scholarly journals Physically interpretable machine learning algorithm on multidimensional non-linear fields

2021 ◽  
Vol 428 ◽  
pp. 110074
Author(s):  
Rem-Sophia Mouradi ◽  
Cédric Goeury ◽  
Olivier Thual ◽  
Fabrice Zaoui ◽  
Pablo Tassi
Keyword(s):  
Machine Learning ◽  
Learning Algorithm ◽  
Machine Learning Algorithm ◽  
Interpretable Machine Learning ◽  
Non Linear

scholarly journals Predicting Venous Thrombosis in Osteoarthritis Using a Machine Learning Algorithm: A Population-Based Cohort Study

2022 ◽  
Vol 12 (1) ◽  
pp. 114
Author(s):  
Chao Lu ◽  
Jiayin Song ◽  
Hui Li ◽  
Wenxing Yu ◽  
Yangquan Hao ◽  
...  
Keyword(s):  
Machine Learning ◽  
Cohort Study ◽  
High Risk ◽  
Venous Thrombosis ◽  
Learning Algorithm ◽  
Population Based ◽  
Joint Disease ◽  
Recursive Feature Elimination ◽  
Machine Learning Algorithm ◽  
Interpretable Machine Learning

Osteoarthritis (OA) is the most common joint disease associated with pain and disability. OA patients are at a high risk for venous thrombosis (VTE). Here, we developed an interpretable machine learning (ML)-based model to predict VTE risk in patients with OA. To establish a prediction model, we used six ML algorithms, of which 35 variables were employed. Recursive feature elimination (RFE) was used to screen the most related clinical variables associated with VTE. SHapley additive exPlanations (SHAP) were applied to interpret the ML mode and determine the importance of the selected features. Overall, 3169 patients with OA (average age: 66.52 ± 7.28 years) were recruited from Xi’an Honghui Hospital. Of these, 352 and 2817 patients were diagnosed with and without VTE, respectively. The XGBoost algorithm showed the best performance. According to the RFE algorithms, 15 variables were retained for further modeling with the XGBoost algorithm. The top three predictors were Kellgren–Lawrence grade, age, and hypertension. Our study showed that the XGBoost model with 15 variables has a high potential to predict VTE risk in patients with OA.

Uncovering the Most Important Factors for Predicting Sexual Desire using Interpretable Machine Learning

2020 ◽  
Author(s):  
Laura Marika Vowels ◽  
Matthew J Vowels ◽  
Kristen P Mark
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Sexual Desire ◽  
Learning Algorithm ◽  
Predictor Variable ◽  
Machine Learning Algorithm ◽  
Societal Factors ◽  
Interpretable Machine Learning ◽  
Several Variables ◽  
Two Samples

Previous studies have found a number of individual, relational, and societal factors that are associated with sexual desire. However, no studies to date have examined which of these variables are the most predictive of sexual desire. We used a machine learning algorithm, random forest (a type of interpretable highly non-linear decision tree), to predict sexual desire from a large number of predictors across two samples (N = 1846; includes 754 individuals forming 377 couples). We also used a Shapley value technique to estimate the size and direction of the effect of each predictor variable on the model outcome. The model predicted around 40% of variance in dyadic and solitary desire with women’s desire being more predictable than men’s. Several variables consistently predicted sexual desire including individual, relational, and societal factors. The study provides the strongest evidence to date of the most important predictors for dyadic and solitary desire.

scholarly journals ESG Investments: Filtering versus Machine Learning Approaches

2021 ◽  
Vol 8 (2) ◽  
pp. 1
Author(s):  
Vincent Margot ◽  
Christophe Geissler ◽  
Carmine De Franco ◽  
Bruno Monnier
Keyword(s):  
Machine Learning ◽  
Learning Algorithm ◽  
Dimensional Space ◽  
Machine Learning Algorithm ◽  
Learning Approaches ◽  
Non Linear ◽  
New Ideas ◽  
Stock Picking ◽  
A Company ◽  
Large Investment

We designed a machine learning algorithm that identifies patterns between ESG profiles and financial performances for companies in a large investment universe. The algorithm consists of regularly updated sets of rules that map regions into the high-dimensional space of ESG features to excess return predictions. The final aggregated predictions are transformed into scores which allow us to design simple strategies that screen the investment universe for stocks with positive scores. By linking the ESG features with financial performances in a non-linear way, our strategy based upon our machine learning algorithm turns out to be an efficient stock picking tool, which outperforms classic strategies that screen stocks according to their ESG ratings, as the popular best-in-class approach. Our paper brings new ideas in the growing field of financial literature that investigates the links between ESG behavior and the economy. We show indeed that there is clearly some form of alpha in the ESG profile of a company, but that this alpha can be accessed only with powerful, non-linear techniques such as machine learning.

scholarly journals eARDS: A multi-center validation of an interpretable machine learning algorithm of early onset Acute Respiratory Distress Syndrome (ARDS) among critically ill adults with COVID-19

PLoS ONE ◽  
2021 ◽  
Vol 16 (9) ◽  
pp. e0257056
Author(s):  
Lakshya Singhal ◽  
Yash Garg ◽  
Philip Yang ◽  
Azade Tabaie ◽  
A. Ian Wong ◽  
...  
Keyword(s):  
Machine Learning ◽  
Health Systems ◽  
Learning Algorithm ◽  
Model Performance ◽  
Health Science ◽  
Machine Learning Algorithm ◽  
Age Group ◽  
Science Center ◽  
Clinical Criteria ◽  
Interpretable Machine Learning

We present an interpretable machine learning algorithm called ‘eARDS’ for predicting ARDS in an ICU population comprising COVID-19 patients, up to 12-hours before satisfying the Berlin clinical criteria. The analysis was conducted on data collected from the Intensive care units (ICU) at Emory Healthcare, Atlanta, GA and University of Tennessee Health Science Center, Memphis, TN and the Cerner® Health Facts Deidentified Database, a multi-site COVID-19 EMR database. The participants in the analysis consisted of adults over 18 years of age. Clinical data from 35,804 patients who developed ARDS and controls were used to generate predictive models that identify risk for ARDS onset up to 12-hours before satisfying the Berlin criteria. We identified salient features from the electronic medical record that predicted respiratory failure among this population. The machine learning algorithm which provided the best performance exhibited AUROC of 0.89 (95% CI = 0.88–0.90), sensitivity of 0.77 (95% CI = 0.75–0.78), specificity 0.85 (95% CI = 085–0.86). Validation performance across two separate health systems (comprising 899 COVID-19 patients) exhibited AUROC of 0.82 (0.81–0.83) and 0.89 (0.87, 0.90). Important features for prediction of ARDS included minimum oxygen saturation (SpO2), standard deviation of the systolic blood pressure (SBP), O2 flow, and maximum respiratory rate over an observational window of 16-hours. Analyzing the performance of the model across various cohorts indicates that the model performed best among a younger age group (18–40) (AUROC = 0.93 [0.92–0.94]), compared to an older age group (80+) (AUROC = 0.81 [0.81–0.82]). The model performance was comparable on both male and female groups, but performed significantly better on the severe ARDS group compared to the mild and moderate groups. The eARDS system demonstrated robust performance for predicting COVID19 patients who developed ARDS at least 12-hours before the Berlin clinical criteria, across two independent health systems.

Machine Learning Algorithm to Predict Early Complications after Brain Tumor Surgery

2018 ◽  
Author(s):  
C.H.B. van Niftrik ◽  
F. van der Wouden ◽  
V. Staartjes ◽  
J. Fierstra ◽  
M. Stienen ◽  
...  
Keyword(s):  
Machine Learning ◽  
Brain Tumor ◽  
Learning Algorithm ◽  
Machine Learning Algorithm ◽  
Tumor Surgery ◽  
Early Complications ◽  
Brain Tumor Surgery
Sign in / Sign up

Export Citation Format

Share Document