Health equity assessment of machine learning performance (HEAL): a framework and dermatology AI model case study.

Background: Artificial intelligence (AI) has repeatedly been shown to encode historical inequities in healthcare. We aimed to develop a framework to quantitatively assess the performance equity of health AI technologies and to illustrate its utility via a case study. Methods: Here, we propose a methodology to assess whether health AI technologies prioritise performance for patient populations experiencing worse outcomes, that is complementary to existing fairness metrics. We developed the Health Equity Assessment of machine Learning performance (HEAL) framework designed to quantitatively assess the performance equity of health AI technologies via a four-step interdisciplinary process to understand and quantify domain-specific criteria, and the resulting HEAL metric. As an illustrative case study (analysis conducted between October 2022 and January 2023), we applied the HEAL framework to a dermatology AI model. A set of 5420 teledermatology cases (store-and-forward cases from patients of 20 years or older, submitted from primary care providers in the USA and skin cancer clinics in Australia), enriched for diversity in age, sex and race/ethnicity, was used to retrospectively evaluate the AI model's HEAL metric, defined as the likelihood that the AI model performs better for subpopulations with worse average health outcomes as compared to others. The likelihood that AI performance was anticorrelated to pre-existing health outcomes was estimated using bootstrap methods as the probability that the negated Spearman's rank correlation coefficient (i.e., "R") was greater than zero. Positive values of R suggest that subpopulations with poorer health outcomes have better AI model performance. Thus, the HEAL metric, defined as p (R >0), measures how likely the AI technology is to prioritise performance for subpopulations with worse average health outcomes as compared to others (presented as a percentage below). Health outcomes were quantified as disability-adjusted life years (DALYs) when grouping by sex and age, and years of life lost (YLLs) when grouping by race/ethnicity. AI performance was measured as top-3 agreement with the reference diagnosis from a panel of 3 dermatologists per case. Findings: Across all dermatologic conditions, the HEAL metric was 80.5% for prioritizing AI performance of racial/ethnic subpopulations based on YLLs, and 92.1% and 0.0% respectively for prioritizing AI performance of sex and age subpopulations based on DALYs. Certain dermatologic conditions were significantly associated with greater AI model performance compared to a reference category of less common conditions. For skin cancer conditions, the HEAL metric was 73.8% for prioritizing AI performance of age subpopulations based on DALYs. Interpretation: Analysis using the proposed HEAL framework showed that the dermatology AI model prioritised performance for race/ethnicity, sex (all conditions) and age (cancer conditions) subpopulations with respect to pre-existing health disparities. More work is needed to investigate ways of promoting equitable AI performance across age for non-cancer conditions and to better understand how AI models can contribute towards improving equity in health outcomes. Funding: Google LLC.

Continuous hemoadsorption with a cytokine adsorber during sepsis - a review of the literature.

Sepsis is a well-recognized healthcare issue worldwide, ultimately resulting in significant mortality, morbidity and resource utilization during and after critical illness. In its most severe form, sepsis causes multi-organ dysfunction that produces a state of critical illness characterized by severe immune dysfunction and catabolism. Sepsis induces the activation of complement factor via 3 pathways and the release of inflammatory cytokines such as tumor necrosis factor alpha (TNF-α) and interleukin-1beta (IL-1ß), resulting in a systemic inflammatory response. The inflammatory cytokines and nitric oxide release induced by sepsis decrease systemic vascular resistance, resulting in profound hypotension. The combination of hypotension and microvascular occlusion results in tissue ischemia and ultimately leads to multiple organ failure. Several clinical and experimental studies have reported that treatment using adsorption of cytokines is beneficial during endotoxemia and sepsis. This review article analyzes the efficacy of CytoSorb® adsorber in reducing the inflammatory response during sepsis. The CytoSorb® adsorber is known to have excellent adsorption rates for inflammatory cytokines such as IL-1ß, IL-6, IL-8, IL-10, and TNF-α. Studies have demonstrated that treatment with cytokine adsorbing columns has beneficial effects on the survival rate and inflammatory responses in animal septic models. Additionally, several cases have been reported in which treatment with cytokine adsorbing columns is very effective in hemodynamic stabilization and in preventing organ failure in critically ill patients. Although further investigations and clinical trials are needed, treatment with cytokine adsorbing columns may play an important role in the treatment of sepsis in the near future.

Activating Hair Follicle Stem Cells via R-spondin2 to Stimulate Hair Growth.

Wnt signaling is required for the development of the hair follicle, and for inciting the growth (anagen) phase of the hair cycle. Most strategies to enhance Wnt signaling for hair growth create a state of constitutive Wnt activation, which leads to neoplastic transformation of the epithelial hair matrix. Using Axin2(LacZ/+) and Axin2(Cre/+)R26R(mTmG/+) reporter mice and RNA analyses, we show that Wnt signaling is elevated during anagen, is reduced at the onset of catagen, and can be reamplified in the skin and surrounding hair follicles via intradermal injection of recombinant R-spondin2 protein. Using Lgr5(LacZ/+) reporter mice, we demonstrate that this amplified Wnt environment leads to activation of leucine-rich repeat-containing G-protein coupled receptor 5-positive stem cells in the hair follicle. The onset of catagen is repressed by R-spondin2 injection, and the anagen phase persists. As a consequence, hair shafts grow longer. We conclude that R-spondin2 treatment activates hair follicle stem cells and therefore may have therapeutic potential to promote hair growth.

Wnt signaling induces epithelial differentiation during cutaneous wound healing.

Cutaneous wound repair in adult mammals typically does not regenerate original dermal architecture. Skin that has undergone repair following injury is not identical to intact uninjured skin. This disparity may be caused by differences in the mechanisms that regulate postnatal cutaneous wound repair compared to embryonic skin development and thus we seek a deeper understanding of the role that Wnt signaling plays in the mechanisms of skin repair in both fetal and adult wounds. The influence of secreted Wnt signaling proteins in tissue homeostasis has galvanized efforts to identify small molecules that target Wnt-mediated cellular responses. Wnt signaling is activated by wounding and participates in every subsequent stage of the healing process from the control of inflammation and programmed cell death, to the mobilization of stem cell reservoirs within the wound site. Endogenous Wnt signaling augmentation represents an attractive option to aid in the restoration of cutaneous wounds, as the complex mechanisms of the Wnt pathway have been increasingly investigated over the years. In this review, we summarize recent data elucidating the roles that Wnt signaling plays in cutaneous wound healing process.

Medical leech therapy in plastic reconstructive surgery.

The use of Hirudo medicinalis in clinical practice has increased in recent years. The primary indication in plastic surgery has traditionally been venous congestion. However, other reported clinical applications were in varicose veins, thrombophlebitis, and osteoarthritis. In this review, we summarize recent data elucidating the role that medicinal leeches play in the field of plastic surgery.