Thanks to the availablity of preliminary data from the City4Age Athens Pilot, MultiMed Engineers has started to work on Machine Learning algorithms for assessing the potential of early detection of frailty in aging citizens, using unobtrusive technologies available in smart-cities.
Ground truth has been collected by using the Functional Ability Index (FAI) developed by Dapp et al., which is a validated instrument with strong predictive potential for the onset of frailty in aging populations.
First results are encouraging, as they show meaningful association patterns among measures collected by the City4Age technology and FAI scores, as shown for example in the following pictures.
Several Machine Learning schemes, experimented by MultiMed Engineers, seem to be able to pick up the above patters and learn classifiers with AUCs that are significantly different from randomness.
The WEKA software has been used to conduct the experiments, applying the GainRatioAttributeEval class for feature selection, in conjuction with the classes representing 21 Machine Learning schemes. Different learning schemes and different numbers of features have been compared, applying nested cross-validation (to avoid bias in performance estimates) with the MultiScheme class.
The best performing schemes were Stochastic Gradient Descent with Log loss function and Naive Bayes with 9-10 features. They produced AUC values around 0.7-0.8 on average.
Although these are preliminary results, based on few datapoints, they clearly represent an important proof-of-concept that vindicates the ambition of City4Age — i.e. to efficiently detect the onset of aging conditions by leveraging smart-city supported, unobtrusive data collection.