Use of machine learning to identify patients at risk of sub-optimal adherence: study based on real-world data from 10,929 children using a connected auto-injector device

Risks and benefits associated with the primary functions of artificial intelligence powered autoinjectors

Objectives

This research aims to present and assess the Primary Functions of autoinjectors introduced in ISO 11608-1:2022. Investigate the risks in current autoinjector technology, identify and assess risks and benefits associated with Artificial Intelligence (AI) powered autoinjectors, and propose a framework for mitigating these risks. ISO 11608-1:2022 is a standard that specifies requirements and test methods for needle-based injection systems intended to deliver drugs, focusing on design and function to ensure patient safety and product effectiveness. 'KZH' is an FDA product code used to classify autoinjectors, for regulatory purposes, ensuring they meet defined safety and efficacy standards before being marketed.

Method

A comprehensive analysis of autoinjectors problems is conducted using data from the United States Food and Drug Administration (FDA) database. This database records medical device reporting events, including those related to autoinjectors, reported by various sources. The analysis focuses on events associated with the product code KZH, covering data from January 1, 2008, to September 30, 2023. This research employs statistical frequency analysis and incorporates pertinent the FDA, United Kingdom, European Commission regulations, and ISO standards.

Results

500 medical device reporting events are assessed for autoinjectors under the KZH code. Ultimately, 188 of these events are confirmed to be associated with autoinjectors, all 500 medical devices were seen to lack AI capabilities. An analysis of these events for traditional mechanical autoinjectors revealed a predominant occurrence of malfunctions (72%) and injuries (26%) among event types. Device problems, such as breakage, defects, jams, and others, accounted for 45% of incidents, while 10% are attributed to patient problems, particularly missed and underdoses.

Conclusion

Traditional autoinjectors are designed to assist patients in medication administration, underscoring the need for quality control, reliability, and design enhancements. AI autoinjectors, sharing this goal, bring additional cybersecurity and software risks, requiring a comprehensive risk management framework that includes standards, tools, training, and ongoing monitoring. The integration of AI promises to improve functionality, enable real-time monitoring, and facilitate remote clinical trials, timely interventions, and tailored medical treatments.

Artificial intelligence in paediatric endocrinology: conflict or cooperation

Artificial intelligence (AI) in medicine is transforming healthcare by automating system tasks, assisting in diagnostics, predicting patient outcomes and personalising patient care, founded on the ability to analyse vast datasets. In paediatric endocrinology, AI has been developed for diabetes, for insulin dose adjustment, detection of hypoglycaemia and retinopathy screening; bone age assessment and thyroid nodule screening; the identification of growth disorders; the diagnosis of precocious puberty; and the use of facial recognition algorithms in conditions such as Cushing syndrome, acromegaly, congenital adrenal hyperplasia and Turner syndrome. AI can also predict those most at risk from childhood obesity by stratifying future interventions to modify lifestyle. AI will facilitate personalised healthcare by integrating data from 'omics' analysis, lifestyle tracking, medical history, laboratory and imaging, therapy response and treatment adherence from multiple sources. As data acquisition and processing becomes fundamental, data privacy and protecting children's health data is crucial. Minimising algorithmic bias generated by AI analysis for rare conditions seen in paediatric endocrinology is an important determinant of AI validity in clinical practice. AI cannot create the patient-doctor relationship or assess the wider holistic determinants of care. Children have individual needs and vulnerabilities and are considered in the context of family relationships and dynamics. Importantly, whilst AI provides value through augmenting efficiency and accuracy, it must not be used to replace clinical skills.

Exploring patient medication adherence and data mining methods in clinical big data: A contemporary review

BACKGROUND

Increasingly, patient medication adherence data are being consolidated from claims databases and electronic health records (EHRs). Such databases offer an indirect avenue to gauge medication adherence in our data-rich healthcare milieu. The surge in data accessibility, coupled with the pressing need for its conversion to actionable insights, has spotlighted data mining, with machine learning (ML) emerging as a pivotal technique. Nonadherence poses heightened health risks and escalates medical costs. This paper elucidates the synergistic interaction between medical database mining for medication adherence and the role of ML in fostering knowledge discovery.

METHODS

We conducted a comprehensive review of EHR applications in the realm of medication adherence, leveraging ML techniques. We expounded on the evolution and structure of medical databases pertinent to medication adherence and harnessed both supervised and unsupervised ML paradigms to delve into adherence and its ramifications.

RESULTS

Our study underscores the applications of medical databases and ML, encompassing both supervised and unsupervised learning, for medication adherence in clinical big data. Databases like SEER and NHANES, often underutilized due to their intricacies, have gained prominence. Employing ML to excavate patient medication logs from these databases facilitates adherence analysis. Such findings are pivotal for clinical decision-making, risk stratification, and scholarly pursuits, aiming to elevate healthcare quality.

CONCLUSION

Advanced data mining in the era of big data has revolutionized medication adherence research, thereby enhancing patient care. Emphasizing bespoke interventions and research could herald transformative shifts in therapeutic modalities.

Curve matching to predict growth in patients receiving growth hormone therapy: An interpretable & explainable method

Curve matching may be used to predict growth outcomes using data of patients whose growth curves resemble those of a new patient with growth hormone deficiency (GHD) and those born small for gestational age (SGA). We aimed to investigate the validity of curve matching to predict growth in patients with GHD and those born SGA receiving recombinant human growth hormone (r-hGH). Height data collected between 0-48 months of treatment were extracted from the easypod™ connect ecosystem and the easypod™ connect observational study. Selected patients with height standard deviation scores (HSDS) [-4, <-1] and age [3, <16y] at start were included. The 'Matching Database' consisted of patients' monthly HSDS obtained by the broken stick method and imputation. Standard deviation (SD) was obtained from the observed minus the predicted HSDS (error) based on matched patients within the 'Matching Database'. Data were available for 3,213 patients in the 'Matching Database', and 2,472 patients with 16,624 HSDS measurements in the observed database. When ≥2 HSDS measurements were available, the error SD for a one-year prediction was approximately 0.2, which corresponds to 1.1 cm, 1.3 cm, and 1.5 cm at 7, 11, and 15 years of age, respectively. Indication and age at treatment start (<11 vs ≥11 years) had a small impact on the error SD, with patients born SGA and patients aged <11 years at treatment start generally having slightly lower values. We conclude that curve matching is a simple and valid technique for predicting growth in patients with GHD and those born SGA.