Personalized Adaptive CBR Bolus Recommender System for Type 1 Diabetes

Abstract

Type 1 Diabetes Mellitus (T1DM) is a chronic disease. Those who have it must administer themselves with insulin to control their blood glucose level. It is difficult to estimate the correct insulin dosage due to the complex glucose metabolism, which can lead to less than optimal blood glucose levels. This paper presents PepperRec, a case-based reasoning (CBR) bolus insulin recommender system capable of dealing with an unrestricted number of situations in which T1DM persons can find themselves. PepperRec considers several factors that affect glucose metabolism, such as data about the physical activity of the user, and can also cope with missing values for these factors. Based on CBR methodology, PepperRec uses new methods to adapt past recommendations to the current state of the user, and retains updated historical patient information to deal with slow and gradual changes in the patient over time (concept drift). The proposed approach is tested using the UVA/PADOVA simulator with 33 virtual subjects and compared with other methods in the literature, and with the default insulin therapy of the simulator. The achieved results demonstrate that PepperRec increases the amount of time the users are in their target glycaemic range, reduces the time spent below it, while maintaining, or even reducing, the time spent above it.

Recommender System Based on Temporal Models: A Systematic Review

“…The DRS has been extensively researched in recent ages across different fields such as multimedia [7,22,62,63] e-commerce [35,[63][64][65][66]85,[97][98][99][100][101], e-documents [67][68][69][102][103][104][105][106][107][108], Travel, Tourism and Places [8,10,13,30,37,109], and others [9,12,87,88,94,[110][111][112][113][114][115][116]. This was said to have started after the Netflix competition in 2009, where the time changing user behaviors were considered to improve recommendation accuracy [70][71][72][73][74][75]…”

Multidimensional Group Recommendations in the Health Domain

“…The work in [37] is a recommender system proposing physical activities using only user's history and employing machine learning, whereas for chronic conditions, other works focus on integrating recommender systems with electronic health records [38,39], proposing the best course of treatment. Other approaches adapt past recommendations to the current state of the user for Diabetes patients [40] or propose context-aware recommendation methods [41] to establish personalized healthcare services. However, all these works use techniques that are principally found in pure group recommendations systems for composing the group recommendation list.…”

Adaptive basal insulin recommender system based on Kalman filter for type 1 diabetes

“…This has led researchers to develop new methods that automatically adjust BC parameters such as those of (Herrero et al, 2015a,b;Torrent-Fontbona et al, 2017) which then recommend an accurate dosage. However, these methods require an optimised basal insulin to achieve good results.…”

“…In this regard, (Torrent-Fontbona et al, 2017) propose a method that uses physical activity, time of day and meal size to define the cases of a CBR system. The method models the changes in terms of ICR and ISF according to physical activity, time of day and meal size throughout a set of cases, to then estimate the value of ICR and ISF under a new situation.…”

“…Nevertheless, a variation of the algorithm that does not require the use of CGM is also presented. The methodology is tested in combination with a traditional BC and with an adaptive BC based on Case-Based Reasoning (CBR) presented in the previous work (Torrent-Fontbona et al, 2017). The system is tested with ten virtual adults and an average adult using the UVa/PADOVA T1DM simulator.…”