Behavioral Analytics for Myopic Agents.

Many multi-agent systems have a single coordinator providing incentives to a large number of agents. Two challenges faced by the coordinator are a finite budget from which to allocate incentives, and an initial lack of knowledge about the utility function of the agents. Here, we present a behavioral analytics approach for solving the coordinator's problem when the agents make decisions by maximizing utility functions that depend on prior system states, inputs, and other parameters that are initially unknown. Our behavioral analytics framework involves three steps: first, we develop a model that describes the decision-making process of an agent; second, we use data to estimate the model parameters for each agent and predict their future decisions; and third, we use these predictions to optimize a set of incentives that will be provided to each agent. The framework and approaches we propose in this paper can then adapt incentives as new information is collected. Furthermore, we prove that the incentives computed by this approach are asymptotically optimal with respect to a loss function that describes the coordinator's objective. We optimize incentives with a decomposition scheme, where each sub-problem solves the coordinator's problem for a single agent, and the master problem is a pure integer program. We conclude with a simulation study to evaluate the effectiveness of our approach for designing a personalized weight loss program. The results show that our approach maintains efficacy of the program while reducing its costs by up to 60%, while adaptive heuristics provide substantially less savings.

Association of Funding and Meal Preparation Time With Nutritional Quality of Meals of Supplemental Nutritional Assistance Program Recipients.

Importance: The Supplemental Nutrition Assistance Program (SNAP) is a federal program that provides food-purchasing assistance to low-income people; however, its current design does not account for the time availability of SNAP recipients to prepare meals. Objective: To evaluate the association of the availability of funding for food purchases and time for meal preparation with the nutritional quality of meals of SNAP recipients. Design, Setting, and Participants: This study used decision analytical modeling to evaluate the nutritional quality of meals of SNAP recipients. The model was developed from February 6, 2017, to December 12, 2020, using data from 2017 and is based on discrete optimization. The model describes food and grocery purchasing, in-home meal preparation, and meal plan choices of a family of SNAP participants (2 adults and 2 children) while considering food preferences, meal preparation time, and food costs. The model assumes food preferences match the foods typically purchased by SNAP households. Costs of food ingredients and prepared foods are taken from a single zip code. Exposures: Time availability and total amount and type of funding were varied. Allowing prepared delicatessen foods and disallowing frozen prepared foods for purchase using SNAP funds were considered. Main Outcomes and Measures: The primary outcome was the number of home-cooked meals and the amounts of fruits, vegetables, protein, sodium, sugar, and fiber consumed from generated meal plans. Amounts were evaluated as a percentage of the quantity recommended by established dietary guidelines. Results: Increased time availability was associated with increases in the percentage of home-cooked meals and servings of fruits/vegetables and decreased sodium consumption. Higher levels of funding were associated with increased consumption of fiber, fruits/vegetables, protein, sodium, and sugar. With 20 min/d of cooking time, $400/mo of SNAP benefits, and $100/mo of self-funding, the meal plan had a mean (SE) of 20.1% (0.3%) of meals home cooked, 0.5 (<0.1) servings/d per person of fruits/vegetables, 100.3% (0.6%) of daily recommended protein per person, 115.1% (0.8%) of daily recommended sodium per person, 241.8% (1.0%) of daily recommended sugar per person, and 31.2% (0.3%) of daily recommended fiber per person. With 20 min/d of cooking time, $400/mo of SNAP benefits, and $600/mo of self-funding, the meal plan had a mean (SE) of 23.9% (1.0%) of meals home cooked, 2.8 (0.1) servings/d per person of fruits/vegetables, 134.9% (1.6%) of daily recommended protein per person, 200.9% (3.1%) of daily recommended sodium per person, 295.1% (3.1%) of daily recommended sugar per person, and 90.1% (1.0%) of daily recommended fiber per person. With 60 min/d of cooking time, $400/mo of SNAP benefits, and $100/mo of self-funding, the meal plan had a mean (SE) of 52.7% (0.9%) of meals home cooked, 1.4 (<0.1) servings/d per person of fruits/vegetables, 109.0% (1.1%) of daily recommended protein per person, 108.7% (1.0%) of daily recommended sodium per person, 298.6% (2.0%) of daily recommended sugar per person, and 38.8% (0.4%) of daily recommended fiber per person. With 60 min/d of cooking time, $400/mo of SNAP benefits, and $600/mo of self-funding, the meal plan had a mean (SE) of 42.8% (1.2%) meals home cooked, 4.3 (0.1) servings/d per person of fruits/vegetables, 144.4% (1.8%) of daily recommended protein per person, 165.2% (2.8%) of daily recommended sodium per person, 322.4% (2.4%) of daily recommended sugar per person, and 91.0% (0.9%) of daily recommended fiber per person. Conclusions and Relevance: In this decision analytical model, meal preparation time was associated with the ability of SNAP recipient families to consume nutritious meals, suggesting that increased funding alone may be insufficient for improving the nutritional profiles of SNAP recipients. Given the current US food supply, governmental interventions that provide the equivalence in increased time availability to achieve nutritious meals may be needed.

Predicting NICU admissions in near-term and term infants with low illness acuity.

OBJECTIVE: Describe NICU admission rate variation among hospitals in infants with birthweight ≥2500 g and low illness acuity, and describe factors that predict NICU admission. STUDY DESIGN: Retrospective study from the Vizient Clinical Data Base/Resource Manager®. Support vector machine methodology was used to develop statistical models using (1) patient characteristics (2) only the indicator for the inborn hospital and (3) patient characteristics plus indicator for the inborn hospital. RESULTS: NICU admission rates of 427,449 infants from 154 hospitals ranged from 0 to 28.6%. C-statistics for the patient characteristics model: 0.64 (Confidence Interval (CI) 0.62-0.65), hospital only model: 0.81 (CI, 0.81-0.82), and patient characteristic plus hospital variable model: 0.84 (CI, 0.83-0.84). CONCLUSION/RELEVANCE: There is wide variation in NICU admission rates in infants with low acuity diagnoses. In all cohorts, birth hospital better predicted NICU admission than patient characteristics alone.

Correction: Modeling differentiation-state transitions linked to therapeutic escape in triple-negative breast cancer.

[This corrects the article DOI: 10.1371/journal.pcbi.1006840.].

Applying machine learning to predict future adherence to physical activity programs.

BACKGROUND: Identifying individuals who are unlikely to adhere to a physical exercise regime has potential to improve physical activity interventions. The aim of this paper is to develop and test adherence prediction models using objectively measured physical activity data in the Mobile Phone-Based Physical Activity Education program (mPED) trial. To the best of our knowledge, this is the first to apply Machine Learning methods to predict exercise relapse using accelerometer-recorded physical activity data. METHODS: We use logistic regression and support vector machine methods to design two versions of a Discontinuation Prediction Score (DiPS), which uses objectively measured past data (e.g., steps and goal achievement) to provide a numerical quantity indicating the likelihood of exercise relapse in the upcoming week. The respective prediction accuracy of these two versions of DiPS are compared, and then numerical simulation is performed to explore the potential of using DiPS to selectively allocate financial incentives to participants to encourage them to increase physical activity. RESULTS: we had access to a physical activity trial data that were continuously collected every 60 sec every day for 9 months in 210 participants. By using the first 15 weeks of data as training and test on weeks 16-30, we show that both versions of DiPS have a test AUC of 0.9 with high sensitivity and specificity in predicting the probability of exercise adherence. Simulation results assuming different intervention regimes suggest the potential benefit of using DiPS as a score to allocate resources in physical activity intervention programs in reducing costs over other allocation schemes. CONCLUSIONS: DiPS is capable of making accurate and robust predictions for future weeks. The most predictive features are steps and physical activity intensity. Furthermore, the use of DiPS scores can be a promising approach to determine when or if to provide just-in-time messages and step goal adjustments to improve compliance. Further studies on the use of DiPS in the design of physical activity promotion programs are warranted. TRIAL REGISTRATION: ClinicalTrials.gov NCT01280812 Registered on January 21, 2011.

Modeling differentiation-state transitions linked to therapeutic escape in triple-negative breast cancer.

Drug resistance in breast cancer cell populations has been shown to arise through phenotypic transition of cancer cells to a drug-tolerant state, for example through epithelial-to-mesenchymal transition or transition to a cancer stem cell state. However, many breast tumors are a heterogeneous mixture of cell types with numerous epigenetic states in addition to stem-like and mesenchymal phenotypes, and the dynamic behavior of this heterogeneous mixture in response to drug treatment is not well-understood. Recently, we showed that plasticity between differentiation states, as identified with intracellular markers such as cytokeratins, is linked to resistance to specific targeted therapeutics. Understanding the dynamics of differentiation-state transitions in this context could facilitate the development of more effective treatments for cancers that exhibit phenotypic heterogeneity and plasticity. In this work, we develop computational models of a drug-treated, phenotypically heterogeneous triple-negative breast cancer (TNBC) cell line to elucidate the feasibility of differentiation-state transition as a mechanism for therapeutic escape in this tumor subtype. Specifically, we use modeling to predict the changes in differentiation-state transitions that underlie specific therapy-induced changes in differentiation-state marker expression that we recently observed in the HCC1143 cell line. We report several statistically significant therapy-induced changes in transition rates between basal, luminal, mesenchymal, and non-basal/non-luminal/non-mesenchymal differentiation states in HCC1143 cell populations. Moreover, we validate model predictions on cell division and cell death empirically, and we test our models on an independent data set. Overall, we demonstrate that changes in differentiation-state transition rates induced by targeted therapy can provoke distinct differentiation-state aggregations of drug-resistant cells, which may be fundamental to the design of improved therapeutic regimens for cancers with phenotypic heterogeneity.

Behavioral Modeling in Weight Loss Interventions.

Designing systems with human agents is difficult because it often requires models that characterize agents' responses to changes in the system's states and inputs. An example of this scenario occurs when designing treatments for obesity. While weight loss interventions through increasing physical activity and modifying diet have found success in reducing individuals' weight, such programs are difficult to maintain over long periods of time due to lack of patient adherence. A promising approach to increase adherence is through the personalization of treatments to each patient. In this paper, we make a contribution towards treatment personalization by developing a framework for predictive modeling using utility functions that depend upon both time-varying system states and motivational states evolving according to some modeled process corresponding to qualitative social science models of behavior change. Computing the predictive model requires solving a bilevel program, which we reformulate as a mixed-integer linear program (MILP). This reformulation provides the first (to our knowledge) formulation for Bayesian inference that uses empirical histograms as prior distributions. We study the predictive ability of our framework using a data set from a weight loss intervention, and our predictive model is validated by comparison to standard machine learning approaches. We conclude by describing how our predictive model could be used for optimization, unlike standard machine learning approaches which cannot.

Differentiation-state plasticity is a targetable resistance mechanism in basal-like breast cancer.

Intratumoral heterogeneity in cancers arises from genomic instability and epigenomic plasticity and is associated with resistance to cytotoxic and targeted therapies. We show here that cell-state heterogeneity, defined by differentiation-state marker expression, is high in triple-negative and basal-like breast cancer subtypes, and that drug tolerant persister (DTP) cell populations with altered marker expression emerge during treatment with a wide range of pathway-targeted therapeutic compounds. We show that MEK and PI3K/mTOR inhibitor-driven DTP states arise through distinct cell-state transitions rather than by Darwinian selection of preexisting subpopulations, and that these transitions involve dynamic remodeling of open chromatin architecture. Increased activity of many chromatin modifier enzymes, including BRD4, is observed in DTP cells. Co-treatment with the PI3K/mTOR inhibitor BEZ235 and the BET inhibitor JQ1 prevents changes to the open chromatin architecture, inhibits the acquisition of a DTP state, and results in robust cell death in vitro and xenograft regression in vivo.

Applying Natural Language Processing to Understand Motivational Profiles for Maintaining Physical Activity After a Mobile App and Accelerometer-Based Intervention: The mPED Randomized Controlled Trial.

BACKGROUND: Regular physical activity is associated with reduced risk of chronic illnesses. Despite various types of successful physical activity interventions, maintenance of activity over the long term is extremely challenging. OBJECTIVE: The aims of this original paper are to 1) describe physical activity engagement post intervention, 2) identify motivational profiles using natural language processing (NLP) and clustering techniques in a sample of women who completed the physical activity intervention, and 3) compare sociodemographic and clinical data among these identified cluster groups. METHODS: In this cross-sectional analysis of 203 women completing a 12-month study exit (telephone) interview in the mobile phone-based physical activity education study were examined. The mobile phone-based physical activity education study was a randomized, controlled trial to test the efficacy of the app and accelerometer intervention and its sustainability over a 9-month period. All subjects returned the accelerometer and stopped accessing the app at the last 9-month research office visit. Physical engagement and motivational profiles were assessed by both closed and open-ended questions, such as "Since your 9-month study visit, has your physical activity been more, less, or about the same (compared to the first 9 months of the study)?" and, "What motivates you the most to be physically active?" NLP and cluster analysis were used to classify motivational profiles. Descriptive statistics were used to compare participants' baseline characteristics among identified groups. RESULTS: Approximately half of the 2 intervention groups (Regular and Plus) reported that they were still wearing an accelerometer and engaging in brisk walking as they were directed during the intervention phases. These numbers in the 2 intervention groups were much higher than the control group (overall P=.01 and P=.003, respectively). Three clusters were identified through NLP and named as the Weight Loss group (n=19), the Illness Prevention group (n=138), and the Health Promotion group (n=46). The Weight Loss group was significantly younger than the Illness Prevention and Health Promotion groups (overall P<.001). The Illness Prevention group had a larger number of Caucasians as compared to the Weight Loss group (P=.001), which was composed mostly of those who identified as African American, Hispanic, or mixed race. Additionally, the Health Promotion group tended to have lower BMI scores compared to the Illness Prevention group (overall P=.02). However, no difference was noted in the baseline moderate-to-vigorous intensity activity level among the 3 groups (overall P>.05). CONCLUSIONS: The findings could be relevant to tailoring a physical activity maintenance intervention. Furthermore, the findings from NLP and cluster analysis are useful methods to analyze short free text to differentiate motivational profiles. As more sophisticated NL tools are developed in the future, the potential of NLP application in behavioral research will broaden. TRIAL REGISTRATION: ClinicalTrials.gov NCT01280812; https://clinicaltrials.gov/ct2/show/NCT01280812 (Archived by WebCite at http://www.webcitation.org/70IkGagAJ).

Objectively Measured Baseline Physical Activity Patterns in Women in the mPED Trial: Cluster Analysis.

BACKGROUND: Determining patterns of physical activity throughout the day could assist in developing more personalized interventions or physical activity guidelines in general and, in particular, for women who are less likely to be physically active than men. OBJECTIVE: The aims of this report are to identify clusters of women based on accelerometer-measured baseline raw metabolic equivalent of task (MET) values and a normalized version of the METs ≥3 data, and to compare sociodemographic and cardiometabolic risks among these identified clusters. METHODS: A total of 215 women who were enrolled in the Mobile Phone Based Physical Activity Education (mPED) trial and wore an accelerometer for at least 8 hours per day for the 7 days prior to the randomization visit were analyzed. The k-means clustering method and the Lloyd algorithm were used on the data. We used the elbow method to choose the number of clusters, looking at the percentage of variance explained as a function of the number of clusters. RESULTS: The results of the k-means cluster analyses of raw METs revealed three different clusters. The unengaged group (n=102) had the highest depressive symptoms score compared with the afternoon engaged (n=65) and morning engaged (n=48) groups (overall P<.001). Based on a normalized version of the METs ≥3 data, the moderate-to-vigorous physical activity (MVPA) evening peak group (n=108) had a higher body mass index (P=.03), waist circumference (P=.02), and hip circumference (P=.03) than the MVPA noon peak group (n=61). CONCLUSIONS: Categorizing physically inactive individuals into more specific activity patterns could aid in creating timing, frequency, duration, and intensity of physical activity interventions for women. Further research is needed to confirm these cluster groups using a large national dataset. TRIAL REGISTRATION: ClinicalTrials.gov NCT01280812; https://clinicaltrials.gov/ct2/show/NCT01280812 (Archived by WebCite at http://www.webcitation.org/6vVyLzwft).

Evaluating Machine Learning-Based Automated Personalized Daily Step Goals Delivered Through a Mobile Phone App: Randomized Controlled Trial.

BACKGROUND: Growing evidence shows that fixed, nonpersonalized daily step goals can discourage individuals, resulting in unchanged or even reduced physical activity. OBJECTIVE: The aim of this randomized controlled trial (RCT) was to evaluate the efficacy of an automated mobile phone-based personalized and adaptive goal-setting intervention using machine learning as compared with an active control with steady daily step goals of 10,000. METHODS: In this 10-week RCT, 64 participants were recruited via email announcements and were required to attend an initial in-person session. The participants were randomized into either the intervention or active control group with a one-to-one ratio after a run-in period for data collection. A study-developed mobile phone app (which delivers daily step goals using push notifications and allows real-time physical activity monitoring) was installed on each participant's mobile phone, and participants were asked to keep their phone in a pocket throughout the entire day. Through the app, the intervention group received fully automated adaptively personalized daily step goals, and the control group received constant step goals of 10,000 steps per day. Daily step count was objectively measured by the study-developed mobile phone app. RESULTS: The mean (SD) age of participants was 41.1 (11.3) years, and 83% (53/64) of participants were female. The baseline demographics between the 2 groups were similar (P>.05). Participants in the intervention group (n=34) had a decrease in mean (SD) daily step count of 390 (490) steps between run-in and 10 weeks, compared with a decrease of 1350 (420) steps among control participants (n=30; P=.03). The net difference in daily steps between the groups was 960 steps (95% CI 90-1830 steps). Both groups had a decrease in daily step count between run-in and 10 weeks because interventions were also provided during run-in and no natural baseline was collected. CONCLUSIONS: The results showed the short-term efficacy of this intervention, which should be formally evaluated in a full-scale RCT with a longer follow-up period. TRIAL REGISTRATION: ClinicalTrials.gov: NCT02886871; https://clinicaltrials.gov/ct2/show/NCT02886871 (Archived by WebCite at http://www.webcitation.org/6wM1Be1Ng).

Feasibility of Reidentifying Individuals in Large National Physical Activity Data Sets From Which Protected Health Information Has Been Removed With Use of Machine Learning.

Importance: Despite data aggregation and removal of protected health information, there is concern that deidentified physical activity (PA) data collected from wearable devices can be reidentified. Organizations collecting or distributing such data suggest that the aforementioned measures are sufficient to ensure privacy. However, no studies, to our knowledge, have been published that demonstrate the possibility or impossibility of reidentifying such activity data. Objective: To evaluate the feasibility of reidentifying accelerometer-measured PA data, which have had geographic and protected health information removed, using support vector machines (SVMs) and random forest methods from machine learning. Design, Setting, and Participants: In this cross-sectional study, the National Health and Nutrition Examination Survey (NHANES) 2003-2004 and 2005-2006 data sets were analyzed in 2018. The accelerometer-measured PA data were collected in a free-living setting for 7 continuous days. NHANES uses a multistage probability sampling design to select a sample that is representative of the civilian noninstitutionalized household (both adult and children) population of the United States. Exposures: The NHANES data sets contain objectively measured movement intensity as recorded by accelerometers worn during all walking for 1 week. Main Outcomes and Measures: The primary outcome was the ability of the random forest and linear SVM algorithms to match demographic and 20-minute aggregated PA data to individual-specific record numbers, and the percentage of correct matches by each machine learning algorithm was the measure. Results: A total of 4720 adults (mean [SD] age, 40.0 [20.6] years) and 2427 children (mean [SD] age, 12.3 [3.4] years) in NHANES 2003-2004 and 4765 adults (mean [SD] age, 45.2 [19.9] years) and 2539 children (mean [SD] age, 12.1 [3.4] years) in NHANES 2005-2006 were included in the study. The random forest algorithm successfully reidentified the demographic and 20-minute aggregated PA data of 4478 adults (94.9%) and 2120 children (87.4%) in NHANES 2003-2004 and 4470 adults (93.8%) and 2172 children (85.5%) in NHANES 2005-2006 (P < .001 for all). The linear SVM algorithm successfully reidentified the demographic and 20-minute aggregated PA data of 4043 adults (85.6%) and 1695 children (69.8%) in NHANES 2003-2004 and 4041 adults (84.8%) and 1705 children (67.2%) in NHANES 2005-2006 (P < .001 for all). Conclusions and Relevance: This study suggests that current practices for deidentification of accelerometer-measured PA data might be insufficient to ensure privacy. This finding has important policy implications because it appears to show the need for deidentification that aggregates the PA data of multiple individuals to ensure privacy for single individuals.

Personalizing Mobile Fitness Apps using Reinforcement Learning.

Despite the vast number of mobile fitness applications (apps) and their potential advantages in promoting physical activity, many existing apps lack behavior-change features and are not able to maintain behavior change motivation. This paper describes a novel fitness app called CalFit, which implements important behavior-change features like dynamic goal setting and self-monitoring. CalFit uses a reinforcement learning algorithm to generate personalized daily step goals that are challenging but attainable. We conducted the Mobile Student Activity Reinforcement (mSTAR) study with 13 college students to evaluate the efficacy of the CalFit app. The control group (receiving goals of 10,000 steps/day) had a decrease in daily step count of 1,520 (SD ± 740) between baseline and 10-weeks, compared to an increase of 700 (SD ± 830) in the intervention group (receiving personalized step goals). The difference in daily steps between the two groups was 2,220, with a statistically significant p = 0.039.

Expression-level optimization of a multi-enzyme pathway in the absence of a high-throughput assay.

Engineered metabolic pathways often suffer from flux imbalances that can overburden the cell and accumulate intermediate metabolites, resulting in reduced product titers. One way to alleviate such imbalances is to adjust the expression levels of the constituent enzymes using a combinatorial expression library. Typically, this approach requires high-throughput assays, which are unfortunately unavailable for the vast majority of desirable target compounds. To address this, we applied regression modeling to enable expression optimization using only a small number of measurements. We characterized a set of constitutive promoters in Saccharomyces cerevisiae that spanned a wide range of expression and maintained their relative strengths irrespective of the coding sequence. We used a standardized assembly strategy to construct a combinatorial library and express for the first time in yeast the five-enzyme violacein biosynthetic pathway. We trained a regression model on a random sample comprising 3% of the total library, and then used that model to predict genotypes that would preferentially produce each of the products in this highly branched pathway. This generalizable method should prove useful in engineering new pathways for the sustainable production of small molecules.

Nonparametric variable selection and modeling for spatial and temporal regulatory networks.

Because of the increasing diversity of data sets and measurement techniques in biology, a growing spectrum of modeling methods is being developed. It is generally recognized that it is critical to pick the appropriate method to exploit the amount and type of biological data available for a given system. Here, we describe a method for use in situations where temporal data from a network is collected over multiple time points, and in which little prior information is available about the interactions, mathematical structure, and statistical distribution of the network. Our method results in models that we term Nonparametric exterior derivative estimation Ordinary Differential Equation (NODE) model's. We illustrate the method's utility using spatiotemporal gene expression data from Drosophila melanogaster embryos. We demonstrate that the NODE model's use of the temporal characteristics of the network leads to quantifiable improvements in its predictive ability over nontemporal models that only rely on the spatial characteristics of the data. The NODE model provides exploratory visualizations of network behavior and structure, which can identify features that suggest additional experiments. A new extension is also presented that uses the NODE model to generate a comb diagram, a figure that presents a list of possible network structures ranked by plausibility. By being able to quantify a continuum of interaction likelihoods, this helps to direct future experiments.

Nonparametric identification of regulatory interactions from spatial and temporal gene expression data.

BACKGROUND: The correlation between the expression levels of transcription factors and their target genes can be used to infer interactions within animal regulatory networks, but current methods are limited in their ability to make correct predictions. RESULTS: Here we describe a novel approach which uses nonparametric statistics to generate ordinary differential equation (ODE) models from expression data. Compared to other dynamical methods, our approach requires minimal information about the mathematical structure of the ODE; it does not use qualitative descriptions of interactions within the network; and it employs new statistics to protect against over-fitting. It generates spatio-temporal maps of factor activity, highlighting the times and spatial locations at which different regulators might affect target gene expression levels. We identify an ODE model for eve mRNA pattern formation in the Drosophila melanogaster blastoderm and show that this reproduces the experimental patterns well. Compared to a non-dynamic, spatial-correlation model, our ODE gives 59% better agreement to the experimentally measured pattern. Our model suggests that protein factors frequently have the potential to behave as both an activator and inhibitor for the same cis-regulatory module depending on the factors' concentration, and implies different modes of activation and repression. CONCLUSIONS: Our method provides an objective quantification of the regulatory potential of transcription factors in a network, is suitable for both low- and moderate-dimensional gene expression datasets, and includes improvements over existing dynamic and static models.