Can the artificial intelligence technique of reinforcement learning use continuously-monitored digital data to optimize treatment for weight loss?

Abstract

Behavioral weight loss (WL) trials show that, on average, participants regain lost weight unless provided long-term, intensive-and thus costly-intervention. Optimization solutions have shown mixed success. The artificial intelligence principle of "reinforcement learning" (RL) offers a new and more sophisticated form of optimization in which the intensity of each individual's intervention is continuously adjusted depending on patterns of response. In this pilot, we evaluated the feasibility and acceptability of a RL-based WL intervention, and whether optimization would achieve equivalent benefit at a reduced cost compared to a non-optimized intensive intervention. Participants (n = 52) completed a 1-month, group-based in-person behavioral WL intervention and then (in Phase II) were randomly assigned to receive 3 months of twice-weekly remote interventions that were non-optimized (NO; 10-min phone calls) or optimized (a combination of phone calls, text exchanges, and automated messages selected by an algorithm). The Individually-Optimized (IO) and Group-Optimized (GO) algorithms selected interventions based on past performance of each intervention for each participant, and for each group member that fit into a fixed amount of time (e.g., 1 h), respectively. Results indicated that the system was feasible to deploy and acceptable to participants and coaches. As hypothesized, we were able to achieve equivalent Phase II weight losses (NO = 4.42%, IO = 4.56%, GO = 4.39%) at roughly one-third the cost (1.73 and 1.77 coaching hours/participant for IO and GO, versus 4.38 for NO), indicating strong promise for a RL system approach to weight loss and maintenance.

The potential of artificial intelligence in enhancing adult weight loss: a scoping review

“…This could be attributed to the personalisation of goals that were coherent with each users' lifestyle habits based on the information retrieved from their calendar apps (indicates availability for exercise) and health app (indicates activity patterns) (26) . Weight loss outcomes ranged from an average of 2•4 -4•7 % (29,(48)(49)(50)106) of which only two were statistically significant (P < 0•05) (26,50) . Three studies reported the use of Bluetooth enabled weighing machines that synchronise weight data to the users' phone apps, while the rest used manually-input weight.…”

“…Three studies (26,53,54) focused on only improving physical activity, four studies focused on only improving dietary behaviours (48)(49)(50)106) and three studies (29,51,52) focused on both. All five studies (29,48,49,50,53) on dietary lapse prevention reported percentage increases in dietary adherence, but only one study reported statistically significant results (P < 0•05), suggesting mixed findings (50) . Two of the three studies on preventing exercise lapses reported significant (P < 0•05) increases in step count and metabolic equivalent task (26,53) .…”

“…Therefore, individuals trying to lose weight often attempt to either increase self-regulation capacity through sheer willpower (25) or reduce the self-regulation effort needed through weight-loss mobile apps (26) , clinical weight management programmes (27) and commercial weight-loss programmes (28) . However, such methods are often expensive, resource-intensive and unsustainable (29) . An emerging strategy to tackle this problem of poor selfregulation is to apply artificial intelligence (AI) (30) .…”

A Clustering Approach for Modeling and Analyzing Changes in Physical Activity Behaviors From Accelerometers

“…Such techniques have been used for educational purposes. For instance, [28] created a decision-making algorithm to optimize a reinforcement learning system for a weight loss intervention program using pre-classified data from Fitbit sensors, together with intervention responses. [29] created a gamified intervention to motivate children to increase every-day physical activity where Support Vector Machine and Random Forests (RF) data mining techniques were applied to smartphone accelerometer data in order to (i) classify children's PA into activities (sitting, standing, walking, jogging, walking upstairs, walking downstairs and intense physical activity) and (ii) calculate a score based on the amount of time spent in these activities.…”

Evaluating text message-delivered, extended contact interventions targeting healthy weight, diet and physical activity behaviour

“…Alternatively a program that incorporates text messages which are supplemented with additional telephone support for emotional and social well-being may address this deficit and may be more appropriate extended contact for some participants (Singleton et al, 2019, Gell et al, 2019b, Job et al, 2017b. Furthermore, with recent developments, this support may be feasible through triaging levels of intervention intensity via artificial intelligence , Forman et al, 2019 or supplementing with chatbots to support skill development for psychological health (Greer et al, 2019).…”