scholarly journals A highly predictive signature of cognition and brain atrophy for progression to Alzheimer’s dementia

2018 ◽  
Author(s):  
Angela Tam ◽  
Christian Dansereau ◽  
Yasser Itturia-Medina ◽  
Sebastian Urchs ◽  
Pierre Orban ◽  
...  
Keyword(s):  
Machine Learning ◽  
Clinical Trials ◽  
High Risk ◽  
Positive Predictive Value ◽  
Predictive Value ◽  
High Specificity ◽  
Alzheimer’S Dementia ◽  
Learning Tools ◽  
Operating Characteristics ◽  
Alzheimer's Dementia

AbstractClinical trials in Alzheimer’s disease need to enroll patients whose cognition will decline over time, if left untreated, in order to demonstrate the efficacy of an intervention. Machine learning models used to screen for patients at risk of progression to dementia should therefore favor specificity (detecting only progressors) over sensitivity (detecting all progressors), especially when the prevalence of progressors is low. Here, we explore whether such high-risk patients can be identified using cognitive assessments and structural neuroimaging, by training machine learning tools in a high specificity regime. A multimodal signature of Alzheimer’s dementia was first extracted from ADNI1. We then validated the predictive value of this signature on ADNI1 patients with mild cognitive impairment (N=235). The signature was optimized to predict progression to dementia over three years with low sensitivity (55.1%) but high specificity (95.6%), resulting in only moderate accuracy (69.3%) but high positive predictive value (80.4%, adjusted for a “typical” 33% prevalence rate of true progressors). These results were replicated in ADNI2 (N=235), with 87.8% adjusted positive predictive value (96.7% specificity, 47.3% sensitivity, 85.1% accuracy). We found that cognitive measures alone could identify high-risk individuals, with structural measurements providing a slight improvement. The signature had comparable receiver operating characteristics to standard machine learning tools, yet a marked improvement in positive predictive value was achieved over the literature by selecting a high specificity operating point. The multimodal signature can be readily applied for the enrichment of clinical trials.

scholarly journals Comparability of Event Adjudication Versus Administrative Billing Claims for Outcome Ascertainment in the DAPT Study

2021 ◽  
Vol 14 (1) ◽  
Author(s):  
Kamil F. Faridi ◽  
Hector Tamez ◽  
Neel M. Butala ◽  
Yang Song ◽  
Changyu Shen ◽  
...  
Keyword(s):  
Myocardial Infarction ◽  
Clinical Trials ◽  
Positive Predictive Value ◽  
Negative Predictive Value ◽  
Predictive Value ◽  
Claims Data ◽  
High Specificity ◽  
Kappa Statistics ◽  
Administrative Claims ◽  
Medicare Claims

Background: Data from administrative claims may provide an efficient alternative for end point ascertainment in clinical trials. However, it is uncertain how well claims data compare to adjudication by a clinical events committee in trials of patients with cardiovascular disease. Methods: We matched 1336 patients ≥65 years old who received percutaneous coronary intervention in the DAPT (Dual Antiplatelet Therapy) Study with the National Cardiovascular Data Registry CathPCI Registry linked to Medicare claims as part of the EXTEND (Extending Trial-Based Evaluations of Medical Therapies Using Novel Sources of Data) Study. Adjudicated trial end points were compared with Medicare claims data with International Classification of Diseases, Ninth Revision codes from inpatient hospitalizations using time-to-event analyses, sensitivity, specificity, positive predictive value, negative predictive value, and kappa statistics. Results: At 21-month follow-up, the cumulative incidence of major adverse cardiovascular and cerebrovascular events (combined mortality, myocardial infarction, and stroke) was similar between trial-adjudicated events and claims data (7.9% versus 7.2%, respectively; P =0.50). Bleeding rates were lower using adjudicated events compared with claims (5.0% versus 8.6%, respectively; P <0.001). The sensitivity and positive predictive value of comprehensive billing codes for identifying adjudicated events were 65.6% and 85.7% for myocardial infarction, 61.5% and 47.1% for stroke, and 76.8% and 39.3% for bleeding, respectively. Specificity and negative predictive value for all outcomes ranged from 93.7% to 99.5%. All 39 adjudicated deaths were identified using Medicare data. Kappa statistics assessing agreement between events for myocardial infarction, stroke, and bleeding were 0.73, 0.52, and 0.49, respectively. Conclusions: Claims data had moderate agreement with adjudication for myocardial infarction and poor agreement but high specificity for bleeding and stroke in the DAPT Study. Deaths were identified equivalently. Using claims data in clinical trials could be an efficient way to assess mortality among Medicare patients and may help detect other outcomes, although additional monitoring is likely needed to ensure accurate assessment of events.

scholarly journals Machine Learning to Predict the Likelihood of Acute Myocardial Infarction

Circulation ◽  
2019 ◽  
Vol 140 (11) ◽  
pp. 899-909 ◽  
Author(s):  
Martin P. Than ◽  
John W. Pickering ◽  
Yader Sandoval ◽  
Anoop S.V. Shah ◽  
Athanasios Tsanas ◽  
...  
Keyword(s):  
Machine Learning ◽  
Myocardial Infarction ◽  
High Risk ◽  
Positive Predictive Value ◽  
Predictive Value ◽  
Characteristic Curve ◽  
Training Set ◽  
Test Set ◽  
Risk Patients ◽  
European Society Of Cardiology

Background: Variations in cardiac troponin concentrations by age, sex, and time between samples in patients with suspected myocardial infarction are not currently accounted for in diagnostic approaches. We aimed to combine these variables through machine learning to improve the assessment of risk for individual patients. Methods: A machine learning algorithm (myocardial-ischemic-injury-index [MI 3 ]) incorporating age, sex, and paired high-sensitivity cardiac troponin I concentrations, was trained on 3013 patients and tested on 7998 patients with suspected myocardial infarction. MI 3 uses gradient boosting to compute a value (0–100) reflecting an individual’s likelihood of a diagnosis of type 1 myocardial infarction and estimates the sensitivity, negative predictive value, specificity and positive predictive value for that individual. Assessment was by calibration and area under the receiver operating characteristic curve. Secondary analysis evaluated example MI 3 thresholds from the training set that identified patients as low risk (99% sensitivity) and high risk (75% positive predictive value), and performance at these thresholds was compared in the test set to the 99th percentile and European Society of Cardiology rule-out pathways. Results: Myocardial infarction occurred in 404 (13.4%) patients in the training set and 849 (10.6%) patients in the test set. MI 3 was well calibrated with a very high area under the receiver operating characteristic curve of 0.963 [0.956–0.971] in the test set and similar performance in early and late presenters. Example MI 3 thresholds identifying low- and high-risk patients in the training set were 1.6 and 49.7, respectively. In the test set, MI 3 values were <1.6 in 69.5% with a negative predictive value of 99.7% (99.5–99.8%) and sensitivity of 97.8% (96.7–98.7%), and were ≥49.7 in 10.6% with a positive predictive value of 71.8% (68.9–75.0%) and specificity of 96.7% (96.3–97.1%). Using these thresholds, MI 3 performed better than the European Society of Cardiology 0/3-hour pathway (sensitivity, 82.5% [74.5–88.8%]; specificity, 92.2% [90.7–93.5%]) and the 99th percentile at any time point (sensitivity, 89.6% [87.4–91.6%]); specificity, 89.3% [88.6–90.0%]). Conclusions: Using machine learning, MI 3 provides an individualized and objective assessment of the likelihood of myocardial infarction, which can be used to identify low- and high-risk patients who may benefit from earlier clinical decisions. Clinical Trial Registration: URL: https://www.anzctr.org.au . Unique identifier: ACTRN12616001441404.

Personalized stratification of back to work risk amidst COVID-19: A machine learning approach (Preprint)

2020 ◽  
Author(s):  
Carson Lam ◽  
Jacob Calvert ◽  
Gina Barnes ◽  
Emily Pellegrini ◽  
Anna Lynn-Palevsky ◽  
...  
Keyword(s):  
Machine Learning ◽  
High Risk ◽  
Learning Algorithm ◽  
Severe Disease ◽  
High Specificity ◽  
Population Level ◽  
The United States ◽  
Health Condition ◽  
Available P ◽  
Severe Illness

BACKGROUND In the wake of COVID-19, the United States has developed a three stage plan to outline the parameters to determine when states may reopen businesses and ease travel restrictions. The guidelines also identify subpopulations of Americans that should continue to stay at home due to being at high risk for severe disease should they contract COVID-19. These guidelines were based on population level demographics, rather than individual-level risk factors. As such, they may misidentify individuals at high risk for severe illness and who should therefore not return to work until vaccination or widespread serological testing is available. OBJECTIVE This study evaluated a machine learning algorithm for the prediction of serious illness due to COVID-19 using inpatient data collected from electronic health records. METHODS The algorithm was trained to identify patients for whom a diagnosis of COVID-19 was likely to result in hospitalization, and compared against four U.S policy-based criteria: age over 65, having a serious underlying health condition, age over 65 or having a serious underlying health condition, and age over 65 and having a serious underlying health condition. RESULTS This algorithm identified 80% of patients at risk for hospitalization due to COVID-19, versus at most 62% that are identified by government guidelines. The algorithm also achieved a high specificity of 95%, outperforming government guidelines. CONCLUSIONS This algorithm may help to enable a broad reopening of the American economy while ensuring that patients at high risk for serious disease remain home until vaccination and testing become available.

Symptomatic Treatment of Pain-Related Pediatric Functional Gastrointestinal Disorders in a Biopsychosocial Construct, and a Validity Study of Rome III Criteria

2021 ◽  
Vol 60 (6-7) ◽  
pp. 304-313
Author(s):  
Shailender Madani ◽  
Rohit Madani ◽  
Suchi Parikh ◽  
Ahila Manivannan ◽  
Wilma R. Orellana ◽  
...  
Keyword(s):  
Positive Predictive Value ◽  
Predictive Value ◽  
Functional Gastrointestinal Disorders ◽  
High Specificity ◽  
Symptomatic Treatment ◽  
Gastrointestinal Disorders ◽  
Organic Disease ◽  
Rome Iii Criteria ◽  
Rome Criteria ◽  
Rome Iii

Our study aims to assess improvement with symptomatic treatment of pain-related functional gastrointestinal disorders (FGIDs) in a biopsychosocial construct and evaluate validity of Rome III criteria. Children with chronic abdominal pain diagnosed with an FGID or organic disease were followed for 1 year: 256/334 were diagnosed with an FGID and 78/334 were diagnosed with a possible organic disease due to alarm signs or not meeting Rome III criteria. After 1 year, 251 had true FGID and 46 had organic diseases. Ninety percent of FGID patients improved with symptomatic treatment over an average of 5.4 months. With a 95% confidence interval, Rome criteria predicted FGIDs with sensitivity 0.89, specificity 0.90, positive predictive value 0.98, and negative predictive value 0.59. We conclude that symptomatic treatment of pain-related FGIDs results in clinical improvement and could reduce invasive/expensive testing. Rome III criteria’s high specificity and positive predictive value suggest they can rule in a diagnosis of FGID.

scholarly journals Serum D-dimer level as a biomarker for identifying patients with isolated injury to prevent unnecessary whole-body computed tomography in blunt trauma care

2021 ◽  
Vol 29 (1) ◽  
Author(s):  
Rakuhei Nakama ◽  
Ryo Yamamoto ◽  
Yoshimitsu Izawa ◽  
Keiichi Tanimura ◽  
Takashi Mato
Keyword(s):  
Computed Tomography ◽  
Positive Predictive Value ◽  
Predictive Value ◽  
High Specificity ◽  
Whole Body ◽  
Cutoff Value ◽  
Whole Body Ct ◽  
Isolated Injury ◽  
D Dimer ◽  
Body Ct

Abstract Background Unnecessary whole-body computed tomography (CT) may lead to excess radiation exposure. Serum D-dimer levels have been reported to correlate with injury severity. We examined the predictive value of serum D-dimer level for identifying patients with isolated injury that can be diagnosed with selected-region CT rather than whole-body CT. Methods This single-center retrospective cohort study included patients with blunt trauma (2014–2017). We included patients whose serum D-dimer levels were measured before they underwent whole-body CT. “Isolated” injury was defined as injury with Abbreviated Injury Scale (AIS) score ≤ 5 to any of five regions of interest or with AIS score ≤ 1 to other regions, as revealed by a CT scan. A receiver operating characteristic curve (ROC) was drawn for D-dimer levels corresponding to isolated injury; the area under the ROC (AUROC) was evaluated. Sensitivity, specificity, positive predictive value, and negative predictive value were calculated for several candidate cut-off values for serum D-dimer levels. Results Isolated injury was detected in 212 patients. AUROC was 0.861 (95% confidence interval [CI]: 0.815–0.907) for isolated injury prediction. Serum D-dimer level ≤ 2.5 μg/mL was an optimal cutoff value for predicting isolated injury with high specificity (100.0%) and positive predictive value (100.0%). Approximately 30% of patients had serum D-dimer levels below this cutoff value. Conclusion D-dimer level ≤ 2.5 μg/mL had high specificity and high positive predictive value in cases of isolated injury, which could be diagnosed with selected-region CT, reducing exposure to radiation associated with whole-body CT.

Identifying children at high risk for infection-related decompensation using a predictive emergency department-based electronic assessment tool

Diagnosis ◽  
2020 ◽  
Vol 0 (0) ◽  
Author(s):  
Robert J. Sepanski ◽  
Arno L. Zaritsky ◽  
Sandip A. Godambe
Keyword(s):  
Emergency Department ◽  
High Risk ◽  
Positive Predictive Value ◽  
Firing Rate ◽  
Predictive Value ◽  
Assessment Tool ◽  
Predictive Tool ◽  
High Severity ◽  
Electronic Assessment ◽  
Ed Visits

AbstractObjectivesElectronic alert systems to identify potential sepsis in children presenting to the emergency department (ED) often either alert too frequently or fail to detect earlier stages of decompensation where timely treatment might prevent serious outcomes.MethodsWe created a predictive tool that continuously monitors our hospital’s electronic health record during ED visits. The tool incorporates new standards for normal/abnormal vital signs based on data from ∼1.2 million children at 169 hospitals. Eighty-two gold standard (GS) sepsis cases arising within 48 h were identified through retrospective chart review of cases sampled from 35,586 ED visits during 2012 and 2014–2015. An additional 1,027 cases with high severity of illness (SOI) based on 3 M’s All Patient Refined – Diagnosis-Related Groups (APR-DRG) were identified from these and 26,026 additional visits during 2017. An iterative process assigned weights to main factors and interactions significantly associated with GS cases, creating an overall “score” that maximized the sensitivity for GS cases and positive predictive value for high SOI outcomes.ResultsTool implementation began August 2017; subsequent improvements resulted in 77% sensitivity for identifying GS sepsis within 48 h, 22.5% positive predictive value for major/extreme SOI outcomes, and 2% overall firing rate of ED patients. The incidence of high-severity outcomes increased rapidly with tool score. Admitted alert positive patients were hospitalized nearly twice as long as alert negative patients.ConclusionsOur ED-based electronic tool combines high sensitivity in predicting GS sepsis, high predictive value for physiologic decompensation, and a low firing rate. The tool can help optimize critical treatments for these high-risk children.

Hepatitis C virus core antigen assay: an alternative method for hepatitis C diagnosis

2016 ◽  
Vol 54 (2) ◽  
pp. 279-285 ◽  
Author(s):  
Lijuan Wang ◽  
Hong Lv ◽  
Guojun Zhang
Keyword(s):  
Hepatitis C ◽  
Positive Predictive Value ◽  
Predictive Value ◽  
Characteristic Curve ◽  
High Specificity ◽  
Receiver Operator Characteristic Curve ◽  
Core Antigen ◽  
Hcv Rna ◽  
Serum Samples ◽  
Alternative Method

Background The study aimed to evaluate a fully automated chemiluminescent immunoassay and compared it with a quantitative RNA assay and anti-HCV assay to verify the utility of this automated Ag assay as an alternative method for hepatitis C diagnosis. Methods A total of 229 serum samples previously tested for anti-HCV concentrations by the Architect Anti-HCV assay, were selected for HCV RNA testing by real time RT-PCR kit (Shanghai ZJ Bio-Tec Co., Ltd) and 125 specimens were tested for HCV Ag by the Architect HCV core Antigen kit. Results The log10 HCVAg and HCV RNA concentrations were highly correlated [ r = 0.834); with HCV RNA as the comparator test, HCVAg had 100% specificity, 100% positive predictive value (PPV) and 94.8% sensitivity. We found 1 pg/mL of total HCV core Ag is equivalent to approximately 6607HCV RNA international units (IU)/mL. Receiver operator characteristic curve analysis showed that the area under the curve of HCV core Ag (0.989) was greater than HCV Ab (0.871). HCV Ag concentrations and RNA-to-Ag ratio of the groups for HCV RNA concentrations ≤105 and >105 IU/mL were both significantly different from each other ( P < 0.05). Conclusion The Architect HCV core Ag assay may be an alternative method for hepatitis C diagnosis, performed on the same analytical platform and sample as the anti-HCV assay, shortening the diagnostic window period, demonstrating good correlation with HCV RNA assay with high specificity and positive predictive value.

scholarly journals Agreement in the diagnosis of dental fluorosis in central incisors performed by a standardized photographic method and clinical examination

2009 ◽  
Vol 25 (5) ◽  
pp. 1017-1024 ◽  
Author(s):  
Carolina Castro Martins ◽  
Loliza Chalub ◽  
Ynara Bosco Lima-Arsati ◽  
Isabela Almeida Pordeus ◽  
Saul Martins Paiva
Keyword(s):  
Positive Predictive Value ◽  
Negative Predictive Value ◽  
Clinical Examination ◽  
Predictive Value ◽  
Gold Standard ◽  
Dental Fluorosis ◽  
Digital Camera ◽  
High Specificity ◽  
Photographic Method ◽  
Pediatric Dentists

The aim of this study was to assess agreement in the diagnosis of dental fluorosis performed by a standardized digital photographic method and a clinical examination (gold standard). 49 children (aged 7-9 years) were clinically evaluated by a trained examiner for the assessment of dental fluorosis. Central incisors were evaluated for the presence or absence of dental fluorosis and were photographed with a digital camera. Photographs were presented to three pediatric dentists, who examined the images. Data were analyzed using Cohen's kappa and validity values. Agreement in the diagnosis performed by the photographic method and clinical examination was good (0.67) and accuracy was 83.7%. The prevalence of dental fluorosis was reported to be higher in the clinical examination (49%) compared with the photographic method (36.7%). The photographic method presented higher specificity (96%) than sensitivity (70.8%), a positive predictive value (PPV) of 94.4% and a negative predictive value (NPV) of 77.4%. The diagnosis of dental fluorosis performed using the photographic method presented high specificity and PPV, which indicates that the method is reproducible and reliable for recording dental fluorosis.

Sign in / Sign up

Export Citation Format

Share Document