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.

The effect of variable labels on deep learning models trained to predict breast density

Purpose. High breast density is associated with reduced efficacy of mammographic screening and increased risk of developing breast cancer. Accurate and reliable automated density estimates can be used for direct risk prediction and passing density related information to further predictive models. Expert reader assessments of density show a strong relationship to cancer risk but also inter-reader variation. The effect of label variability on model performance is important when considering how to utilise automated methods for both research and clinical purposes.Methods. We utilise subsets of images with density labels from the same 13 readers and 12 reader pairs, and train a deep transfer learning model which is used to assess how label variability affects the mapping from representation to prediction. We then create two end-to-end models: one that is trained on averaged labels across the reader pairs and the second that is trained using individual reader scores, with a novel alteration to the objective function. The combination of these two end-to-end models allows us to investigate the effect of label variability on the model representation formed.Results. We show that the trained mappings from representations to labels are altered considerably by the variability of reader scores. Training on labels with distribution variation removed causes the Spearman rank correlation coefficients to rise from 0.751 ± 0.002 to either 0.815 ± 0.026 when averaging across readers or 0.844 ± 0.002 when averaging across images. However, when we train different models to investigate the representation effect we see little difference, with Spearman rank correlation coefficients of 0.846 ± 0.006 and 0.850 ± 0.006 showing no statistically significant difference in the quality of the model representation with regard to density prediction.Conclusions. We show that the mapping between representation and mammographic density prediction is significantly affected by label variability. However, the effect of the label variability on the model representation is limited.

Digital Transformation in Healthcare: Technology Acceptance and Its Applications

Technological innovation has become an integral aspect of our daily life, such as wearable and information technology, virtual reality and the Internet of Things which have contributed to transforming healthcare business and operations. Patients will now have a broader range and more mindful healthcare choices and experience a new era of healthcare with a patient-centric culture. Digital transformation determines personal and institutional health care. This paper aims to analyse the changes taking place in the field of healthcare due to digital transformation. For this purpose, a systematic bibliographic review is performed, utilising Scopus, Science Direct and PubMed databases from 2008 to 2021. Our methodology is based on the approach by Wester and Watson, which classify the related articles based on a concept-centric method and an ad hoc classification system which identify the categories used to describe areas of literature. The search was made during August 2022 and identified 5847 papers, of which 321 fulfilled the inclusion criteria for further process. Finally, by removing and adding additional studies, we ended with 287 articles grouped into five themes: information technology in health, the educational impact of e-health, the acceptance of e-health, telemedicine and security issues.

MetaboVariation: Exploring Individual Variation in Metabolite Levels

To date, most metabolomics biomarker research has focused on identifying disease biomarkers. However, there is a need for biomarkers of early metabolic dysfunction to identify individuals who would benefit from lifestyle interventions. Concomitantly, there is a need to develop strategies to analyse metabolomics data at an individual level. We propose "MetaboVariation", a method that models repeated measurements on individuals to explore fluctuations in metabolite levels at an individual level. MetaboVariation employs a Bayesian generalised linear model to flag individuals with intra-individual variations in their metabolite levels across multiple measurements. MetaboVariation models repeated metabolite levels as a function of explanatory variables while accounting for intra-individual variation. The posterior predictive distribution of metabolite levels at the individual level is available, and is used to flag individuals with observed metabolite levels outside the 95% highest posterior density prediction interval at a given time point. MetaboVariation was applied to a dataset containing metabolite levels for 20 metabolites, measured once every four months, in 164 individuals. A total of 28% of individuals with intra-individual variations in three or more metabolites were flagged. An R package for MetaboVariation was developed with an embedded R Shiny web application. To summarize, MetaboVariation has made considerable progress in developing strategies for analysing metabolomics data at the individual level, thus paving the way toward personalised healthcare.

Personal Digital Twin: A Close Look into the Present and a Step towards the Future of Personalised Healthcare Industry

Digital twins (DTs) play a vital role in revolutionising the healthcare industry, leading to more personalised, intelligent, and proactive healthcare. With the evolution of personalised healthcare, there is a significant need to represent a virtual replica for individuals to provide the right type of care in the right way and at the right time. Therefore, in this paper, we surveyed the concept of a personal digital twin (PDT) as an enhanced version of the DT with actionable insight capabilities. In particular, PDT can bring value to patients by enabling more accurate decision making and proper treatment selection and optimisation. Then, we explored the progression of PDT as a revolutionary technology in healthcare research and industry. However, although several research works have been performed for smart healthcare using DT, PDT is still at an early stage. Consequently, we believe that this work can be a step towards smart personalised healthcare industry by guiding the design of industrial personalised healthcare systems. Accordingly, we introduced a reference framework that empowers smart personalised healthcare using PDTs by bringing together existing advanced technologies (i.e., DT, blockchain, and AI). Then, we described some selected use cases, including the mitigation of COVID-19 contagion, COVID-19 survivor follow-up care, personalised COVID-19 medicine, personalised osteoporosis prevention, personalised cancer survivor follow-up care, and personalised nutrition. Finally, we identified further challenges to pave the PDT paradigm toward the smart personalised healthcare industry.

Personalised Medicine Using N-of-1 Trials: Overcoming Barriers to Delivery

In this paper we discuss the value of N-of-1 trials in personalising health care. We describe the challenges faced in implementing N-of-1 trials in the United Kingdom’s National Health Service and suggest how making greater use of these personalised trials might be facilitated.

[Artificial intelligence in healthcare and the transformation of healthcare professions]

By integrating artificial intelligence (AI) to their practice, healthcare professions will evolve towards more efficient patient management and better quality of care. These changes require new competencies to which all professionals must be trained. A methodology to quantify the impact of AI on healthcare occupations can be used to support and anticipate these changes.

Personalised measures of obesity using waist to height ratios from an Australian health screening program

Objectives

The aim of the current study is to generate waist circumference to height ratio cut-off values for obesity categories from a model of the relationship between body mass index and waist circumference to height ratio. We compare the waist circumference to height ratio discovered in this way with cut-off values currently prevalent in practice that were originally derived using pragmatic criteria.

Method

Personalized data including age, gender, height, weight, waist circumference and presence of diabetes, hypertension and cardiovascular disease for 847 participants over eight years were assembled from participants attending a rural Australian health review clinic (DiabHealth). Obesity was classified based on the conventional body mass index measure (weight/height²) and compared to the waist circumference to height ratio. Correlations between the measures were evaluated on the screening data, and independently on data from the National Health and Nutrition Examination Survey that included age categories.

Results

This article recommends waist circumference to height ratio cut-off values based on an Australian rural sample and verified using the National Health and Nutrition Examination Survey database that facilitates the classification of obesity in clinical practice. Gender independent cut-off values are provided for waist circumference to height ratio that identify healthy (waist circumference to height ratio ≥0.45), overweight (0.53) and the three obese (0.60, 0.68, 0.75) categories verified on the National Health and Nutrition Examination Survey dataset. A strong linearity between the waist circumference to height ratio and the body mass index measure is demonstrated.

Conclusion

The recommended waist circumference to height ratio cut-off values provided a useful index for assessing stages of obesity and risk of chronic disease for improved healthcare in clinical practice.

Clinical efficacy and effectiveness of 3D printing: a systematic review

OBJECTIVE

To evaluate the clinical efficacy and effectiveness of using 3D printing to develop medical devices across all medical fields.

DESIGN

Systematic review compliant with Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

DATA SOURCES

PubMed, Web of Science, OVID, IEEE Xplore and Google Scholar.

METHODS

A double-blinded review method was used to select all abstracts up to January 2017 that reported on clinical trials of a three-dimensional (3D)-printed medical device. The studies were ranked according to their level of evidence, divided into medical fields based on the International Classification of Diseases chapter divisions and categorised into whether they were used for preoperative planning, aiding surgery or therapy. The Downs and Black Quality Index critical appraisal tool was used to assess the quality of reporting, external validity, risk of bias, risk of confounding and power of each study.

RESULTS

Of the 3084 abstracts screened, 350 studies met the inclusion criteria. Oral and maxillofacial surgery contained 58.3% of studies, and 23.7% covered the musculoskeletal system. Only 21 studies were randomised controlled trials (RCTs), and all fitted within these two fields. The majority of RCTs were 3D-printed anatomical models for preoperative planning and guides for aiding surgery. The main benefits of these devices were decreased surgical operation times and increased surgical accuracy.

CONCLUSIONS

All medical fields that assessed 3D-printed devices concluded that they were clinically effective. The fields that most rigorously assessed 3D-printed devices were oral and maxillofacial surgery and the musculoskeletal system, both of which concluded that the 3D-printed devices outperformed their conventional comparators. However, the efficacy and effectiveness of 3D-printed devices remain undetermined for the majority of medical fields. 3D-printed devices can play an important role in healthcare, but more rigorous and long-term assessments are needed to determine if 3D-printed devices are clinically relevant before they become part of standard clinical practice.

Towards personalised management of atherosclerosis via computational models in vascular clinics: technology based on patient-specific simulation approach

The development of a new technology based on patient-specific modelling for personalised healthcare in the case of atherosclerosis is presented. Atherosclerosis is the main cause of death in the world and it has become a burden on clinical services as it manifests itself in many diverse forms, such as coronary artery disease, cerebrovascular disease/stroke and peripheral arterial disease. It is also a multifactorial, chronic and systemic process that lasts for a lifetime, putting enormous financial and clinical pressure on national health systems. In this Letter, the postulate is that the development of new technologies for healthcare using computer simulations can, in the future, be developed as in-silico management and support systems. These new technologies will be based on predictive models (including the integration of observations, theories and predictions across a range of temporal and spatial scales, scientific disciplines, key risk factors and anatomical sub-systems) combined with digital patient data and visualisation tools. Although the problem is extremely complex, a simulation workflow and an exemplar application of this type of technology for clinical use is presented, which is currently being developed by a multidisciplinary team following the requirements and constraints of the Vascular Service Unit at the University College Hospital, London.

Public health in the genomic era: will Public Health Genomics contribute to major changes in the prevention of common diseases?

The completion of the Human Genome Project triggered a whole new field of genomic research which is likely to lead to new opportunities for the promotion of population health. As a result, the distinction between genetic and environmental diseases has faded. Presently, genomics and knowledge deriving from systems biology, epigenomics, integrative genomics or genome-environmental interactions give a better insight on the pathophysiology of common diseases. However, it is barely used in the prevention and management of diseases. Together with the boost in the amount of genetic association studies, this demands for appropriate public health actions. The field of Public Health Genomics analyses how genome-based knowledge and technologies can responsibly and effectively be integrated into health services and public policy for the benefit of population health. Environmental exposures interact with the genome to produce health information which may help explain inter-individual differences in health, or disease risk. However today, prospects for concrete applications remain distant. In addition, this information has not been translated into health practice yet. Therefore, evidence-based recommendations are few. The lack of population-based research hampers the evaluation of the impact of genomic applications. Public Health Genomics also evaluates the benefits and risks on a larger scale, including normative, legal, economic and social issues. These new developments are likely to affect all domains of public health and require rethinking the role of genomics in every condition of public health interest. This article aims at providing an introduction to the field of and the ideas behind Public Health Genomics.