Characterization of Synthetic Health Data Using Rule-Based Artificial Intelligence Models.

The aim of this study is to apply and characterize eXplainable AI (XAI) to assess the quality of synthetic health data generated using a data augmentation algorithm. In this exploratory study, several synthetic datasets are generated using various configurations of a conditional Generative Adversarial Network (GAN) from a set of 156 observations related to adult hearing screening. A rule-based native XAI algorithm, the Logic Learning Machine, is used in combination with conventional utility metrics. The classification performance in different conditions is assessed: models trained and tested on synthetic data, models trained on synthetic data and tested on real data, and models trained on real data and tested on synthetic data. The rules extracted from real and synthetic data are then compared using a rule similarity metric. The results indicate that XAI may be used to assess the quality of synthetic data by (i) the analysis of classification performance and (ii) the analysis of the rules extracted on real and synthetic data (number, covering, structure, cut-off values, and similarity). These results suggest that XAI can be used in an original way to assess synthetic health data and extract knowledge about the mechanisms underlying the generated data.

Final results of the NINFA project: impact of new technologies in the daily life of elderly people.

BACKGROUND: The paper presents the work carried out within NINFA (iNtelligent Integrated Network For Aged people), a project for the wellbeing of the elderly people at home. AIMS: The impact of new technologies on elderly people is evaluated with respect to the three main topics faced by NINFA. METHODS: NINFA was structured into three main topics: (1) active user engagement from the very beginning of the planning stage: the use of specially designed questionnaires to evaluate the acceptability of new technology in general and robot caregiver specifically; (2) assessment of the well-being through non-invasive techniques: natural language processing for language change monitoring in elderly subjects; (3) automated assessment of motor and cognitive functions at home: systems to deliver tests and exergames through user interfaces compliant with elderly subjects. RESULTS: The analysis shows that there is no a priori closure to support the technology, but it must not be invasive and must allow social interactions. The study of speech transcripts shows that a large variations in the number of words used to describe the same situation could be a sign on the onset of cognitive impairments. The specifically designed systems highlight, after the training period, significant improvements in the performances of the participants and a satisfaction with regards to the systems usability. CONCLUSIONS: The outcomes of NINFA project highlight some important aspects of the relationship between elderly people and new technologies concerning: engagement and acceptability, assessment of the wellbeing and of the modifications of motor, cognitive and language functions.