Diagnostic Prediction Model Research Articles

A diagnostic prediction model for colorectal cancer in elderlies via internet of medical things.

Internet of Medical Things (IoMT) and embedded systems have improved the healthcare systems by enabling remote monitoring the patients’ health conditions anywhere and anytime especially during novel COVID-19 pandemic. In this paper, an IoT-based predicting model is proposed to predict colorectal cancer (CRC) in elderlies. It provides a CRC predicting model for the involved medical team to continuously trace an elderly’s biological indicators using smart wearable embedded systems and medical IoT devices. In this model, vital medical data is collected by IoMT devices and sensors, then analytical information is derived via machine learning (ML) methods for early CRC diagnosis and elderly’s health parameters changes. The experimental results confirm that the suggested model meets the proper accuracy of predicting the CRC in aged people.

Diagnosis of Early-Stage Non-Small Cell Lung Cancer Using DNA Methylation in Tissue and Plasma

Background: To explore the diagnostic value of DNA methylation in early-stage lung cancer using tissue and plasma. Methods: A tissue cohort (n=100) and a plasma cohort (n=146) of patients diagnosed with lung cancer and benign diseases were enrolled and randomly divided into training and validation groups. Methylation profiles were compared to select lung cancer-specific methylation markers. Diagnostic prediction models were constructed using these markers. Their efficacies were estimated by sensitivity, specificity, and AUC. Findings: Diagnostic models were developed to differentiate lung cancer from benignity. In the tissue cohort, 276 markers were significantly differentially methylated in lung cancer (FDR<0.05). A diagnostic prediction model of 6 markers from these achieved outstanding performances in both training cohort (Sensitivity 90%, Specificity 97%, AUC 0.988) and validation cohort (Sensitivity 92%, Specificity 94%, AUC 0.977). These markers were not significantly differentially methylated in the plasma samples. Thus, another 15 markers were selected independently in the plasma samples, yielding a sensitivity of 98% and a specificity of 100% (AUC 0.998) in the training cohort and a sensitivity of 81% and a specificity of 59% (AUC 0.797) in the validation cohort. Analysis based on the methylation haplotype was also conducted, identifying 1,222 differentially methylated regions in tissue samples (FDR<0.05) that were enriched in DNA replication-related pathways. Correlations between DNA methylation and clinical characteristics revealed significant differential methylation patterns between smokers and nonsmokers (FDR<0.05) in both tissue and plasma. Interpretation: Both tissue and ctDNA methylation can potentially be used as biomarkers for the early diagnosis of lung cancer. Funding: National Natural Science Foundation of China (81871890, 91859203). Declaration of Interest: The authors declare no potential conflicts of interest. Ethical Approval: The clinical protocol was approved by the Ethics Committee of West China Medical Center, and written informed consent was provided by all participants in this research.

Could Ovarian Cancer Prediction Models Improve the Triage of Symptomatic Women in Primary Care? A Modelling Study Using Routinely Collected Data.

Simple SummaryEarlier detection of ovarian cancer has the potential to improve patient outcomes, including survival. However, determining which women presenting in primary care to refer for specialist assessment and investigation is a clinical dilemma. In this study, we used routinely collected English primary care data from 29,962 women with symptoms of possible ovarian cancer who were tested for the ovarian cancer biomarker CA125. We developed diagnostic prediction models to estimate the probability of the disease. A relatively simple model, consisting of age and CA125 level, performed well for the identification of ovarian cancer. Including additional risk factors within the model did not materially improve model performance. Following further validation, this model could be used to help triage symptomatic women in primary care based on their risk of undiagnosed ovarian cancer, identifying those at high risk for urgent specialist investigation and those at lower (but still elevated) risk for non-urgent investigation or monitoring.CA125 is widely used as an initial investigation in women presenting with symptoms of possible ovarian cancer. We sought to develop CA125-based diagnostic prediction models and to explore potential implications of implementing model-based thresholds for further investigation in primary care. This retrospective cohort study used routinely collected primary care and cancer registry data from symptomatic, CA125-tested women in England (2011–2014). A total of 29,962 women were included, of whom 279 were diagnosed with ovarian cancer. Logistic regression was used to develop two models to estimate ovarian cancer probability: Model 1 consisted of age and CA125 level; Model 2 incorporated further risk factors. Model discrimination (AUC) was evaluated using 10-fold cross-validation. The sensitivity and specificity of various model risk thresholds (≥1% to ≥3%) were compared with that of the current CA125 cut-off (≥35 U/mL). Model 1 exhibited excellent discrimination (AUC: 0.94) on cross-validation. The inclusion of additional variables (Model 2) did not improve performance. At a risk threshold of ≥1%, Model 1 exhibited greater sensitivity (86.4% vs. 78.5%) but lower specificity (89.1% vs. 94.5%) than CA125 (≥35 U/mL). Applying the ≥1% model threshold to the cohort in place of the current CA125 cut-off, 1 in every 74 additional women identified had ovarian cancer. Following external validation, Model 1 could be used as part of a ‘risk-based triage’ system in which women at high risk of undiagnosed ovarian cancer are selected for urgent specialist investigation, while women at ‘low risk but not no risk’ are offered non-urgent investigation or interval CA125 re-testing. Such an approach has the potential to expedite ovarian cancer diagnosis, but further research is needed to evaluate the clinical impact and health–economic implications.

Cell-free DNA promoter hypermethylation as a diagnostic marker for pancreatic ductal adenocarcinoma – An external validation study

BackgroundWe recently identified a diagnostic prediction model based on promoter hypermethylation of eight selected genes in plasma cell-free (cf) DNA, which showed promising results as a diagnostic biomarker for pancreatic ductal adenocarcinoma (PDAC). The aim of the present study was to validate this biomarker profile in an external patient cohort and examine any additional effect of serum CA 19-9. MethodsPatients with PDAC (n = 346, stage I-IV) and chronic pancreatitis (n = 25) were included. Methylation-specific PCR of a 28-gene panel was performed on serum cfDNA samples. The previously developed diagnostic prediction model (age>65 years, BMP3, RASSF1A, BNC1, MESTv2, TFPI2, APC, SFRP1 and SFRP2) was validated alone and in combination with serum CA 19-9 in this external patient cohort. ResultsPatients with PDAC had a higher number of hypermethylated genes (mean 8.11, 95% CI 7.70–8.52) than patients with chronic pancreatitis (mean 5.60, 95% CI 4.42–6.78, p = 0.011). Validation of the diagnostic prediction model yielded an AUC of 0.77 (95% CI 0.69–0.84). The combination of serum CA 19-9 and our test had an AUC of 0.93 (95% CI 0.89–0.96) in the primary study and 0.85 (95% CI 0.79–0.91) in the validation study. ConclusionIn this validation study, PDAC was associated with a higher number of hypermethylated genes in serum cfDNA than chronic pancreatitis. Our diagnostic test was superior to the predictive value of serum CA 19-9 alone in both the primary and the validation study. The combination of our test with CA 19-9 may serve as a clinically useful diagnostic biomarker for PDAC.

Artificial Intelligence Applications for COVID-19 in Intensive Care and Emergency Settings: A Systematic Review.

Background: Little is known about the role of artificial intelligence (AI) as a decisive technology in the clinical management of COVID-19 patients. We aimed to systematically review and critically appraise the current evidence on AI applications for COVID-19 in intensive care and emergency settings. Methods: We systematically searched PubMed, Embase, Scopus, CINAHL, IEEE Xplore, and ACM Digital Library databases from inception to 1 October 2020, without language restrictions. We included peer-reviewed original studies that applied AI for COVID-19 patients, healthcare workers, or health systems in intensive care, emergency, or prehospital settings. We assessed predictive modelling studies and critically appraised the methodology and key findings of all other studies. Results: Of fourteen eligible studies, eleven developed prognostic or diagnostic AI predictive models, all of which were assessed to be at high risk of bias. Common pitfalls included inadequate sample sizes, poor handling of missing data, failure to account for censored participants, and weak validation of models. Conclusions: Current AI applications for COVID-19 are not ready for deployment in acute care settings, given their limited scope and poor quality. Our findings underscore the need for improvements to facilitate safe and effective clinical adoption of AI applications, for and beyond the COVID-19 pandemic.

Predictive Analytic Model for Diagnosis of Ectopic Pregnancy.

Objective: Ectopic pregnancy (EP) is a serious condition. Delayed diagnosis could lead to life-threatening outcomes. The study aimed to develop a diagnostic predictive model for EP to approach suspected cases with prompt intervention before the rupture occurred.Methods: A retrospective cross-sectional study enrolled 347 pregnant women presenting first-trimester complications (abdominal pain or vaginal bleeding) with diagnosis suspected of pregnancy of unknown location, who were eligible and underwent chart review. The data including clinical risk factors, signs and symptoms, serum human chorionic gonadotropin (hCG), and ultrasound findings were analyzed. The statistical predictive score was developed by performing logistic regression analysis. The testing data of 30 patients were performed to test the validation of predictive scoring.Results: From a total of 22 factors, logistic regression method–derived scoring model was based on five potent factors (history of pelvic inflammatory disease, current use of emergency pills, cervical motion tenderness, serum hCG ≥1,000 mIU/ml, and ultrasound finding of adnexal mass) using a cutoff score ≥3. This predictive index score was able to determine ectopic pregnancy with an accuracy of 77.8% [95% confidence interval (CI) = 73.1–82.1], specificity of 91.0% (95% CI = 62.1–72.0), sensitivity of 67.0% (95% CI = 88.0–94.0), and area under the curve of 0.906 (95% CI = 0.875–0.937). In the validation group, no patient with negative result of this score had an EP.Conclusion: Statistical predictive score was derived with high accuracy and applicable performance for EP diagnosis. This score could be used to support clinical decision making in routine practice for management of EP.

Predicting regional coastal sea level changes with machine learning

All ocean basins have been experiencing significant warming and rising sea levels in recent decades. There are, however, important regional differences, resulting from distinct processes at different timescales (temperature-driven changes being a major contributor on multi-year timescales). In view of this complexity, it deems essential to move towards more sophisticated data-driven techniques as well as diagnostic and prognostic prediction models to interpret observations of ocean warming and sea level variations at local or regional sea basins. In this context, we present a machine learning approach that exploits key ocean temperature estimates (as proxies for the regional thermosteric sea level component) to model coastal sea level variability and associated uncertainty across a range of timescales (from months to several years). Our findings also demonstrate the utility of machine learning to estimate the possible tendency of near-future regional sea levels. When compared to actual sea-level records, our models perform particularly well in the coastal areas most influenced by internal climate variability. Yet, the models are widely applicable to evaluate the patterns of rising and falling sea levels across many places around the globe. Thus, our approach is a promising tool to model and anticipate sea level changes in the coming (1–3) years, which is crucial for near-term decision making and strategic planning about coastal protection measures.

Predicting Coronary Artery Disease in Primary Care: Development and Validation of a Diagnostic Risk Score for Major Ethnic Groups in Southeast Asia.

Coronary artery disease (CAD) risk prediction tools are useful decision supports. Their clinical impact has not been evaluated amongst Asians in primary care. We aimed to develop and validate a diagnostic prediction model for CAD in Southeast Asians by comparing it against three existing tools. We prospectively recruited patients presenting to primary care for chest pain between July 2013 and December 2016. CAD was diagnosed at tertiary institution and adjudicated. A logistic regression model was built, with validation by resampling. We validated the Duke Clinical Score (DCS), CAD Consortium Score (CCS), and Marburg Heart Score (MHS). Discrimination and calibration quantify model performance, while net reclassification improvement and net benefit provide clinical insights. CAD prevalence was 9.5% (158 of 1658 patients). Our model included age, gender, type 2 diabetes mellitus, hypertension, smoking, chest pain type, neck radiation, Q waves, and ST-T changes. The C-statistic was 0.808 (95% CI 0.776-0.840) and 0.815 (95% CI 0.782-0.847), for model without and with ECG respectively. C-statistics for DCS, CCS-basic, CCS-clinical, and MHS were 0.795 (95% CI 0.759-0.831), 0.756 (95% CI 0.717-0.794), 0.787 (95% CI 0.752-0.823), and 0.661 (95% CI 0.621-0.701). Our model (with ECG) correctly reclassified 100% of patients when compared with DCS and CCS-clinical respectively. At 5% threshold probability, the net benefit for our model (with ECG) was 0.063. The net benefit for DCS, CCS-basic, and CCS-clinical was 0.056, 0.060, and 0.065. PRECISE (Predictive Risk scorE for CAD In Southeast Asians with chEst pain) performs well and demonstrates utility as a clinical decision support for diagnosing CAD among Southeast Asians.

Diagnostic prediction models of stratifying chronic heart failure patients based on the underlying disease

Aim. To develop classification criteria for stratifying congestive heart failure (CHF) patients based on the underlying disease.Methods. 61 patients with CHF were recruited in a study. All patients were assigned to three groups according to the underlying disease: patients with coronary artery disease (CAD) (n = 29), patients with arterial hypertension (AH) (n = 19), and those present with dilated cardiomyopathy (DCM) (n = 13). Patients underwent routine clinical examination. Biochemical and inflammatory markers (IL-6, its soluble receptor sIL-6R, and sgp130) were measured in all patients. The Mann-Whitney U test, the Kruskal-Wallis H test, the Pearson χ2 test, and Fischer exact test were used to analyze the selected variables. Discriminant analysis was used for generating prediction models. The quality of the models was evaluated with the ROC analysis.Results. Statistically significant variables identified by the pairwise comparison of patients with CAD and AH, CAD and DCM, AH and DCM were included in the discriminant analysis along with clinically valid parameters. Clinical prediction models of stratifying patients to different etiological groups were based on these parameters. The optimal cut-off values were determined for each model. The area under the ROC curve (AUC) was used to evaluate the quality of the model. The AUC value for CAD and AH groups was 1, for AH and DCM - 72±0.024, and for CAD and DCM - 0.907±0.053.Conclusion. Diagnostic prediction models were developed using the discriminant analysis. These models allow stratifying CHF patients according to the underlying disease (CAD, AH, and DCM). The ROC curves have confirmed the good classifying quality of the models.

Abstract PD6-06: Radiomic phenotypes from dynamic contrast-enhanced MRI (DCE-MRI) parametric maps for early prediction of response to neoadjuvant systemic therapy (NAST) in triple negative breast cancer (TNBC) patients

Abstract Background and Purpose:Early and accurate assessment ofbreast cancer response to NAST is important for patient management. In this study, we investigated the value of radiomic phenotypes derived from semi-quantitative and quantitative DCE-MRI parametric maps for early prediction of NASTresponse in TNBC patients. MATERIALS AND METHODS:This IRB approved study included 74 patients with stage I-III TNBC who were enrolled in the prospective ARTEMIS trial (NCT02276443). Pathologic complete response (pCR) and non-pCR were assessed by surgical histopathology after NAST (pCR=34; non-pCR=40).MRI scans were obtained at 3 time points during the NAST treatment with every 2-week anthracycline-based chemotherapy (AC): at baseline (BSL=74), post-2 cycles of AC (C2= 27) and post-4 cycles of AC (C4= 27). Patients went on to receive taxane-based chemotherapy prior to surgery. Tumor regions of interest (ROIs) were segmented by a breast radiologist at the early-phase subtractions of DCE-MRI scans using in-house developed software, followed by co-registration of the ROIs with quantitative (Ktrans, Veand Kep), and semi-quantitative DCE parametric maps (Maximum Slope Increase (MSI), Positive Enhancement Integral (PEI) and Peak Signal Enhancement Ratio (SER)).A total of 93 first order radiomic features were extracted from the tumor ROIs of each time point semi-quantitative DCE parametric map, while a total of 390 extracted radiomic features (first order-histogram features and second order grey-level-co-occurrence matrix) were extracted from each quantitative DCE parametric map using an in-house developed Matlab software.Radiomic features at each time point and changes between the 3 time points were compared between pCR and non-pCR using Wilcoxon Rank Sum test and Fisher’s exact test. Area under the receiver operating characteristics curve (AUC) was used to determine which features predicted pCR.Logistic regression was performed for feature selection, and used to build the radiomic phenotype model. The model performance was assessed by leave-one-out cross validation and 3-fold cross validation. RESULTS:Thirty-three radiomic features from PEI map were significantly different between pCR and non-pCR. The PEI most significant features were changesbetween BSL and C4 in skewness, mean and median (AUC=0.87, 0.85 and 0.87, p=&amp;lt;0.001, 0.001 and 0.002 respectively). Additionally, 31 MSI features were significantly different between pCR and non-pCR. The top 2 features were the interscan-change in skewness between BSL and C2 (AUC=0.80, P=0.007) and C4 standard deviation (AUC=0.80, P=0.006). Four BSL Veradiomic features were statistically significant between pCR and non-pCR with the best being range of difference variance (AUC=0.64, P=0.03). One BSL Kepfeature (Angular-Variance of Information measure of correlation-2) was able to differentiate pCR from non-pCR (AUC=0.64, P=0.04). Five C4-Ktrans features were able to differentiate pCR and non-pCR, with the most significant being mean value (AUC=0.86, P=0.001). BSL-Kepradiomic model built from 24 features (AUC=0.80, p=0.003) and combined (Ktrans, Veand Kep)C2-radiomic model consisting of 20 features (AUC=0.97, p=0.01) showed the best performance for prediction of pCR. CONCLUSIONS:Radiomic phenotypes form DCE-MRI parametric maps were useful for differentiation between pCR and non-pCR and showed promise as noninvasive imaging biomarkers for early prediction of NAST response in TNBC. Potentially, DCE-MRI radiomic features may be used for development of diagnostic predictive model for early noninvasive assessment of NAST treatment response in TNBC patients. Citation Format: Nabil Elshafeey, Beatriz E Adrada, Rosalind P Candelaria, Abeer H Abdelhafez, Benjamin C Musall, Jia Sun, Medine Boge, Rania M.M Mohamed, Hagar S Mahmoud, Jong Bum Son, Aikaterini Kotrosou, Shu Zhang, Jessica Leung, Deanna Lane, Marion Scoggins, David Spak, Elsa Arribas, Lumarie Santiago, Gary J. Whitman, Huong T Le-Petross, Tanya W Moseley, Jason B White, Elizabeth Ravenberg, Ken-Pin Hwang, Peng Wei, Jennifer K Litton, Lei Huo, Debu Tripathy, Vicente Valero, Alastair M Thompson, Stacy Moulder, Wei T Yang, Mark D Pagel, Jingfei Ma, Gaiane M Rauch. Radiomic phenotypes from dynamic contrast-enhanced MRI (DCE-MRI) parametric maps for early prediction of response to neoadjuvant systemic therapy (NAST) in triple negative breast cancer (TNBC) patients [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PD6-06.

Biomarkers to Distinguish Bacterial From Viral Pediatric Clinical Pneumonia in a Malaria-Endemic Setting.

BackgroundDifferential etiologies of pediatric acute febrile respiratory illness pose challenges for all populations globally, but especially in malaria-endemic settings because the pathogens responsible overlap in clinical presentation and frequently occur together. Rapid identification of bacterial pneumonia with high-quality diagnostic tools would enable appropriate, point-of-care antibiotic treatment. Current diagnostics are insufficient, and the discovery and development of new tools is needed. We report a unique biomarker signature identified in blood samples to accomplish this.MethodsBlood samples from 195 pediatric Mozambican patients with clinical pneumonia were analyzed with an aptamer-based, high-dynamic-range, quantitative assay (~1200 proteins). We identified new biomarkers using a training set of samples from patients with established bacterial, viral, or malarial pneumonia. Proteins with significantly variable abundance across etiologies (false discovery rate <0.01) formed the basis for predictive diagnostic models derived from machine learning techniques (Random Forest, Elastic Net). Validation on a dedicated test set of samples was performed.ResultsSignificantly different abundances between bacterial and viral infections (219 proteins) and bacterial infections and mixed (viral and malaria) infections (151 proteins) were found. Predictive models achieved >90% sensitivity and >80% specificity, regardless of number of pathogen classes. Bacterial pneumonia was strongly associated with neutrophil markers—in particular, degranulation including HP, LCN2, LTF, MPO, MMP8, PGLYRP1, RETN, SERPINA1, S100A9, and SLPI.ConclusionsBlood protein signatures highly associated with neutrophil biology reliably differentiated bacterial pneumonia from other causes. With appropriate technology, these markers could provide the basis for a rapid diagnostic for field-based triage for antibiotic treatment of pediatric pneumonia.

Recent Advances and Future Directions of Diagnostic and Prognostic Prediction Models in Ovarian Cancer

Ovarian cancer has one of the highest mortality rates among gynecological malignancies. This disease has a low early detection rate, a high postoperative recurrence rate, and a 5-year survival rate of only 40%. Hence, there is an urgent need to improve the early diagnosis and prognosis of ovarian cancer. Prediction models can effectively estimate the risk of disease occurrence, as well as its prognosis. Recently, many studies have established multiple ovarian cancer prediction models based on different regions and populations. These models can improve the detection rate and optimize the prognosis management to a certain extent. Herein, the construction principle of the ovarian cancer risk prediction model and its validation are summarized; furthermore, comprehensive reviews and comparisons of the different types of these models are made. Therefore, our review may be of great significance for the whole course of ovarian cancer management.

Metal-Organic Framework Hybrids Aid Metabolic Profiling for Colorectal Cancer.

Colorectal cancer (CRC) is the third most common fatal cancer worldwide, accounting for ≈10% of cancer-related mortality. Metabolic shift occurs from the very early stage during the development of CRC, which is of significant etiological and diagnostic importance toward precision medicine. Here, an advanced molecular tool to characterize the metabolic alterations in CRC, based on metal-organic framework (MOF) hybrids is reported. Consuming only 500 nL of plasma without any sample pretreatment, MOF hybrids yield direct metabolic fingerprints by laser desorption/ionization mass spectrometry in seconds. A diagnostic prediction model by a machine learning algorithm is constructed, to discriminate CRC patients from normal controls with an average area under the curve of 0.947 for the discovery cohort and 0.912 for the independent validation cohort. In addition, CRC-specific metabolic signature consisting of 34 potential biomarkers, based on the aforementioned diagnostic model is identified. The results advance the design of nanomaterial-based platforms for metabolic analysis and establish a new liquid biopsy tool for CRC screening compatible with the current clinical workflow in practice.

Neonatal cholestasis: development of a diagnostic decision algorithm from multivariate predictive models.

Despite the recent advances involving molecular studies, the neonatal cholestasis (NC) diagnosis still relays on the expertise of medical teams. Our aim was to develop models of etiological diagnosis and unfavourable prognosis which may support a rationale diagnostic approach. We retrospectively analysed 154 patients born between January 1985 and October 2019. The cohort was divided into two main groups: (A) transient cholestasis and (B) other diagnosis (with subgroups) and also in two groups of outcomes: (I) unfavourable and (II) favourable. Multivariate logistic regression analysis identified the lower gestational age as the only variable independently associated with an increased risk of transient cholestasis and signs and/or symptoms of sepsis with infectious or metabolic diseases. Gamma-glutamyl transferase serum levels > 300 IU/L had a positive predictive value for both diagnosis of biliary atresia and for alpha-1-antitrypsin deficiency (A1ATD) and for unfavourable prognosis. A model of diagnosis for A1ATD (n = 34) showed an area under the ROC curve = 0.843 [confidence interval (CI): 0.773-0.912].Conclusion: This study identified some predictors of diagnosis and prognosis which helped to build a diagnostic decision algorithm. The unusually large subgroup of patients with A1ATD in this cohort emphasizes its predictive diagnostic model. What Is Known • The etiological diagnosis of neonatal cholestasis (NC) requires a step-by-step guided approach, and diagnostic models have been developed only for biliary atresia. • Current algorithms neither address the epidemiology changes nor the application of the new molecular diagnostic tools. What Is New • This study provides diagnostic predictive models for patients with A1ATD, metabolic/infectious diseases, and transient cholestasis, and two models of unfavourable prognosis for NC. • A diagnostic decision algorithm is proposed based on this study, authors expertise and the literature.

Distinguishing brain abscess from necrotic glioblastoma using MRI-based intranodular radiomic features and peritumoral edema/tumor volume ratio.

A correct preoperative diagnosis is essential for the treatment and prognosis of necrotic glioblastoma and brain abscess, but the differentiation between them remains challenging. We constructed a diagnostic prediction model with good performance and enhanced clinical applicability based on data from 86 patients with necrotic glioblastoma and 32 patients with brain abscess that were diagnosed between January 2012 and January 2020. The diagnostic values of three regions of interest based on contrast-enhanced T1 weighted images (including whole tumor, brain-tumor interface, and an amalgamation of both regions) were compared using Logistics Regression and Random Forest. Feature reduction based on the optimal regions of interest was performed using principal component analysis with varimax rotation. The performance of the classifiers was assessed by receiver operator curves. Finally, clinical predictors were utilized to detect the diagnostic power. The mean area under curve (AUC) values of the whole tumor model was significantly higher than other two models obtained from Brain-Tumor Interface (BTI) and combine regions both in training (AUC mean = 0.850) and test/validation set (AUC mean = 0.896) calculated by Logistics Regression and in the testing set (AUC mean = 0.876) calculated by Random Forest. Among these three diagnostic prediction models, the combined model provided superior discrimination performance and yielded an AUC of 0.993, 0.907, and 0.974 in training, testing, and combined datasets, respectively. Compared with the brain-tumor interface and the combined regions, features obtained from the whole tumor showed the best differential value. The radiomic features combined with the peritumoral edema/tumor volume ratio provided the prediction model with the greatest diagnostic performance.

MiRNA based tumor mutation burden diagnostic and prognostic prediction models for endometrial cancer

ABSTRACT Uterus Corpus Endometrial cancer (UCEC) is the sixth most common malignant tumor worldwide. In this research, we identified diagnostic and prognostic biomarkers to reflect patients’ immune microenvironment and prognostic. Various data of UCEC patients from the TCGA database were obtained. Firstly, patients were divided into a high tumor mutation burden (TMB) level group and a low TMB level group according to the level of TMB. Then, differentially expressed miRNAs between the two groups were obtained. LASSO logistic regression analysis was used to construct a diagnostic model to predict the level of TMB. Univariate, multivariate, and LASSO regression analysis were used to construct a prognostic risk signature (PRS) to predict the prognosis of UCEC patients. Twenty-one miRNAs were used to construct a diagnostic model for predicting TMB levels. The AUC values of ROC curves for 21-miRNA-based diagnostic models were 0.911 in the training set, 0.827 in the test set, and 0.878 in the entire set. This diagnostic model showed positive correlation with TMB, PDL1 expression, and the infiltration of immune cells. In addition, three prognostic miRNAs were finally used to construct the PRS. The PRS was related to the expression of multiple immune checkpoints and the infiltration of multiple immune cells. Furthermore, the PRS can also reflect the response to some commonly used chemotherapy regimens. We have established a miRNA-based diagnostic model and a prognostic model that can predict the prognosis of UCEC patients and their response to chemotherapy and immunotherapy, thus providing valuable information on the choice of treatment regimen.

Predictive model using hemoglobin, albumin, fibrinogen, and neutrophil-to-lymphocyte ratio to distinguish patients with colorectal cancer from those with benign adenoma.

Colorectal cancer (CRC) is associated with inflammation, activation of coagulation, and mild anemia. Hematological parameters reflecting ongoing cancer may have the potency to be effective for early diagnostics of CRC. The aim of this study was to examine the validity and relationship between some biochemical and hematological parameters for the early diagnosis of CRC. We designed a prospective observational cohort study to analyze whether these laboratory markers have the potency to distinguish benign tumors from malignant before planned surgery. The clinical data were collected from 208 patients with suspected benign or malignant colorectal tumors who were eligible for elective surgery between September 2018 and January 2020. Blood samples were collected one day before surgery, examined parameters included: complete blood count, hemoglobin (HGB) concentration, albumin (ALB), C-reactive protein (CRP), interleukin-6 (IL-6), and fibrinogen (FG). Absolute neutrophil and lymphocyte counts were used for the calculation of the neutrophil-to-lymphocyte ratio (NLR). The diagnosis was confirmed by histopathological examination. The 197 patients were divided into the group of benign (B group = 52 patients) or malignant tumors (CRC group = 145 patients). ROC curves and cut-off values of NLR, HGB, FG, and ALB concentration, serum CRP and IL-6 levels. In the cohort of 197 adult patients submitted for the elective colorectal surgery, the complete blood samples drawn one day before surgery showed significant differences between patients with benign tumors and colorectal carcinoma: HGB (mean 139.9 g/l vs. 129.9; p<0.001), FG (mean 3.36 g/l vs. 3.94 g/l; p<0.001), ALB (mean 43.4 g/l vs. 41.1 g/l; p=0.001), NLR (mean 2.73 vs. 3.88; p=0.016), respectively. CRP (mean 2.9 mg/l vs. 4.4 mg/l; p=0.011), thrombocyte count (mean 235×109/l vs. 265×109/l; p<0.029). Differences in IL-6 concentrations were not significant (2.9 pg/ml vs. 4.15 pg/ml; p=0.052). Using multivariable logistic regression analysis, four valid parameters (HGB, FG, ALB, and NLR) were suitable for the construction of a diagnostic predictive model for the identification of CRC. In conclusion, a panel of routinely examined blood parameters like HGB, FG, ALB, and NLR has the potency to distinguish patients with benign tumors from malignant by applying a diagnostic predictive model for early laboratory detection of CRC.

Preliminary Exploration of the Initial Diagnostic Prediction Model of Moderate Coronavirus Disease 2019 (2019-nCoV) Based on Clinical Data

Objective: To explore those differences and relationships of the initial diagnostic clinical data between confirmed cases of 2019-nCoV and suspected cases of COVID-19, and then to establish prediction models for predicting the probability of the first diagnosis of 2019-nCoV. Methods: A total of 81 suspected cases and 87 confirmed cases of moderate 2019-nCoV diagnosed initially in the isolation wards of the First People’s Hospital of Wuhu and the People’s Hospital of Wuwei and Wuhan Caidian Module Hospital with the help of our hospital doctors were gathered, and retrospectively analyzed. Results: The most common symptoms were fever (76.79%) and cough (64.29%) in the total of 168 cases. The median age was 45 (35 - 56) years old in the 87 confirmed cases of moderate 2019-nCoV, older than the median age 36 (29 - 50) in the 81 suspected cases. There were significant more in the former than in the latter in the incidence of myalgia, ground-glass opacity (GGO), invasions of lesion in the peripheral lobes, vascular thickening and bronchial wall thickening, interlobular septal thicking, and small pulmonary nodules. On the contrary, there were less in the former than in the latter in the total number of leukocytes and neutrophils in blood routine examination and the levels of procalcitonin (PCT). Two groups were statistically significantly different (P P = ex/(1 + ex), x = −6.226 + (0.071 × ages) + (1.720 × fever) + (2.858 × myalgia) + (2.131 × GGO) + (3.000 × vascular thickening and bron-chial wall thickening) + (3.438 × invasions of lesion in the peripheral lobes) + (−0.304 × the number of leukocytes) + (−1.478 × cough) + (−1.830 × pharyngalgia) + (−2.413 × headache), where e is a natural logarithm. The area under the ROC curve (AUC) of this model was calculated to be 0.945 (0.915 - 0.976). The sensitivity is 0.920 and the specificity is 0.827 when the appropriate critical point is 0.360. Conclusions: A mathematical equation prediction model for predicting the probability of the first diagnosis of 2019-nCoV can be established based on the initial diagnostic clinical data of moderate 2019-nCoV. The prediction model is a good assistant diagnostic method for its high accurateness.

Integrated Analysis of Gene Co-Expression Network and Prediction Model Development Indicates Immune-Related Roles of the Identified Biomarkers in Sepsis and Sepsis-Induced Acute Respiratory Distress Syndrome

Background: Sepsis has emerged as a life-threatening condition and acute respiratory distress syndrome (ARDS) induced by sepsis has added more massive destruction for the limitation of effective prevention and therapy. However, well-described studies of analysis-based researches especially bioinformatics analysis on revealing specific targets and underlying molecular mechanisms of sepsis and sepsis-induced ARDS (sepsis/ARDS) still remain limited and delayed despite the era of data-driven medicine. This study might provide a novel insight into the pathophysiology of sepsis/ARDS. Methods: Weight gene co-expression network based on data from the public database was constructed and followed by functional enrichment analysis. By performing differential expression analysis, least absolute shrink and selection operator (LASSO) method and multivariable logistic regression analysis, candidate biomarkers were identified. risk nomogram and various machine learning algorithms were employed to develop diagnostic prediction models, which were then assessed and validated by receiver operating characteristic (ROC) curve analysis. Analysis of marker gene expressions and immune cell infiltration was performed to further evaluate their correlations during the development of sepsis/ARDS. Findings: 2 key modules and 36 hub genes were identified and extensive immune-involved enrichment was significantly associated with sepsis/ARDS. 4 biomarkers (SIGLEC9, TSPO, CKS1B and PTTG3P) were identified as candidate biomarkers and a 4-gene-based prediction model were then developed and proved to be of great efficiency and performance for sepsis/ARDS. The expressions of both SIGLEC9 and TSPO were regulated significantly and exhibit a notable correlation with the stages of sepsis/ARDS. CD4 memory T cells, mast cells and monocytes might be involved in the process of sepsis/ARDS. Besides, SIGLEC9 and TSPO were significantly correlated with the infiltration of the above immune cells. Interpretation: Our study identified 4 genes, SIGLEC9, TSPO, CKS1B and PTTG3P, and established 4-gene-based prediction model for sepsis/ARDS. The relevant gene co-expression network and immune signatures might provide comprehensive insights into integrated analysis for efficient risk prediction, relatively accurate genetic diagnosis and potential immune-mediated molecular intervention of sepsis/ARDS. Funding: The present study was supported by grants from the National Natural Science Foundation of China (NO. 81970722, 81901952 and 82172144). Declaration of Interest: The authors declared no conflict of interest in the study.

Applying Convolutional Neural Networks to Predict the ICD-9 Codes of Medical Records.

The International Statistical Classification of Disease and Related Health Problems (ICD) is an international standard system for categorizing and reporting diseases, injuries, disorders, and health conditions. Most previously-proposed disease predicting systems need clinical information collected by the medical staff from the patients in hospitals. In this paper, we propose a deep learning algorithm to classify disease types and identify diagnostic codes by using only the subjective component of progress notes in medical records. In this study, we have a dataset, consisting of about one hundred and sixty-eight thousand medical records, from a medical center, collected during 2003 and 2017. First, we apply standard text processing procedures to parse the sentences and word embedding techniques for vector representations. Next, we build a convolution neural network model on the medical records to predict the ICD-9 code by using a subjective component of the progress note. The prediction performance is evaluated by ten-fold cross-validation and yields an accuracy of 0.409, recall of 0.409 and precision of 0.436. If we only consider the “chapter match” of ICD-9 code, our model achieves an accuracy of 0.580, recall of 0.580, and precision of 0.582. Since our diagnostic code prediction model is solely based on subjective components (mainly, patients’ self-report descriptions), the proposed approach could serve as a remote and self-diagnosis assistance tool, prior to seeking medical advice or going to the hospital. In addition, our work may be used as a primary evaluation tool for discomfort in the rural area where medical resources are restricted.