Minimum labelling requirements for dermatology artificial intelligence-based Software as Medical Device (SaMD): A consensus statement.

BACKGROUND/OBJECTIVES: Artificial intelligence (AI) holds remarkable potential to improve care delivery in dermatology. End users (health professionals and general public) of AI-based Software as Medical Devices (SaMD) require relevant labelling information to ensure that these devices can be used appropriately. Currently, there are no clear minimum labelling requirements for dermatology AI-based SaMDs. METHODS: Common labelling recommendations for AI-based SaMD identified in a recent literature review were evaluated by an Australian expert panel in digital health and dermatology via a modified Delphi consensus process. A nine-point Likert scale was used to indicate importance of 10 items, and voting was conducted to determine the specific characteristics to include for some items. Consensus was achieved when more than 75% of the experts agreed that inclusion of information was necessary. RESULTS: There was robust consensus supporting inclusion of all proposed items as minimum labelling requirements; indication for use, intended user, training and test data sets, algorithm design, image processing techniques, clinical validation, performance metrics, limitations, updates and adverse events. Nearly all suggested characteristics of the labelling items received endorsement, except for some characteristics related to performance metrics. Moreover, there was consensus that uniform labelling criteria should apply across all AI categories and risk classes set out by the Therapeutic Goods Administration. CONCLUSIONS: This study provides critical evidence for setting labelling standards by the Therapeutic Goods Administration to safeguard patients, health professionals, consumers, industry, and regulatory bodies from AI-based dermatology SaMDs that do not currently provide adequate information about how they were developed and tested.

Informing a position statement on the use of artificial intelligence in dermatology in Australia.

Artificial Intelligence (AI) is the ability for computers to simulate human intelligence. In dermatology, there is substantial interest in using AI to identify skin lesions from images. Due to increasing research and interest in the use of AI, the Australasian College of Dermatologists has developed a position statement to inform its members of appropriate use of AI. This article presents the ACD Position Statement on the use of AI in dermatology, and provides explanatory information that was used to inform the development of this statement.

Global Globin Network Consensus Paper: Classification and Stratified Roadmaps for Improved Thalassaemia Care and Prevention in 32 Countries.

The Global Globin Network (GGN) is a project-wide initiative of the Human Variome/Global Variome Project (HVP) focusing on haemoglobinopathies to build the capacity for genomic diagnosis, clinical services, and research in low- and middle-income countries. At present, there is no framework to evaluate the improvement of care, treatment, and prevention of thalassaemia and other haemoglobinopathies globally, despite thalassaemia being one of the most common monogenic diseases worldwide. Here, we propose a universally applicable system for evaluating and grouping countries based on qualitative indicators according to the quality of care, treatment, and prevention of haemoglobinopathies. We also apply this system to GGN countries as proof of principle. To this end, qualitative indicators were extracted from the IthaMaps database of the ITHANET portal, which allowed four groups of countries (A, B, C, and D) to be defined based on major qualitative indicators, supported by minor qualitative indicators for countries with limited resource settings and by the overall haemoglobinopathy carrier frequency for the target countries of immigration. The proposed rubrics and accumulative scores will help analyse the performance and improvement of care, treatment, and prevention of haemoglobinopathies in the GGN and beyond. Our proposed criteria complement future data collection from GGN countries to help monitor the quality of services for haemoglobinopathies, provide ongoing estimates for services and epidemiology in GGN countries, and note the contribution of the GGN to a local and global reduction of disease burden.

Implementation of patient-reported outcome measures and patient-reported experience measures in melanoma clinical quality registries: a systematic review.

OBJECTIVES: To identify patient-reported outcome measures (PROMs) and patient-reported experience measures (PREMs) in clinical quality registries, for people with cutaneous melanoma, to inform a new Australian Melanoma Clinical Outcomes Registry; and describe opportunities and challenges of routine PROM/PREM collection, especially in primary care. DESIGN: Systematic review. PRIMARY AND SECONDARY OUTCOME MEASURES: Which PROMs and PREMs are used in clinical quality registries for people with cutaneous melanoma, how they are collected, frequency of collection, participant recruitment methods and funding models for each registry. RESULTS: 1134 studies were identified from MEDLINE, PreMEDLINE, Embase, PsychInfo, Cochrane Database of Abstracts of Reviews of Effects databases and TUFTS Cost-Effectiveness Analysis Registry, alongside grey literature, from database inception to 5th February 2020. Following screening, 14 studies were included, identifying four relevant registries: Dutch Melanoma Registry, Adelphi Real-World Disease-Specific Programme (Melanoma), Patient-Reported Outcomes Following Initial treatment and Long-term Evaluation of Survivorship Registry, and Cancer Experience Registry. These used seven PROMs: EuroQol-5 Dimensions, Functional Assessment of Cancer-General (FACT-G) and FACT-Melanoma (FACT-M), European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-Cancer 30 (EORTC QLQ-C30), Fatigue Assessment Scale Hospital Anxiety and Depression Scale, Patient-Reported Outcome Measures Information System-29 and one PREM; EORTC QLQ-Information Module 26. PROMs/PREMs in registries were reported to improve transparency of care; facilitate clinical auditing for quality assessment; enable cost-effectiveness analyses and create large-scale research platforms. Challenges included resource burden for data entry and potential collection bias toward younger, more affluent respondents. Feedback from patients with melanoma highlighted the relevance of PROMs/PREMs in assessing patient outcomes and patient experiences. CONCLUSIONS: Clinical registries indicate PROMs/PREMs for melanoma care can be incorporated and address important gaps, however cost and collection bias may limit generalisability. PROSPERO REGISTRATION NUMBER: CRD42018086737.

Practice guidelines for teledermatology in Australia.

Despite the potential of teledermatology to increase access to dermatology services and improve patient care, it is not widely practised in Australia. In an effort to increase uptake of teledermatology by Australian dermatologists and support best practice, guidelines for teledermatology for the Australian context have been developed by The University of Queensland's Centre for Online Health in collaboration with The Australasian College of Dermatologists' E-Health Committee. The guidelines are presented in two sections: 1. Guidelines and 2. Notes to support their application in practice, when feasible and appropriate. Content was last updated March 2020 and includes modalities of teledermatology; patient selection and consent; imaging; quality and safety; privacy and security; communication; and documentation and retention of clinical images. The guidelines educate dermatologists about the benefits and limitations of telehealth while articulating how to enhance patient care and reduce risk when practicing teledermatology.

A review of literature supporting the development of practice guidelines for teledermatology in Australia.

Despite the potential of teledermatology to increase access to dermatology services and improve patient care, it is not widely practised in Australia. In an effort to increase uptake of teledermatology, Australian-specific practice guidelines for teledermatology are being developed by the Australasian College of Dermatologist. This paper reports finding from literature reviews that were undertaken to inform the development of these guidelines. Results cover the following sections: Modalities of teledermatology; Patient selection and consent; Imaging; Quality and safety; Privacy and security; Communication; and Documentation and retention. The document educates providers about the benefits and limitations of telehealth while articulating how to enhance patient care and reduce risk when practicing teledermatology.

Aluminum Exposure from Parenteral Nutrition: Early Bile Canaliculus Changes of the Hepatocyte.

Background: Neonates on long-term parenteral nutrition (PN) may develop parenteral nutrition-associated liver disease (PNALD). Aluminum (Al) is a known contaminant of infant PN, and we hypothesize that it substantially contributes to PNALD. In this study, we aim to assess the impact of Al on hepatocytes in a piglet model. Methods: We conducted a randomized control trial using a Yucatan piglet PN model. Piglets, aged 3⁻6 days, were placed into two groups. The high Al group (n = 8) received PN with 63 µg/kg/day of Al, while the low Al group (n = 7) received PN with 24 µg/kg/day of Al. Serum samples for total bile acids (TBA) were collected over two weeks, and liver tissue was obtained at the end of the experiment. Bile canaliculus morphometry were studied by transmission electron microscopy (TEM) and ImageJ software analysis. Results: The canalicular space was smaller and the microvilli were shorter in the high Al group than in the low Al group. There was no difference in the TBA between the groups. Conclusions: Al causes structural changes in the hepatocytes despite unaltered serum bile acids. High Al in PN is associated with short microvilli, which could decrease the functional excretion area of the hepatocytes and impair bile flow.

Infant Parenteral Nutrition Remains a Significant Source for Aluminum Toxicity.

BACKGROUND: Aluminum toxicity is associated with anemia, impaired bone metabolism, neurologic defects, and parenteral nutrition (PN)-associated liver disease. This element is a ubiquitous contaminant of PN components, especially in infant formulations. We assessed the current levels of aluminum contamination in infant PN at a level III neonatal intensive care unit. MATERIALS AND METHODS: Thirty samples of PN prepared in the same hospital for infants aged <30 days (mean [SD] weight, 1.54 [0.71] kg) were collected from discarded solution. Each sample was analyzed for aluminum content via inductively coupled plasma mass spectrometry. The components of PN (from label) and measured aluminum content were then compared using linear regression and 1-way analysis of variance. RESULTS: The mean (SD) aluminum contamination of infant PN was 14.02 (6.51) mcg/kg/d. Only 3 samples were <5 mcg/kg/d. Aluminum levels and infant weight were not associated. Linear regression revealed a significant correlation between aluminum and both calcium gluconate ( P < .0001) and phosphate ( P = .05), with a trend between aluminum and potassium ( P = .07). CONCLUSIONS: Aluminum contamination in infant PN remains almost 3 times higher than the advised maximum exposure (<5 mcg/kg/d, Food and Drug Administration 2004). Unexpectedly, an association between infant weight and aluminum exposure was not apparent, likely due to the homogeneity of our population. Isolating the source of aluminum contamination is difficult, as multiple components appear to be involved. Calcium gluconate is likely still a major contributor, but further investigations into individual components are warranted to promote the reduction of aluminum in infant PN.

Influence of nutrition provision during the first two weeks of life in premature infants on adolescent body composition and blood pressure.

OBJECTIVE: Adequate nutrition is paramount for premature infants. Longitudinal information is scant on the effects of early nutrition and later growth. The purpose of this study was to determine the influence of early energy and protein provision in premature infants on adolescent body composition and blood pressure. METHODS: In 2007-2008 we obtained data from 36 male (12.3±1.7 years) and 25 female (11.5±1.8 years) adolescents born preterm at <34 weeks gestation (range 23-34 weeks) between October 1st 1989 and December 31st 1995 (birth weight <1850 g). The adolescents were divided into groups depending on infant intake mode (enteral vs parenteral), energy provision (<70 kcal/kg/d and ≥70 kcal/kg/d) and protein provision (>2.5 g/kg/d for ≥5 days and >2.5 g/kg/d for <5 days) during the first 14 days of life. RESULTS: After controlling for birth weight and biological maturity, adolescents who received ≥70 kcal/kg/d during infancy were significantly taller (163±11 cm vs. 156±11 cm) and heavier (58±16 kg vs. 49±16 kg) than adolescents who received <70 kcal/kg/d. There were no significant differences in systolic and diastolic BP and total percent body fat between the two groups. CONCLUSIONS: Our data suggests that higher infant energy provision appears to be related to adolescent size, it does not appear to contribute to adverse risk factors such as higher systolic BP or increased body fat.

Reduced aluminum contamination decreases parenteral nutrition associated liver injury.

PURPOSE: Parenteral nutrition-associated cholestasis remains a significant problem, especially for the surgical neonates. Aluminum is a toxic element known to contaminate parenteral nutrition. We hypothesize that parenterally administered aluminum causes liver injury similar to that seen in parenteral nutrition-associated cholestasis. METHODS: Twenty 3- to 6-day-old domestic pigs were divided into 5 equal groups. A control group received daily intravenous 0.9% NaCl. Each subject in experimental groups received intravenous aluminum chloride at 1500 µg kg(-1) d(-1) for 1, 2, 3, or 4 weeks. At the end of the study, blood was sampled for direct bilirubin and total bile acid levels. Liver, bile, and urine were sampled for aluminum content. Liver tissue was imaged by transmission electron microscopy for ultrastructural changes. RESULTS: Transmission electron microscopy revealed marked blunting of bile canaliculi microvilli in all experimental subjects but not the controls. Serum total bile acids correlated with the duration of aluminum exposure. The hepatic aluminum concentration correlated with the duration of aluminum exposure. CONCLUSIONS: Parenterally infused aluminum resulted in liver injury as demonstrated by elevated bile acids and by blunting of the bile canaliculi microvilli. These findings are similar to those reported in early parenteral nutrition-associated liver disease.

Parenteral aluminum induces liver injury in a newborn piglet model.

PURPOSE: Parenteral nutrition-associated cholestasis remains a significant problem, especially for the surgical neonate. Aluminum is a toxic element known to contaminate parenteral nutrition. We hypothesize that parenterally administered aluminum causes liver injury similar to that seen in parenteral nutrition-associated cholestasis. METHODS: Twenty 3- to 6-day-old domestic pigs were divided into 5 equal groups. A control group received daily intravenous 0.9% sodium chloride. Each subject in experimental groups received intravenous aluminum chloride at 1500 µg/kg per day for 1, 2, 3, or 4 weeks. At the end of the study, blood was sampled for direct bilirubin and total bile acid levels. Liver, bile, and urine were sampled for aluminum content. Liver tissue was imaged by transmission electron microscopy for ultrastructural changes. RESULTS: Transmission electron microscopy revealed marked blunting of bile canaliculi microvilli in all experimental subjects but not the controls. Serum total bile acids correlated with the duration of aluminum exposure. The hepatic aluminum concentration correlated with the duration of aluminum exposure. CONCLUSIONS: Parenterally infused aluminum resulted in liver injury as demonstrated by elevated bile acids and by blunting of the bile canaliculi microvilli. These findings are similar to those reported in early parenteral nutrition-associated liver disease.

Preterm birth and adolescent bone mineral content.

The purpose of this study was to determine the influence of preterm low birth weight on bone mineral content in adolescence. In 2007 to 2008, data on adolescents were obtained for study, including 16 females and 25 males who were born preterm (≤37 weeks' gestation) between October 1, 1989, and December 31, 1995, with a birth weight of less than 1850 g. Preterm low-birth-weight individuals were age- and sex-matched to full-term (>37 weeks) normal-birth-weight (>2500 g) controls. Total body, hip, and spine bone mineral content (BMC) was assessed using dual energy X-ray absorptiometry. Male preterm individuals had less BMC at the proximal femur in adolescence compared with controls ( p < 0.05). However, once adjusted for age, maturity, height, weight, physical activity, and diet, there were no differences between groups ( p < 0.05) in any bone parameters. These findings suggest that preterm birth and low birth weight did not influence bone accrual in these individuals at adolescence.

Effects of thickened beverages fortified with inulin on beverage acceptance, gastrointestinal function, and bone resorption in institutionalized adults.

OBJECTIVES: We wanted to develop thickened beverages that contain soluble fiber (inulin) with acceptable consistency, taste, and texture and to determine the effects of these beverages on bone resorption markers (to determine calcium retention), bowel frequency, and indicators of gastrointestinal function in institutionalized adults bound to wheelchairs. METHODS: A double-blind, 3-wk, cross-over study testing 13-g/d inulin-fortified versus isocaloric standard modified starch-thickened beverages was conducted in institutionalized adults who were bound to wheelchairs and had dysphagia or did not have dysphagia. Beverage acceptability, as assessed by discriminative and descriptive sensory testing, bowel frequency, fecal output, and laxative use, were determined by direct testing or by nursing charts. Bone resorption was measured by using the urinary excretion of fasting calcium and of cross-linked N-telopeptides of collagen. RESULTS: Sensory panelists were unable to detect a difference between beverages thickened with modified starch and those fortified with inulin. Few differences were found between the control and inulin-fortified beverages for sensory descriptors. No significant difference was found in frequency of bowel movements between treatments; however, weighted bowel movement frequency increased by 13% with inulin (P < 0.01), whereas enema and laxative administration decreased by 13% (P < 0.05). Bone resorption, as an indicator of calcium retention, remained unchanged. CONCLUSIONS: Inulin was incorporated into thickened beverages, with no decrease in acceptability; when consumed, perceived stool output increased in residents of long-term care facilities.

Parenteral nutrition-associated cholestasis in neonates: the role of aluminum.

Parenteral nutrition (PN) is an essential component in the care of premature and ill infants. The incidence of parenteral nutrition-associated cholestasis (PNAC) ranges from 7.4 to 84%. One substance in PN solutions that has been implicated in PNAC is aluminum. Aluminum loading in animals and humans causes hepatic accumulation and damage. The degree of aluminum contamination of PN solutions has decreased over time, but contamination still significantly exceeds levels that are safe for human neonates. Further study into the relationship between aluminum contamination in neonatal PN solutions and the development of PNAC is necessary.