Are all quantitative postmarketing signal detection methods equal? Performance characteristics of logistic regression and Multi-item Gamma Poisson Shrinker

2011 ◽  
Vol 21 (6) ◽  
pp. 622-630 ◽  
Author(s):  
Conny Berlin ◽  
Carles Blanch ◽  
David J. Lewis ◽  
Dionigi D. Maladorno ◽  
Christiane Michel ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Signal Detection ◽  
Performance Characteristics ◽  
Detection Methods ◽  
Gamma Poisson Shrinker

Do logistic regression and signal detection identify different subgroups at risk? Implications for the design of tailored interventions.

2001 ◽  
Vol 6 (1) ◽  
pp. 35-48 ◽  
Author(s):  
Michaela Kiernan ◽  
Helena C. Kraemer ◽  
Marilyn A. Winkleby ◽  
Abby C. King ◽  
C. Barr Taylor
Keyword(s):  
At Risk ◽  
Logistic Regression ◽  
Signal Detection ◽  
Tailored Interventions

Feature Selection for Kick Detection With Machine Learning Using Laboratory Data

2019 ◽  
Author(s):  
Suranga C. H. Geekiyanage ◽  
Adrian Ambrus ◽  
Dan Sui
Keyword(s):  
Machine Learning ◽  
Logistic Regression ◽  
Feature Selection ◽  
Data Analysis ◽  
Laboratory Data ◽  
Surface Flow ◽  
Full Scale ◽  
Rate Of Change ◽  
Laboratory Scale ◽  
Detection Methods

Abstract Conventional kick detection methods mainly include monitoring pit gains, surface flow data (flow in and flow out), surface and down-hole pressure variations, and outputs from physics-based models. Kick detection times depend on a driller’s individual ability to interpret these drilling measurements, symptoms and model predictions. Furthermore, testing a novel data-driven solution in a full-scale operation may induce non-productive time, safety risks and crew fatigue adding to false alarms that inevitably occur during testing. Therefore, the development of better, faster and less human intervention-dependent kick detection on a laboratory scale system is a valuable step before full-scale testing. We have generated a dataset containing seven typical drilling measurements and a sequence of gas kicks from experiments conducted in the laboratory scale. First, we employ data analysis tools following data pre-processing steps, data scaling, outlier detection, and natural feature selection. Next, we consider additional “engineered features” and apply different feature combinations to logistic regression with an ensemble method (boosting) for developing kick detection algorithms. In our data analysis, ‘Delta flow’ (difference between flow in and flow out of the well) and ‘Rate of change of delta flow’ designed features, combined with logistic regression and boosting, give promising results in detecting kicks. Finally, we propose an intelligent algorithm and alarm architecture for a complete kick alarm system, which draws from both data analysis and machine learning models developed in this work.

Circulating tumor cell count from a blood sample for colorectal cancer (CRC) prevention: A 627-patient prospective study.

2019 ◽  
Vol 37 (4_suppl) ◽  
pp. 485-485 ◽  
Author(s):  
Wen-Sy Tsai ◽  
Drew Watson ◽  
Ying Chang ◽  
Ben Hsieh ◽  
Hung-Jen Shao ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Logistic Regression ◽  
Prospective Study ◽  
Compliance Rate ◽  
Screening Method ◽  
Unmet Need ◽  
Detection Methods ◽  
P Value ◽  
Non Invasive ◽  
Adenoma Detection

485 Background: Up to 25% of patients with adenomas progress to having colorectal cancer. If detected early, adenomas can be removed with a diagnostic colonoscopy procedure, preventing cancer. Invasive colonoscopy is the only screening method with the sensitivity to accurately detect adenomas, but has a low compliance rate of 38% for screening. Available non-invasive tests (including stool-based multi-analyte tests) have very limited sensitivity for adenomas. Hence, there is an unmet need for a non-invasive test for adenoma detection. Methods: IRB-approved prospective study was conducted in 627 subjects 50 years or older- recommended for routine CRC screening- 405 subjects had adenoma or CRC, confirmed by colonoscopy with tumor biopsy. Two mL peripheral blood was processed using the CellMax biomimetic platform (CMx), which uses a microfluidic biochip to enumerate circulating tumor cells (CTCs). Nominal logistic regression was used to assess performance while proportional odds logistic regression and Cuzick’s trend test were used to determine association of CTC counts with cancer stage. Results: An increase in CTC count was significantly correlated with an increase in disease burden (Cuzick’s Test p-value < 0.0001). Furthermore, there was a significant association between CTC counts and stages of adenoma-carcinoma progression (Likelihood ratio p-value < 0.0001). The CTC enumeration was able to differentiate between healthy and diseased patients (adenoma + cancer). Conclusions: To the best of our knowledge, these are the first reported results for a blood test that has high accuracy for adenoma detection, and truly enables colorectal cancer prevention. This test can be administered in the primary care setting and drive high compliance.[Table: see text]

Reference set for performance testing of pediatric vaccine safety signal detection methods and systems

Vaccine ◽  
2016 ◽  
Vol 34 (51) ◽  
pp. 6626-6633 ◽  
Author(s):  
Yolanda Brauchli Pernus ◽  
Cassandra Nan ◽  
Thomas Verstraeten ◽  
Mariia Pedenko ◽  
Osemeke U. Osokogu ◽  
...  
Keyword(s):  
Signal Detection ◽  
Performance Testing ◽  
Vaccine Safety ◽  
Detection Methods ◽  
Safety Signal ◽  
Reference Set
Sign in / Sign up

Export Citation Format

Share Document