ABSTRACT
With the advent of Large Language Models (LLMs) like ChatGPT, the integration of Generative Artificial Intelligence (GAI) into clinical medicine is becoming increasingly feasible. This study aimed to evaluate the ability of the freely available ChatGPT-3.5 to generate complex differential diagnoses, comparing its output to case records of the Massachusetts General Hospital published in the New England Journal of Medicine (NEJM). Forty case records were presented to ChatGPT-3.5, prompting it to provide a differential diagnosis and then narrow it down to the most likely diagnosis. The results indicated that the final diagnosis was included in ChatGPT-3.5's original differential list in 42.5% of the cases. After narrowing, ChatGPT correctly determined the final diagnosis in 27.5% of the cases, demonstrating a decrease in accuracy compared to previous studies using common chief complaints. These findings emphasize the necessity for further investigation into the capabilities and limitations of LLMs in clinical scenarios while highlighting the potential role of GAI as an augmented clinical opinion. Anticipating the growth and enhancement of GAI tools like ChatGPT, physicians and other healthcare workers will likely find increasing support in generating differential diagnoses. However, continued exploration and regulation are essential to ensure the safe and effective integration of GAI into healthcare practice. Future studies may seek to compare newer versions of ChatGPT or investigate patient outcomes with physicians integrating this GAI technology. Understanding and expanding GAI's capabilities, particularly in differential diagnosis, may foster innovation and provide additional resources, especially in underserved areas in the medical field.
ABSTRACT
Tissue engineering has been successful in reproducing human skin equivalents while incorporating new approaches such as three-dimensional (3D) bioprinting. The latter method offers a plethora of advantages including increased production scale, ability to incorporate multiple cell types and printing on demand. However, the quality of printed skin equivalents compared to those developed manually has never been assessed. To leverage the benefits of this method, it is imperative that 3D-printed skin should be structurally and functionally similar to real human skin. Here, we developed four bilayered human skin epidermal-dermal equivalents: non-printed dermis and epidermis (NN), printed dermis and epidermis (PP), printed epidermis and non-printed dermis (PN), and non-printed epidermis and printed dermis (NP). The effects of printing induced shear stress [0.025 kPa (epidermis); 0.049 kPa (dermis)] were characterized both at the cellular and at the tissue level. At cellular level, no statistically significant differences in keratinocyte colony-forming efficiency (CFE) (p = 0.1641) were observed. In the case of fibroblasts, no significant differences in the cell alignment index (p < 0.1717) and their ability to contract collagen gel (p = 0.851) were detected. At the tissue levels, all the four skin equivalents were characterized using histological and immunohistochemical analysis with no significant differences found in either epidermal basal cell count, thickness of viable epidermis, and relative intensity of filaggrin and claudin-1. Our results demonstrated that 3D printing can achieve the same high-quality skin constructs as have been developed traditionally, thus opening new avenues for numerous high-throughput industrial and clinical applications.
