Weight loss trajectories and related factors in a 16-week mobile obesity intervention: A retrospective observational study (Preprint)

BACKGROUND In obesity management, whether patients lose 5% or more of their initial weight is a critical factor in their clinical outcome. However, evaluations that only take this approach cannot identify and distinguish between individuals whose weight change varies and those who steadily lose weight. Evaluation of weight loss considering the volatility of weight change through a mobile-based intervention for obesity can facilitate the understanding of individuals’ behavior and weight changes from a longitudinal perspective. OBJECTIVE With machine learning approach, we examined weight loss trajectories and explored the factors related to behavioral and app usage characteristics that induce weight loss. METHODS We used the lifelog data of 19,784 individuals who enrolled in a 16-week obesity management program on the healthcare app Noom in the US during August 8, 2013 to August 8, 2019. We performed K-means clustering with dynamic time warping to cluster the weight loss time series and inspected the quality of clusters with the total sum of distance within the clusters. To identify the usage factors to determine clustering assignment, we longitudinally compared weekly usage statistics with effect size on a weekly basis. RESULTS Initial Body Mass Index (BMI) of participants was 33.9±5.9 kg/m2, and ultimately reached an average BMI of 32.0±5.7 kg/m2. In their weight log, 5 Clusters were identified: Cluster 1 (sharp decrease) showed a high proportion of weight reduction class between 10% and 15%—the highest among the five clusters (n=2,364/12,796, 18.9%)—followed by Cluster 2 (moderate decrease), Cluster 3 (increase), Cluster 4 (yoyo), Cluster 5 (other). In comparison between cluster 2 and cluster 4, although the effect size of difference in the average meal input adherence and average weight input adherence did not show a significant difference in the first week, it increased continuously for 7 weeks (Cohen’s d=0.408; 0.38). CONCLUSIONS With machine learning approach clustering shape-based timeseries similarity, this study identified 5 weight loss trajectories in mobile weight management app. Overall adherence and early adherence related to self-monitoring emerged as a potential predictor of these trajectories.

Download Full-text

Application of multi-omics data integration and machine learning approaches to identify epigenetic and transcriptomic differences between in vitro and in vivo produced bovine embryos

PLoS ONE ◽

10.1371/journal.pone.0252096 ◽

2021 ◽

Vol 16 (5) ◽

pp. e0252096

Author(s):

Maria B. Rabaglino ◽

Alan O’Doherty ◽

Jan Bojsen-Møller Secher ◽

Patrick Lonergan ◽

Poul Hyttel ◽

...

Keyword(s):

Machine Learning ◽

Data Integration ◽

Focal Adhesion ◽

Learning Approach ◽

Omics Data ◽

Machine Learning Approach ◽

Cluster 2 ◽

Omics Data Integration

Pregnancy rates for in vitro produced (IVP) embryos are usually lower than for embryos produced in vivo after ovarian superovulation (MOET). This is potentially due to alterations in their trophectoderm (TE), the outermost layer in physical contact with the maternal endometrium. The main objective was to apply a multi-omics data integration approach to identify both temporally differentially expressed and differentially methylated genes (DEG and DMG), between IVP and MOET embryos, that could impact TE function. To start, four and five published transcriptomic and epigenomic datasets, respectively, were processed for data integration. Second, DEG from day 7 to days 13 and 16 and DMG from day 7 to day 17 were determined in the TE from IVP vs. MOET embryos. Third, genes that were both DE and DM were subjected to hierarchical clustering and functional enrichment analysis. Finally, findings were validated through a machine learning approach with two additional datasets from day 15 embryos. There were 1535 DEG and 6360 DMG, with 490 overlapped genes, whose expression profiles at days 13 and 16 resulted in three main clusters. Cluster 1 (188) and Cluster 2 (191) genes were down-regulated at day 13 or day 16, respectively, while Cluster 3 genes (111) were up-regulated at both days, in IVP embryos compared to MOET embryos. The top enriched terms were the KEGG pathway "focal adhesion" in Cluster 1 (FDR = 0.003), and the cellular component: "extracellular exosome" in Cluster 2 (FDR<0.0001), also enriched in Cluster 1 (FDR = 0.04). According to the machine learning approach, genes in Cluster 1 showed a similar expression pattern between IVP and less developed (short) MOET conceptuses; and between MOET and DKK1-treated (advanced) IVP conceptuses. In conclusion, these results suggest that early conceptuses derived from IVP embryos exhibit epigenomic and transcriptomic changes that later affect its elongation and focal adhesion, impairing post-transfer survival.

Download Full-text

Machine-learning analysis identifies digital behavioral phenotypes for engagement and health outcome efficacy of mHealth interventions for obesity: post-hoc analyses of a randomized trial (Preprint)

10.2196/preprints.27218 ◽

2021 ◽

Author(s):

Meelim Kim ◽

Jaeyeong Yang ◽

Woo-Young Ahn ◽

Hyung Jin Choi

Keyword(s):

Machine Learning ◽

Cognitive Behavioral Therapy ◽

Health Outcomes ◽

Clinical Outcomes ◽

Weight Change ◽

Behavioral Therapy ◽

Cognitive Behavioral ◽

Learning Approach ◽

Machine Learning Approach ◽

Learning Analysis

BACKGROUND The digital healthcare community has been urged to enhance engagement and clinical outcomes by analyzing multidimensional digital phenotypes. OBJECTIVE This study aimed to investigate the performance of multivariate phenotypes predicting the engagement rate and health outcomes of digital cognitive behavioral therapy (dCBT) using a machine learning approach. METHODS We leveraged both conventional phenotypes assessed by validated psychological questionnaires and multidimensional digital phenotypes within time-series data from a mobile app of 45 participants undergoing digital cognitive behavioral therapy (dCBT) for eight weeks. To discriminate the important characteristics, we conducted a machine-learning analysis. RESULTS A higher engagement rate was associated with higher weight loss at 8 weeks (r = -0.59, p < 0001) and 24 weeks (r = -0.52, p = 0001). The machine learning approach revealed distinct multivariate profiles associated with varying impacts on the outcomes. Lower self-esteem on the conventional phenotype and higher in-app motivational measures on digital phenotypes commonly accounted for both engagement and health outcomes. In addition, eight types of digital phenotypes predicted engagement rates (mean R2 = 0416, SD = 0006). The prediction of short-term weight change (mean R2 = 0382, SD = 0015) was associated with six different digital phenotypes. Lastly, two behavioral measures of digital phenotypes were associated with a long-term weight change (mean R2 = 0590, SD = 0011). CONCLUSIONS Our findings successfully demonstrated how multiple psychological constructs, such as emotional, cognitive, behavioral, and motivational phenotypes, elucidate the mechanisms and clinical efficacy of digital intervention with the machine learning method. Our results also highlight the importance of assessing multiple aspects of motivation before and during the intervention to improve both engagement rate and clinical outcomes. This line of research may shed light on the development of advanced prevention and personalized digital therapeutics. CLINICALTRIAL ClinicalTrials.gov NCT03465306 (Retrieved September 18, 2017, https://register.clinicaltrials.gov/NCT03465306)

Download Full-text

Machine Learning Consensus Clustering of Hospitalized Patients with Admission Hyponatremia

Diseases ◽

10.3390/diseases9030054 ◽

2021 ◽

Vol 9 (3) ◽

pp. 54

Author(s):

Charat Thongprayoon ◽

Panupong Hansrivijit ◽

Michael A. Mao ◽

Pradeep K. Vaitla ◽

Andrea G. Kattah ◽

...

Keyword(s):

Machine Learning ◽

Kidney Function ◽

Serum Sodium ◽

Learning Approach ◽

Serum Bicarbonate ◽

Unsupervised Machine Learning ◽

Disease Cluster ◽

Machine Learning Approach ◽

One Year ◽

Cluster 2

Background: The objective of this study was to characterize patients with hyponatremia at hospital admission into clusters using an unsupervised machine learning approach, and to evaluate the short- and long-term mortality risk among these distinct clusters. Methods: We performed consensus cluster analysis based on demographic information, principal diagnoses, comorbidities, and laboratory data among 11,099 hospitalized adult hyponatremia patients with an admission serum sodium below 135 mEq/L. The standardized mean difference was utilized to identify each cluster’s key features. We assessed the association of each hyponatremia cluster with hospital and one-year mortality using logistic and Cox proportional hazard analysis, respectively. Results: There were three distinct clusters of hyponatremia patients: 2033 (18%) in cluster 1, 3064 (28%) in cluster 2, and 6002 (54%) in cluster 3. Among these three distinct clusters, clusters 3 patients were the youngest, had lowest comorbidity burden, and highest kidney function. Cluster 1 patients were more likely to be admitted for genitourinary disease, and have diabetes and end-stage kidney disease. Cluster 1 patients had the lowest kidney function, serum bicarbonate, and hemoglobin, but highest serum potassium and prevalence of acute kidney injury. In contrast, cluster 2 patients were the oldest and were more likely to be admitted for respiratory disease, have coronary artery disease, congestive heart failure, stroke, and chronic obstructive pulmonary disease. Cluster 2 patients had lowest serum sodium and serum chloride, but highest serum bicarbonate. Cluster 1 patients had the highest hospital mortality and one-year mortality, followed by cluster 2 and cluster 3, respectively. Conclusion: We identified three clinically distinct phenotypes with differing mortality risks in a heterogeneous cohort of hospitalized hyponatremic patients using an unsupervised machine learning approach.

Download Full-text