Prediction of preterm birth based on machine learning using bacterial risk score in cervicovaginal fluid

2021 ◽  
Author(s):  
Sunwha Park ◽  
Daejoong Oh ◽  
Hanna Heo ◽  
Gain Lee ◽  
Soo Min Kim ◽  
...  
Keyword(s):  
Machine Learning ◽  
Preterm Birth ◽  
Risk Score ◽  
Cervicovaginal Fluid

scholarly journals All-cause Dementia Prediction by Machine Learning: The Health, Aging, and Body Composition Study

2020 ◽  
Vol 4 (Supplement_1) ◽  
pp. 487-487
Author(s):  
Chenkai Wu ◽  
Xurui Jin
Keyword(s):  
Machine Learning ◽  
Body Composition ◽  
Risk Prediction ◽  
Risk Score ◽  
Body Composition Study ◽  
C Statistic ◽  
Dementia Risk ◽  
Health Aging ◽  
Composition Study ◽  
Traditional Approaches

Abstract There are several shortcomings of the currently available risk prediction models for dementia. We developed a risk prediction model for dementia using machine-learning approach and compared its performance with traditional approaches. Data were from the Health, Aging, and Body Composition Study, comprising 3,075 older adults (at least 70 years). Dementia was defined as (1) use of a prescribed dementia medication, (2) adjudicated dementia diagnosis, or (3) a race-stratified cognitive decline>1.5 SDs from the baseline mean. We selected 275 predictors collected from questionnaires, imaging data, performance testing, and biospecimen. We used random survival forest (RSF) to build the full model and rank the importance of predictors. Subsequently, we built parsimonious models with top-20 predictors using RSF and Cox regression. A dementia risk score was developed using top-ranked variables. We used the C-statistic for performance evaluation. Over a median of 11.4 years of follow-up, 659 dementias (21.4%) occurred. The RSF model (both including all and top-20 variables) showed a higher C-statistic than the regression model. Digit symbol score, physical performance battery, finger tapping score, weight change since age 50, serum adiponectin, and APOE genotype were the top-6 variables. We created a dementia risk score (0-10) using the top-6 variables. A 1-unit increase in the risk score was associated with an 8% higher risk of dementia. The risk score demonstrated good discrimination (C-statistic=0.75). Machine learning methods offered improvement over traditional approaches in predicting dementia. The risk prediction score derived from a parsimonious model had good prediction performance.

Effect of a Real-Time Risk Score on 30-day Readmission Reduction in Singapore

2021 ◽  
Vol 12 (02) ◽  
pp. 372-382
Author(s):  
Christine Xia Wu ◽  
Ernest Suresh ◽  
Francis Wei Loong Phng ◽  
Kai Pik Tai ◽  
Janthorn Pakdeethai ◽  
...  
Keyword(s):  
Machine Learning ◽  
High Risk ◽  
Real Time ◽  
Risk Score ◽  
Patient Specific ◽  
Learning Models ◽  
Medicine Department ◽  
High Risk Patients ◽  
Risk Patients ◽  
Machine Learning Models

Abstract Objective To develop a risk score for the real-time prediction of readmissions for patients using patient specific information captured in electronic medical records (EMR) in Singapore to enable the prospective identification of high-risk patients for enrolment in timely interventions. Methods Machine-learning models were built to estimate the probability of a patient being readmitted within 30 days of discharge. EMR of 25,472 patients discharged from the medicine department at Ng Teng Fong General Hospital between January 2016 and December 2016 were extracted retrospectively for training and internal validation of the models. We developed and implemented a real-time 30-day readmission risk score generation in the EMR system, which enabled the flagging of high-risk patients to care providers in the hospital. Based on the daily high-risk patient list, the various interfaces and flow sheets in the EMR were configured according to the information needs of the various stakeholders such as the inpatient medical, nursing, case management, emergency department, and postdischarge care teams. Results Overall, the machine-learning models achieved good performance with area under the receiver operating characteristic ranging from 0.77 to 0.81. The models were used to proactively identify and attend to patients who are at risk of readmission before an actual readmission occurs. This approach successfully reduced the 30-day readmission rate for patients admitted to the medicine department from 11.7% in 2017 to 10.1% in 2019 (p < 0.01) after risk adjustment. Conclusion Machine-learning models can be deployed in the EMR system to provide real-time forecasts for a more comprehensive outlook in the aspects of decision-making and care provision.

EHR-based machine learning modeling for preterm birth prediction: A systematic review (Preprint)

2021 ◽  
Author(s):  
Zahra Sharifiheris ◽  
Juho Laitala ◽  
Antti Airola ◽  
Amir M Rahmani ◽  
Miriam Bender
Keyword(s):  
Machine Learning ◽  
Systematic Review ◽  
Preterm Birth ◽  
Model Performance ◽  
Bibliographic Databases ◽  
Complementary Approach ◽  
Prisma Statement ◽  
Comprehensive Search ◽  
First Time

BACKGROUND Preterm birth (PTB) as a common pregnancy complication is responsible for 35% of the 3.1 million pregnancy-related deaths each year and significantly impacts around 15 million children annually across the world. Conventional approaches to predict PTB may neither be applicable for first-time mothers nor possess reliable predictive power. Recently, machine learning (ML) models have shown the potential as an appropriate complementary approach for PTB prediction. OBJECTIVE In this article we systematically reviewed the literature concerned with PTB prediction using ML modeling. METHODS This systematic review was conducted in accordance with the PRISMA statement. A comprehensive search was performed in seven bibliographic databases up until 15 May 2021. The quality of studies was assessed, and the descriptive information including socio-demographic characteristics, ML modeling processes, and model performance were extracted and reported. RESULTS A total of 732 papers were screened through title and abstract. Of these, 23 studies were screened by full text resulting in 13 papers that met the inclusion criteria. CONCLUSIONS We identified various ML models used for different EHR data resulting in a desirable performance for PTB prediction. However, evaluation metrics, software/package used, data size and type, and selected features, and importantly data management method often varied from study to study threatening the reliability and generalizability of the model. CLINICALTRIAL n.a.

259: Evaluation of a cumulative first trimester characteristic and serum marker risk score for predicting early spontaneous preterm birth

2015 ◽  
Vol 212 (1) ◽  
pp. S142
Author(s):  
Laura L. Jelliffe-Pawlowski ◽  
Rebecca Baer ◽  
Yair Blumenfeld ◽  
Christina Chambers ◽  
Maurice Druzin ◽  
...  
Keyword(s):  
Preterm Birth ◽  
Risk Score ◽  
First Trimester ◽  
Serum Marker ◽  
Spontaneous Preterm Birth

scholarly journals Prediction of Spontaneous Preterm Birth in At-risk Women Using Thrombospondin 1 from Cervicovaginal Fluid: A Prospective Observational Study

2021 ◽  
Vol 81 (09) ◽  
pp. 1055-1064
Author(s):  
Johannes Stubert ◽  
Kathleen Gründler ◽  
Bernd Gerber ◽  
Dagmar-Ulrike Richter ◽  
Max Dieterich
Keyword(s):  
At Risk ◽  
Preterm Birth ◽  
Predictive Value ◽  
Transvaginal Ultrasound ◽  
Area Under The Curve ◽  
Cervical Length ◽  
Multiple Gestation ◽  
Spontaneous Preterm Birth ◽  
Thrombospondin 1 ◽  
Cervicovaginal Fluid

Abstract Introduction Thrombospondin 1, desmoplakin and stratifin are putative biomarkers for the prediction of preterm birth. This study aimed to validate the predictive capability of these biomarkers in patients at risk of preterm birth. Materials and Methods We included 109 women with symptoms of threatened spontaneous preterm birth between weeks 20 0/7 and 31 6/7 of gestation. Inclusion criteria were uterine contractions, cervical length of less than 25 mm, or a personal history of spontaneous preterm birth. Multiple gestations were also included. Samples of cervicovaginal fluid were taken before performing a digital examination and transvaginal ultrasound. Levels of cervicovaginal thrombospondin 1, desmoplakin and stratifin were quantified by enzyme-linked immunosorbent assays. The primary endpoint was spontaneous preterm birth before 34 + 0 weeks of gestation. Results Sixteen women (14.7%) delivered before 34 + 0 weeks. Median levels of thrombospondin 1 were higher in samples where birth occurred before 34 weeks vs. ≥ 34 weeks of gestation (4904 vs. 469 pg/mL, p < 0.001). Receiver operator characteristics analysis resulted in an area under the curve of 0.86 (p < 0.0001). At an optimal cut-off value of 2163 pg/mL, sensitivity, specificity, positive predictive value and negative predictive value were 0.94, 0.77, 0.42 and 0.99, respectively, with an adjusted odds ratio of 32.9 (95% CI: 3.1 – 345, p = 0.004). Multiple gestation, cervical length, and preterm labor had no impact on the results. Survival analysis revealed a predictive period of more than eight weeks. Levels of desmoplakin and stratifin did not differ between groups. Conclusion Thrombospondin 1 allowed long-term risk estimation of spontaneous preterm birth.

scholarly journals Smart Policing Technique With Crime Type and Risk Score Prediction Based on Machine Learning for Early Awareness of Risk Situation

IEEE Access ◽  
2021 ◽  
Vol 9 ◽  
pp. 131906-131915
Author(s):  
Myung-Sun Baek ◽  
Wonjoo Park ◽  
Jaehong Park ◽  
Kwang-Ho Jang ◽  
Yong-Tae Lee
Keyword(s):  
Machine Learning ◽  
Risk Score ◽  
Crime Type ◽  
Risk Situation

scholarly journals Machine Learning Algorithm Using Clinical Data and Demographic Data for Preterm Birth Prediction

2022 ◽  
Vol 226 (1) ◽  
pp. S362-S363
Author(s):  
Matthew Hoffman ◽  
Wei Liu ◽  
Jade Tunguhan ◽  
Ghamar Bitar ◽  
Kaveeta Kumar ◽  
...  
Keyword(s):  
Machine Learning ◽  
Preterm Birth ◽  
Clinical Data ◽  
Learning Algorithm ◽  
Demographic Data ◽  
Machine Learning Algorithm

Abstract 153: Novel Machine Learning Proves Stroke Risk is Not Linear

Stroke ◽  
2020 ◽  
Vol 51 (Suppl_1) ◽  
Author(s):  
Agni Orfanoudaki ◽  
Amre M Nouh ◽  
Emma Chesley ◽  
Christian Cadisch ◽  
Barry Stein ◽  
...  
Keyword(s):  
Machine Learning ◽  
Risk Factors ◽  
Risk Score ◽  
Stroke Risk ◽  
Assessment Tools ◽  
Risk Calculator ◽  
Stroke Incidence ◽  
Non Linear ◽  
User Friendly ◽  
The Impact

Background: Current stroke risk assessment tools presume the impact of risk factors is linear and cumulative. However, both novel risk factors and their interplay influencing stroke incidence are difficult to reveal using traditional linear models. Objective: To improve upon the Revised-Framingham Stroke Risk Score and design an interactive non-linear Stroke Risk Score (NSRS). Our work aimed at increasing the accuracy of event prediction and uncovering new relationships in an interpretable user-friendly fashion. Methods: A two phase approach was used to develop our stroke risk score predictor. First, clinical examinations of the Framingham offspring cohort were utilized as the training dataset for the predictive model consisting of 14,196 samples where each clinical examination was considered an independent observation. Optimal Classification Trees (OCT) were used to train a model to predict 10-year stroke risk. Second, this model was validated with 17,527 observations from the Boston Medical Center. The NSRS was developed into an online user friendly application in the form of a questionnaire (http://www.mit.edu/~agniorf/files/questionnaire_Cohort2.html). Results: The algorithm suggests a key dichotomy between patients with or without history of cardiovascular disease. While the model agrees with known findings, it also identified 23 unique stroke risk profiles and introduced new non-linear relationships; such as the role of T-wave abnormality on electrocardiography and hematocrit levels in a patient’s risk profile. Our results in both the training and validation populations suggested that the non-linear approach significantly improves upon the existing revised Framingham stroke risk calculator in the c-statistic (training 87.43% (CI 0.85-0.90) vs. 73.74% (CI 0.70-0.76); validation 75.29% (CI 0.74-0.76) vs 65.93% (CI 0.64-0.67), even in multi-ethnicity populations. Conclusions: We constructed a highly predictive, interpretable and user-friendly stroke risk calculator using novel machine-learning uncovering new risk factors, interactions and unique profiles. The clinical implications include prioritization of risk factor modification and personalized care improving targeted intervention for stroke prevention.

Sign in / Sign up

Export Citation Format

Share Document