Resident Salary Compared to Living Wages at United States Training Institutions.

OBJECTIVE: To compare living wages and salaries at US residency programs. SUMMARY BACKGROUND DATA: It is unknown how resident salary compares to living wages across the United States (US). METHODS: Cross-sectional analysis of publicly available resident salary affordability from training centers with post-graduate-year (PGY)-1 through PGY-7 resident compensation for 2022-2023 was compared with the Massachusetts Institute of Technology (MIT) Living-Wage Calculator. Resident salary to living wage ratios were calculated using PGY-4 salary for each family composition. Univariate and multivariable analysis of PGY-4 salary affordability was performed, accounting for proportion of expected living wages to taxes, transportation, housing, healthcare, childcare, and food, as well as unionization and state income-tax. RESULTS: 118 residency programs, representing over 60% of US trainees, were included, 20 (17%) of which were unionized. Single-parent families were unable to earn a living wage until PGY-7. Residents with 1 child in 2-adult (single-income) and 2-adult (dual-income) families earn below living wages until PGY-5 and PGY-3, respectively. Residents with more than 1 child never earn a living wage. Multivariable regression analysis using PGY-4 salary: living wage ratios in single-child, 2-parent homes showed food expense and unionization status were consistent predictors of affordability. Unionization was associated with lower affordability pre-stipend, almost equivalent affordability post-stipend, and lower affordability post-stipend and union dues. CONCLUSIONS: Resident salaries often preclude residents with children from earning a living wage. Unionization is not associated with increased resident affordability in this cross-sectional analysis. All annual reimbursement data should be centrally compiled, and additional stipends should be considered for residents with children.

Incorporation of ChatGPT and Other Large Language Models into a Graduate Level Computational Bioengineering Course.

The remarkable capabilities of generative artificial intelligence and large language models (LLMs) such as ChatGPT have delighted users around the world. Educators have regarded these tools as either a cause for great concern, an opportunity to educate students on cutting-edge technology, or often some combination of the two. Throughout the Fall 2023 semester, we explored the use of ChatGPT (and Bard, among other LLMs) in a graduate level numerical and statistical methods course for PhD-level bioengineers. In this article we share examples of this ChatGPT content, our observations on what worked best in our course, and speculate on how bioengineering students may be best served by this technology in the future.

Biomimetic proteoglycans as a tool to engineer the structure and mechanics of porcine bioprosthetic heart valves.

The utility of bioprosthetic heart valves (BHVs) is limited to certain patient populations because of their poor durability compared to mechanical prosthetic valves. Histological analysis of failed porcine BHVs suggests that degeneration of the tissue extracellular matrix (ECM), specifically the loss of proteoglycans and their glycosaminoglycans (GAGs), may lead to impaired mechanical performance, resulting in nucleation and propagation of tears and ultimately failure of the prosthetic. Several strategies have been proposed to address this deterioration, including novel chemical fixatives to stabilize ECM constituents and incorporation of small molecule inhibitors of catabolic enzymes implicated in the degeneration of the BHV ECM. Here, biomimetic proteoglycans (BPGs) were introduced into porcine aortic valves ex vivo and were shown to distribute throughout the valve leaflets. Incorporation of BPGs into the heart valve leaflet increased tissue overall GAG content. The presence of BPGs also significantly increased the micromodulus of the spongiosa layer within the BHV without compromising the chemical fixation process used to sterilize and strengthen the tissue prior to implantation. These findings suggest that a targeted approach for molecularly engineering valve leaflet ECM through the use of BPGs may be a viable way to improve the mechanical behavior and potential durability of BHVs.

MC3T3 E1 cell response to mineralized nanofiber shish kebab structures.

Block copolymers (BCPs) are of growing interest because of their extensive utility in tissue engineering, particularly in biomimetic approaches where multifunctionality is critical. We synthesized polycaprolactone-polyacrylic acid (PCL-b-PAA) BCP and crystallized it onto PCL nanofibers, making BCP nanofiber shish kebab (BCP NFSK) structures. When mineralized in 2× simulated body fluid, BCP NFSK mimic the structure of mineralized collagen fibrils. We hypothesized that the addition of a calcium phosphate layer of graded roughness on the nano-structure of the nanofiber shish kebabs would enhance preosteoblast alkaline phosphatase (ALP) activity, which has been shown to be a critical component in bone matrix formation. The objectives in the study were to investigate the effect of mineralization on cell proliferation and ALP activity, and to also investigate the effect of BCP NFSK periodicity, a structural feature describing the distance between PCL-b-PAA crystals on the nanofiber core, on cell proliferation, and ALP activity. ALP activity of cells cultured on the mineralized BCP NFSK template was significantly higher than the nonmineralized BCP NFSK templates. Interestingly, no statistical difference was observed in ALP activity when the periodic varied, indicating that surface chemistry seemed to play a larger role than the surface roughness.