This study aimed to investigate the accuracy of large language models (LLMs), specifically ChatGPT and Claude, in surgical decision-making and radiological assessment for spine pathologies compared to experienced spine surgeons. The study employed a comparative analysis between the LLMs and a panel of attending spine surgeons. Five written clinical scenarios encompassing various spine pathologies were presented to the LLMs and surgeons, who provided recommended surgical treatment plans. Additionally, magnetic resonance imaging images depicting spine pathologies were analyzed by the LLMs and surgeons to assess their radiological interpretation abilities. Spino-pelvic parameters were estimated from a scoliosis radiograph by the LLMs. Qualitative content analysis revealed limitations in the LLMs' consideration of patient-specific factors and the breadth of treatment options. Both ChatGPT and Claude provided detailed descriptions of magnetic resonance imaging findings but differed from the surgeons in terms of specific levels and severity of pathologies. The LLMs acknowledged the limitations of accurately measuring spino-pelvic parameters without specialized tools. The accuracy of surgical decision-making for the LLMs (20%) was lower than that of the attending surgeons (100%). Statistical analysis showed no significant differences in accuracy between the groups. The study highlights the potential of LLMs in assisting with radiological interpretation and surgical decision-making in spine surgery. However, the current limitations, such as the lack of consideration for patient-specific factors and inaccuracies in treatment recommendations, emphasize the need for further refinement and validation of these artificial intelligence (AI) models. Continued collaboration between AI researchers and clinical experts is crucial to address these challenges and realize the full potential of AI in spine surgery.
SUMMARY: Conventional stable isotope tracing assays track one or several metabolites. However, cells use an array of nutrients to sustain nitrogen metabolic pathways. This incongruency hampers a system level understanding of cellular nitrogen metabolism. Therefore, we created a platform to simultaneously trace 30 nitrogen isotope-labeled metabolites. This platform revealed that while primitive cells engage both de novo and salvage pyrimidine synthesis pathways, differentiated cells nearly exclusively salvage uridine despite expressing de novo pathway enzymes. This link between cell state and pyrimidine synthesis routes persisted in physiological contexts, including primary murine and human tissues and tumor xenografts. Mechanistically, we found that Ser1900 phosphorylation of CAD, the first enzyme of the de novo pathway, was enriched in primitive cells and that mimicking this modification in differentiated cells abrogated their preference for pyrimidine salvage. Collectively, we establish a method for nitrogen metabolism profiling and define a mechanism of cell state-specific pyrimidine synthesis pathway choice.