Objectives: Throughout history, societies have been impacted by inequality. Many studies have been conducted on the topic more broadly, but only a few have investigated inequalities in out-of-pocket health payments (OHP). This study measures OHP inequality trends among the Iranian households.Methods: This study used data from the Iranian Statistics Center on Iranian household income and expenditures. The analysis included a total of 995 300 households during the 36 years from 1984 to 2019. The Gini coefficient, Atkinson index, and Theil index were calculated for Iranian OHP.Results: Average Iranian household OHP increased from 33 US dollar (USD) in 1984 to 47 USD in 2019. During this 36-year span, the average±standard deviation Gini coefficient for OHP was 0.73±0.04, and the Atkinson and Theil indexes were 0.68±0.05 and 1.14±0.29, respectively. The Gini coefficients for the subcategories of OHP of outpatient diagnostic services, medical assistant accessories, hospital inpatient services, and addiction cessation were 0.70, 0.61, 0.84, and 0.64, respectively.Conclusions: In this study, we scrutinized trends of inequality in the OHP of Iranian households. Inequality in OHP decreased slightly over the past four decades. An analysis of trends among different subgroups revealed that affluent households, such as households with insurance coverage and households in higher income deciles, experienced higher inequality. Therefore, lower inequality in health care expenditures may be related to restricted access to health care services in Iran.

<div>State-of-the-art Heterogeneous System on Chips (HMPSoCs) can perform on-chip embedded inference on its CPU and GPU. Multi-component pipelining is the method of choice to provide high-throughput Convolutions Neural Network (CNN) inference on embedded platforms. In this work, we provide details for the first CPU-GPU pipeline design for CNN inference called Pipe-All. Pipe-All uses the ARM-CL library to integrate an ARM big.Little CPU with an ARM Mali GPU. Pipe-All is the first three-stage CNN inference pipeline design with ARM’s big CPU cluster, Little CPU cluster, and Mali GPU as its stages. Pipe-All provides on average 75.88% improvement in inference throughput (over peak single-component inference) on Amlogic A311D HMPSoC in Khadas Vim 3 embedded platform. We also provide an open-source implementation for Pipe-All.</div><div>This paper is submitted to IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD) as a transaction brief paper (5 pages).</div>

As health care needs are unequally distributed among people in society and various factors influence this inequality, it is expected that health expenditures (unlike other types of expenditures such as food or education expenditures) should be unequally distributed among individuals. In this case, it can be ensured that people have sufficiently benefited from health services.

<div>State-of-the-art Heterogeneous System on Chips (HMPSoCs) can perform on-chip embedded inference on its CPU and GPU. Multi-component pipelining is the method of choice to provide high-throughput Convolutions Neural Network (CNN) inference on embedded platforms. In this work, we provide details for the first CPU-GPU pipeline design for CNN inference called Pipe-All. Pipe-All uses the ARM-CL library to integrate an ARM big.Little CPU with an ARM Mali GPU. Pipe-All is the first three-stage CNN inference pipeline design with ARM’s big CPU cluster, Little CPU cluster, and Mali GPU as its stages. Pipe-All provides on average 75.88% improvement in inference throughput (over peak single-component inference) on Amlogic A311D HMPSoC in Khadas Vim 3 embedded platform. We also provide an open-source implementation for Pipe-All.</div><div>This paper is submitted to IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD) as a transaction brief paper (5 pages).</div>

No abstract