Development and preliminary testing of Health Equity Across the AI Lifecycle (HEAAL): A framework for healthcare delivery organizations to mitigate the risk of AI solutions worsening health inequities.

The use of data-driven technologies such as Artificial Intelligence (AI) and Machine Learning (ML) is growing in healthcare. However, the proliferation of healthcare AI tools has outpaced regulatory frameworks, accountability measures, and governance standards to ensure safe, effective, and equitable use. To address these gaps and tackle a common challenge faced by healthcare delivery organizations, a case-based workshop was organized, and a framework was developed to evaluate the potential impact of implementing an AI solution on health equity. The Health Equity Across the AI Lifecycle (HEAAL) is co-designed with extensive engagement of clinical, operational, technical, and regulatory leaders across healthcare delivery organizations and ecosystem partners in the US. It assesses 5 equity assessment domains-accountability, fairness, fitness for purpose, reliability and validity, and transparency-across the span of eight key decision points in the AI adoption lifecycle. It is a process-oriented framework containing 37 step-by-step procedures for evaluating an existing AI solution and 34 procedures for evaluating a new AI solution in total. Within each procedure, it identifies relevant key stakeholders and data sources used to conduct the procedure. HEAAL guides how healthcare delivery organizations may mitigate the potential risk of AI solutions worsening health inequities. It also informs how much resources and support are required to assess the potential impact of AI solutions on health inequities.

Non-task expert physicians benefit from correct explainable AI advice when reviewing X-rays.

Artificial intelligence (AI)-generated clinical advice is becoming more prevalent in healthcare. However, the impact of AI-generated advice on physicians' decision-making is underexplored. In this study, physicians received X-rays with correct diagnostic advice and were asked to make a diagnosis, rate the advice's quality, and judge their own confidence. We manipulated whether the advice came with or without a visual annotation on the X-rays, and whether it was labeled as coming from an AI or a human radiologist. Overall, receiving annotated advice from an AI resulted in the highest diagnostic accuracy. Physicians rated the quality of AI advice higher than human advice. We did not find a strong effect of either manipulation on participants' confidence. The magnitude of the effects varied between task experts and non-task experts, with the latter benefiting considerably from correct explainable AI advice. These findings raise important considerations for the deployment of diagnostic advice in healthcare.

Feminicide and counterdata production: Activist efforts to monitor and challenge gender-related violence.

Gender-related violence against women and its lethal outcome, feminicide, are a serious problem throughout the world. Official government data on gender violence and feminicide are often absent, incomplete, infrequently updated, and contested. We draw on data feminism to situate feminicide data as missing data. Building on qualitative interviews, this study discusses the informatic work of ten activist and civil society organizations across six countries who combat missing data by producing counterdata. Activists enact alternative epistemological approaches to data science that center care, memory, and justice. Activists also face significant information challenges that increase monitoring labor and add emotional burden to reading about violent deaths. This work contributes to literature on data activism and critical data studies, proposing feminicide data practices as an important research subject. The empirical insights contribute to human-computer interaction (HCI) research, suggesting ways that the field may support and sustain the counterdata production practices of activists.

Do as AI say: susceptibility in deployment of clinical decision-aids.

Artificial intelligence (AI) models for decision support have been developed for clinical settings such as radiology, but little work evaluates the potential impact of such systems. In this study, physicians received chest X-rays and diagnostic advice, some of which was inaccurate, and were asked to evaluate advice quality and make diagnoses. All advice was generated by human experts, but some was labeled as coming from an AI system. As a group, radiologists rated advice as lower quality when it appeared to come from an AI system; physicians with less task-expertise did not. Diagnostic accuracy was significantly worse when participants received inaccurate advice, regardless of the purported source. This work raises important considerations for how advice, AI and non-AI, should be deployed in clinical environments.

Pediatric Endoscopic Ossiculoplasty Following Surgery for Chronic Ear Disease.

OBJECTIVE: Rigid endoscopes can improve visualization of the tympanic space compared to traditional microscopic techniques. This study investigates whether use of transcanal endoscopic ossiculoplasty influences audiologic outcomes compared to microscopic ossiculoplasty following chronic ear surgery in children. STUDY DESIGN: Comparative cohort study at two tertiary care centers. METHODS: Retrospective review of pediatric chronic ear cases where ossiculoplasty was performed from February 2009 to March 2018. RESULTS: We identified 100 ears that underwent endoscopic ossiculoplasty and 100 ears that underwent microscopic ossiculoplasty. The mean age was 11 years (range, 4-18 years) with 63% males. There were no significant differences in these parameters between the two groups. Subjects underwent either primary ossiculoplasty or ossiculoplasty during second-look procedures. There was no significant difference in air conduction pure tone average (PTA) after microscopic cases compared to endoscopic cases (-12.5 dB vs. -10.5 dB, P = .40). These results were independent of prosthesis type. Microscopic ossiculoplasty was significantly more likely to use a post-auricular approach (P = .0001). There was no difference in complication rate between the two groups. The malleus was more likely to be absent or removed prior to endoscopic ossiculoplasty (P = .0004) with no significant difference in the change in PTA between groups. CONCLUSIONS: Transcanal endoscopic ossiculoplasty was found to have equivalent audiometric outcomes with significantly fewer post-auricular approaches and no increase in complications compared to microscopic ossiculoplasty. While the malleus was more likely to be absent in endoscopic cases, this did not appear to influence the change in PTA. LEVEL OF EVIDENCE: 4 Laryngoscope, 2020.

Use of machine-learning algorithms to determine features of systolic blood pressure variability that predict poor outcomes in hypertensive patients.

BACKGROUND: We re-analyzed data from the Systolic Blood Pressure Intervention Trial (SPRINT) trial to identify features of systolic blood pressure (SBP) variability that portend poor cardiovascular outcomes using a nonlinear machine-learning algorithm. METHODS: We included all patients who completed 1 year of the study without reaching any primary endpoint during the first year, specifically: myocardial infarction, other acute coronary syndromes, stroke, heart failure or death from a cardiovascular event (n = 8799; 94%). In addition to clinical variables, features representing longitudinal SBP trends and variability were determined and combined in a random forest algorithm, optimized using cross-validation, using 70% of patients in the training set. Area under the curve (AUC) was measured using a 30% testing set. Finally, feature importance was determined by minimizing node impurity averaging over all trees in the forest for a specific feature. RESULTS: A total of 365 patients (4.1%) reached the combined primary outcome over 37 months of follow-up. The random forest classifier had an AUC of 0.71 on the testing set. The 10 most significant features selected in order of importance by the automated algorithm included the urine albumin/creatinine (CR) ratio, estimated glomerular filtration rate, age, serum CR, history of subclinical cardiovascular disease (CVD), cholesterol, a variable representing SBP signals using wavelet transformation, high-density lipoprotein, the 90th percentile of SBP and triglyceride level. CONCLUSIONS: We successfully demonstrated use of random forest algorithm to define best prognostic longitudinal SBP representations. In addition to known risk factors for CVD, transformed variables for time series SBP measurements were found to be important in predicting poor cardiovascular outcomes and require further evaluation.

Orientation-specific attachment of polymeric microtubes on cell surfaces.

Tubular particles presenting heterogeneous regions of chemistry on the tube-ends versus the side are fabricated and are shown to control the particle orientation on the surface of live lymphocytes. Controlling the orientation of anisotropic microparticles on cell surfaces is of interest for biomedical applications and drug delivery in particular, since it can be used to promote or resist particle internalization.