Can Artificial Intelligence Mitigate Missed Diagnoses by Generating Differential Diagnoses for Neurosurgeons?

BACKGROUND/OBJECTIVE: Neurosurgery emphasizes the criticality of accurate differential diagnoses, with diagnostic delays posing significant health and economic challenges. As large language models (LLMs) emerge as transformative tools in healthcare, this study seeks to elucidate their role in assisting neurosurgeons with the differential diagnosis process, especially during preliminary consultations. METHODS: This study employed 3 chat-based LLMs, ChatGPT (versions 3.5 and 4.0), Perplexity AI, and Bard AI, to evaluate their diagnostic accuracy. Each LLM was prompted using clinical vignettes, and their responses were recorded to generate differential diagnoses for 20 common and uncommon neurosurgical disorders. Disease-specific prompts were crafted using Dynamed, a clinical reference tool. The accuracy of the LLMs was determined based on their ability to identify the target disease within their top differential diagnoses correctly. RESULTS: For the initial differential, ChatGPT 3.5 achieved an accuracy of 52.63%, while ChatGPT 4.0 performed slightly better at 53.68%. Perplexity AI and Bard AI demonstrated 40.00% and 29.47% accuracy, respectively. As the number of considered differentials increased from 2 to 5, ChatGPT 3.5 reached its peak accuracy of 77.89% for the top 5 differentials. Bard AI and Perplexity AI had varied performances, with Bard AI improving in the top 5 differentials at 62.11%. On a disease-specific note, the LLMs excelled in diagnosing conditions like epilepsy and cervical spine stenosis but faced challenges with more complex diseases such as Moyamoya disease and amyotrophic lateral sclerosis. CONCLUSIONS: LLMs showcase the potential to enhance diagnostic accuracy and decrease the incidence of missed diagnoses in neurosurgery.

Excelencia en Neurocirugía: María Cristina García-Sancho, the First Latina Neurosurgeon in the World.

BACKGROUND: Neurosurgery is a specialty that has been dominated by males. Although there has been an increase in the number of women in the field, it is not yet close to being equal. Some noteworthy women who have carved the path for other women to follow in their footsteps include Drs. Sofía Ionescu and Diana Beck, the first and second female neurosurgeons worldwide, respectively. However, there are limited publications on Dr. María Cristina García-Sancho, the first Latina neurosurgeon. METHODS: The purpose of this review was to illuminate the neurosurgical community on the life of Dr. García-Sancho. A thorough literature was performed on medical and non-medical publications that were either authored by Dr. García-Sancho or mentioned her directly. RESULTS: Dr. García-Sancho earned her medical degree at the School of Medicine of the National Autonomous University of Mexico under the guidance of Dr. Clemente Robles, who founded Mexico's first neurosurgical department. Her training took her worldwide. CONCLUSIONS: Her expertise allowed her to pioneer a revolutionary advancement known as the one-step bilateral cordotomy. Her perseverance led her to becoming the head of the Department of Neurosurgery at the National Cancer Institute of Mexico and co-found the Mexican Society of Neurological Surgery, where she served on the board of directors. This review aims to advocate for an equitable environment in the field of neurosurgery with Dr. García-Sancho's story.

Surgical Complication Following Urgent Appendectomy and Considerations in Surgical Management of a Patient With Mitochondrial Encephalopathy, Lactic Acidosis, and Stroke-Like Episodes (MELAS).

This case report details a surgical complication with a delayed presentation in a 23-year-old male with mitochondrial encephalopathy, lactic acidosis, and stroke-like episodes (MELAS). MELAS is a rare entity that can complicate the routine medical and surgical management of patients. Without sufficient research and guidelines, decision-making for patients who require time-sensitive care may be difficult. This patient population may require special consideration and preventative measures to maximize safety in their surgical care. This case serves to highlight a surgical complication that MELAS patients may be susceptible to, as well as to detail possible means of prevention and protection.

Custom order entry for Parkinson's medications in the hospital improves timely administration: an analysis of over 31,000 medication doses.

Background: Patients with Parkinson's disease (PD) are at increased risk for hospital acquired complications. Deviations from home medication schedules and delays in administration are major contributing factors. We had previously developed a protocol to ensure adherence to home medication schedules using "custom" ordering. In this study we are assessing the impact this order type may have on reducing delays in PD medication administration in the hospital. Material and methods: We reviewed 31,404 orders placed for PD medications from January 2, 2016 to April 30 2021. We evaluated the orders to determine if they were placed in a Custom format or using a default non-custom order entry. We further evaluated all orders to determine if there was a relationship with the order type and timely administration of medications. We compared medications that were administered within 1 min, 15 min, 30 min and 60 min of due times across custom orders vs. non-custom default orders. We also evaluated the relationship between ordering providers and type of orders placed as well as hospital unit and type of orders placed. Results: 14,204 (45.23%) orders were placed using a custom schedule and 17,200 (54.77%) orders were placed using non-custom defaults. The custom group showed a significantly lower median delay of 3.06 minutes compared to the non-custom group (p<.001). Custom orders had a significantly more recent median date than non-custom default orders (2019-10-07 vs. 2018-01-06, p<0.001). In additional analyses, medication administration delays were significantly improved for custom orders compared to non-custom orders, with likelihoods 1.64 times higher within 1 minute, 1.40 times higher within 15 minutes, and 1.33 times higher within 30 minutes of the due time (p<0.001 for all comparisons). Conclusion: This is the largest study to date examining the effects of order entry type on timely administration of PD medications in the hospital. Orders placed using a custom schedule may help reduce delays in administration of PD medications.