RESUMEN
In recent years, the application of machine learning for virtual sensing has revolutionized the monitoring and management of information. In particular, electrochemical sensors generate large amounts of data, allowing the application of complex machine learning/AI models able to (1) reproduce the measured data and (2) predict and manage faults in the measuring sensor. In this work, data-driven models based on an autoregressive model and an artificial neural network have been identified and used to (i) evaluate sensor redundancy and (ii) predict and manage faults in the context of electrochemical sensors for the measurement of ethanol. The approach shows encouraging results in terms of both performance and sensitivity analyses, allowing for the reconstruction of the values measured by two sensors in a series of six sensors with different dopant levels and to reproduce their values after a fault.
RESUMEN
Adolescent idiopathic scoliosis (AIS) is a three-dimensional spine and trunk deformity. Bracing is an effective treatment for medium-degree curves. Thermal sensors help monitor patients' adherence (compliance), a critical issue in bracing treatment. Some studies investigated adherence determinants but rarely through sensors or in highly adherent cohorts. We aimed to verify the influence of personal and clinical variables routinely registered by physicians on adherence to brace treatment in a large cohort of consecutive AIS patients from a highly adherent cohort. We performed a cross-sectional study of patients consecutively recruited in the last three years at a tertiary referral institute and treated with braces for one year. To ensure high adherence, for years, we have provided specific support to brace treatment through a series of cognitive-behavioural interventions for patients and parents. We used iButton thermal sensor systematic data collection to precisely analyse the real brace-wearing time. We included 514 adolescents, age 13.8 ± 1.6, with the worst scoliosis curve of 34.5 ± 10.3° Cobb. We found a 95% (95CI 60-101%) adherence to the brace prescription of 21.9 ± 1.7 h per day. Determinants included gender (91% vs. 84%; females vs. males) and age < 14 years (92% vs. 88%). Brace hours prescription, BMI, and all clinical variables (worst curve Cobb degrees, angle of trunk rotation, and TRACE index for aesthetics) did not influence adherence.