-op phase and to develop an individual diagnostic predictive model for MPP in IAC.Methods170 patients who underwent complete resection for pathologically confirmed lung IAC were included in our study. Of these 121 were used as a training cohort and the other 49 as a test cohort. Clinical features and enhanced CT images were collected and assessed. Quantitative CT analysis was performed based on feature types including first order, shape, gray-level co-occurrence matrix-based, gray-level size zone matrix-based, gray-level run length matrix-based, gray-level dependence matrix-based, neighboring gray tone difference matrix-based features and transform types including Log, wavelet and local binary pattern. Receiver operating characteristic (ROC) and area under the curve (AUC) were used to value the ability to identify the lung IAC with MPP using these characteristics.ResultsUsing quantitative CT analysis, one thousand three hundred and seventeen radiomics features were deciphered from R (https://www.r-project.org/). Then these radiomic features were decreased to 14 features after dimension reduction using the least absolute shrinkage and selection operator (LASSO) method in R. After correlation analysis, 5 key features were obtained and used as signatures for predicting MPP within IAC. The individualized prediction model which included age, smoking, family tumor history and radiomics signature had better identification (AUC=0.739) in comparison with the model consisting only of radiomics features (AUC=0.722). DeLong test showed that the difference in AUC between the two models was statistically significant (P<0.01). Compared with the simple radiomics model, the more comprehensive individual prediction model has better prediction performance.ConclusionThe use of radiomics approach is of great value in the diagnosis of tumors by non-invasive means. The individualized prediction model in the study, when incorporated with age, smoking and radiomics signature, had effective predictive performance of lung IAC with MPP lesions. The combination of imaging features and clinical features can provide additional diagnostic value to identify the micropapillary pattern in IAC and can affect clinical diagnosis and treatment.

Circulating cell-free DNA-based methylation patterns for breast cancer diagnosis

Mammography is used to detect breast cancer (BC), but its sensitivity is limited, especially for dense breasts. Circulating cell-free DNA (cfDNA) methylation tests is expected to compensate for the deficiency of mammography. We derived a specific panel of markers based on computational analysis of the DNA methylation profiles from The Cancer Genome Atlas (TCGA). Through training (n = 160) and validation set (n = 69), we developed a diagnostic prediction model with 26 markers, which yielded a sensitivity of 89.37% and a specificity of 100% for differentiating malignant disease from normal lesions [AUROC = 0.9816 (95% CI: 96.09-100%), and AUPRC = 0.9704 (95% CI: 94.54–99.46%)]. A simplified 4-marker model including cg23035715, cg16304215, cg20072171, and cg21501525 had a similar diagnostic power [AUROC = 0.9796 (95% CI: 95.56–100%), and AUPRC = 0.9220 (95% CI: 91.02–94.37%)]. We found that a single cfDNA methylation marker, cg23035715, has a high diagnostic power [AUROC = 0.9395 (95% CI: 89.72–99.27%), and AUPRC = 0.9111 (95% CI: 88.45–93.76%)], with a sensitivity of 84.90% and a specificity of 93.88%. In an independent testing dataset (n = 104), the obtained diagnostic prediction model discriminated BC patients from normal controls with high accuracy [AUROC = 0.9449 (95% CI: 90.07–98.91%), and AUPRC = 0.8640 (95% CI: 82.82–89.98%)]. We compared the diagnostic power of cfDNA methylation and mammography. Our model yielded a sensitivity of 94.79% (95% CI: 78.72–97.87%) and a specificity of 98.70% (95% CI: 86.36–100%) for differentiating malignant disease from normal lesions [AUROC = 0.9815 (95% CI: 96.75–99.55%), and AUPRC = 0.9800 (95% CI: 96.6–99.4%)], with better diagnostic power and had better diagnostic power than that of using mammography [AUROC = 0.9315 (95% CI: 89.95–96.34%), and AUPRC = 0.9490 (95% CI: 91.7–98.1%)]. In addition, hypermethylation profiling provided insights into lymph node metastasis stratifications (p < 0.05). In conclusion, we developed and tested a cfDNA methylation model for BC diagnosis with better performance than mammography.

Relative contribution of vitamin D deficiency to subclinical atherosclerosis in Indian context: Preliminary findings.

Asian Indians have a genetic predisposition to atherothrombotic risk. common carotid intima-media thickness (CCIMT) measured by ultrasound is a quantitative marker for atherosclerotic burden and a derived variable, that is, “CCIMT statistical Z-score (Z-score)” is useful for better quantification. The association between vitamin D deficiency and atherosclerosis is inconclusive. Since, vitamin D deficiency is highly prevalent in India, there is a need to study its relative contribution to subclinical atherosclerotic burden.This prospective cross-sectional study (n = 117) in apparently healthy individuals aged 20 to 60 years sought to identify the determinants of CCIMT Z score with CCIMT measured by “echo-tracking” method. A multivariable linear regression analysis was done with CCIMT Z score as dependent variable and the following as independent variables: age, body mass index, waist-to-height ratio, total cholesterol to HDL ratio (TC-HDL ratio), serum vitamin D3 levels (ng/mL), sex, diabetes mellitus, current cigarette smoking status. A diagnostic prediction model was also developed with a threshold value of 1.96 for CCIMT Z score.The mean (SD) for calendar age (y) was 40 (8). There were 26 (22.22%) individuals in sample with CCIMT Z score ≥1.96 (advanced stage) of whom 14 (23.33%) were <40 y (n = 60). The mean score was 1.28 (90th percentile) in the entire sample. Vitamin D3 deficiency with a mean (SD) blood level (ng/mL) of 14.3 (6.4) was noted and prevalence of deficiency was 81%. The final model wasCCIMT Z-score = 0.80 + (0.841 × current smoking = 1) + (0.156 × TC-HDL ratio) – (0.0263 × vitamin D3 blood level in ng/mL).The decreasing order of association is smoking, TC-HDL ratio, and vitamin D3. With the model, likelihood ratio (95% CIs) was better for positive test 3.5 (1.23–9.94) than that for a negative test 0.83 (0.66–1.02).Internal validation with Bootstrap resampling revealed stability of baseline diagnostic variables.There is substantial subclinical atherosclerotic burden in Indian setting with independent contribution by vitamin D deficiency. The model is valuable in “ruling-in” of the underlying advanced atherosclerosis. The study is limited by convenient sampling and lack of external validation of the model.

Clinical features of COVID-19 infection among patients consulting at an emergency service

COVID-19 is a serious public health problem worldwide. To describe the clinical features of COVID-19 infection in adult patients consulting at an Emergency Service. Descriptive prospective study of adult patients with suspected COVID-19 consulting between April 1 and July 31, 2020, at the Emergency Service of a clinical hospital. Clinical features, chronic comorbidities and demographic data were recorded. We assessed 2,958 adult patients aged 42 ± 15 years (46% males). In 54% of them, COVID-19 infection was confirmed, 40% had preexisting diseases, especially hypertension (15%), hypothyroidism (6%), diabetes (6%), asthma (5%) and obesity (6%). The main clinical manifestations associated with COVID-19 were general malaise (79%), anorexia (38%), myalgia (64%), fever (52%), headache (70%), anosmia/dysgeusia (60%), cough (56%), dyspnea (54%) and diarrhea (36%). In the multivariate analysis, the main clinical predictors of COVID-19 infection were malaise, anorexia, fever, myalgia, headache, nasal congestion, cough, expectoration, anosmia/dysgeusia, and history of close contact with a SARS-CoV-2 patient. Odynophagia and chest discomfort were negative predictors of the disease. The history of fever associated with anorexia, cough, and dyspnea or anosmia/dysgeusia and close contact with a SARS-CoV-2 patient had high specificity and positive predictive value for COVID-19 infection. Clinical features of COVID-19 infection were highly unspecific in these patients. Clinical diagnostic prediction models could be useful to support healthcare decision making at primary care setting.

Deep learning for abdominal ultrasound: A computer-aided diagnostic system for the severity of fatty liver.

The prevalence of nonalcoholic fatty liver disease is increasing over time worldwide, with similar trends to those of diabetes and obesity. A liver biopsy, the gold standard of diagnosis, is not favored due to its invasiveness. Meanwhile, noninvasive evaluation methods of fatty liver are still either very expensive or demonstrate poor diagnostic performances, thus, limiting their applications. We developed neural network-based models to assess fatty liver and classify the severity using B-mode ultrasound (US) images. We followed standards for reporting of diagnostic accuracy guidelines to report this study. In this retrospective study, we utilized B-mode US images from a consecutive series of patients to develop four-class, two-class, and three-class diagnostic prediction models. The images were eligible if confirmed by at least two gastroenterologists. We compared pretrained convolutional neural network models, consisting of visual geometry group (VGG)19, ResNet-50 v2, MobileNet v2, Xception, and Inception v2. For validation, we utilized 20% of the dataset resulting in >100 images for each severity category. There were 21,855 images from 2,070 patients classified as normal (N = 11,307), mild (N = 4,467), moderate (N = 3,155), or severe steatosis (N = 2,926). We used ResNet-50 v2 for the final model as the best ones. The areas under the receiver operating characteristic curves were 0.974 (mild steatosis vs others), 0.971 (moderate steatosis vs others), 0.981 (severe steatosis vs others), 0.985 (any severity vs normal), and 0.996 (moderate-to-severe steatosis/clinically abnormal vs normal-to-mild steatosis/clinically normal). Our deep learning models achieved comparable predictive performances to the most accurate, yet expensive, noninvasive diagnostic methods for fatty liver. Because of the discriminative ability, including for mild steatosis, significant impacts on clinical applications for fatty liver are expected. However, we need to overcome machine-dependent variation, motion artifacts, lacking of second confirmation from any other tools, and hospital-dependent regional bias.

Predictive model for atrial fibrillation in hypertensive diabetic patients.

Several scores to identify patients at high risk of suffering atrial fibrillation have been developed. Their applicability in hypertensive diabetic patients, however, remains uncertain. Our aim is to develop and validate a diagnostic predictive model to calculate the risk of developing atrial fibrillation at five years in a hypertensive diabetic population. The derivation cohort consisted of patients with both hypertension and diabetes attended in any of the 52 primary healthcare centres of Barcelona; the validation cohort came from the 11 primary healthcare centres of Terres de l'Ebre (Catalonia South) from January 2013 to December 2017. Multivariable Cox regression identified clinical risk factors associated with the development of atrial fibrillation. The overall performance, discrimination and calibration of the model were carried out. The derivation data set comprised 54575 patients. The atrial fibrillation rate incidence was 15.3 per 1000 person/year. A 5-year predictive model included age, male gender, overweight, heart failure, valvular heart disease, peripheral vascular disease, chronic kidney disease, number of antihypertensive drugs, systolic and diastolic blood pressure, heart rate, thromboembolism, stroke and previous history of myocardial infarction. The discrimination of the model was good (c-index=0.692; 95% confidence interval, 0.684-0.700), and calibration was adequate. In the validation cohort, the discrimination was lower (c-index=0.670). The model accurately predicts future atrial fibrillation in a population with both diabetes and hypertension. Early detection allows the prevention of possible complications arising from this disease.