Reliability of large language models for advanced head and neck malignancies management: a comparison between ChatGPT 4 and Gemini Advanced.

PURPOSE: This study evaluates the efficacy of two advanced Large Language Models (LLMs), OpenAI's ChatGPT 4 and Google's Gemini Advanced, in providing treatment recommendations for head and neck oncology cases. The aim is to assess their utility in supporting multidisciplinary oncological evaluations and decision-making processes. METHODS: This comparative analysis examined the responses of ChatGPT 4 and Gemini Advanced to five hypothetical cases of head and neck cancer, each representing a different anatomical subsite. The responses were evaluated against the latest National Comprehensive Cancer Network (NCCN) guidelines by two blinded panels using the total disagreement score (TDS) and the artificial intelligence performance instrument (AIPI). Statistical assessments were performed using the Wilcoxon signed-rank test and the Friedman test. RESULTS: Both LLMs produced relevant treatment recommendations with ChatGPT 4 generally outperforming Gemini Advanced regarding adherence to guidelines and comprehensive treatment planning. ChatGPT 4 showed higher AIPI scores (median 3 [2-4]) compared to Gemini Advanced (median 2 [2-3]), indicating better overall performance. Notably, inconsistencies were observed in the management of induction chemotherapy and surgical decisions, such as neck dissection. CONCLUSIONS: While both LLMs demonstrated the potential to aid in the multidisciplinary management of head and neck oncology, discrepancies in certain critical areas highlight the need for further refinement. The study supports the growing role of AI in enhancing clinical decision-making but also emphasizes the necessity for continuous updates and validation against current clinical standards to integrate AI into healthcare practices fully.

Patients' Radiation Exposure During Endovascular Abdominal Aortic Aneurysm Repair.

BACKGROUND: To investigate associations between patient characteristics, intraprocedural complexity factors, and radiation exposure to patients during endovascular abdominal aortic aneurysm repair (EVAR). METHODS: Elective standard EVAR procedures between January 2015 and December 2020 were retrospectively analyzed. Patient characteristics and intraprocedural data (i.e., type of device, endograft configuration, additional procedures, and contralateral gate cannulation time [CGCT]) were collected. Dose area product (DAP) and fluoroscopy time were considered as measurements of radiation exposure. Furthermore, effective dose (ED) and doses to internal organs were calculated using PCXMC 2.0 software. Descriptive statistics, univariable, and multivariable linear regression were applied to investigate predictors of increased radiation exposure. RESULTS: The 99 patients were mostly male (90.9%) with a mean age of 74 ± 7 years. EVAR indications were most frequently abdominal aortic aneurysm (93.9%), penetrating aortic ulceration (2.0%), focal dissection (2.0%), or subacute rupture of infrarenal abdominal aortic aneurysm (2.0%). Median fluoroscopy time was 19.6 minutes (interquartile range [IQR], 14.1-29.4) and median DAP was 86,311 mGy cm2 (IQR, 60,160-130,385). Median ED was 23.2 mSv (IQR, 17.0-34.8) for 93 patients (93.9%). DAP and ED were positively correlated with body mass index (BMI) and CGCT. Kidneys, small intestine, active bone marrow, colon, and stomach were the organs that received the highest equivalent doses during EVAR. Higher DAP and ED values were observed using the Excluder endograft, other bi- and tri-modular endografts, and EVAR with ≥2 additional procedures. Multivariable linear regression analysis revealed that BMI, ≥2 additional procedures during EVAR, and CGCT were independent positive predictors of DAP and ED levels after accounting for endograft type. CONCLUSIONS: Patient-related and procedure-related factors such as BMI, ≥2 additional procedures during EVAR, and CGCT resulted predictors of radiation exposure for patients undergoing EVAR, as quantified by higher DAP and ED levels. The main intraprocedural factor that increased radiation exposure was CGCT. These data can be of importance for better managing radiation exposure during EVAR.

Are artificial intelligence large language models a reliable tool for difficult differential diagnosis? An a posteriori analysis of a peculiar case of necrotizing otitis externa.

Key Clinical Message: Large language models have made artificial intelligence readily available to the general public and potentially have a role in healthcare; however, their use in difficult differential diagnosis is still limited, as demonstrated by a case of necrotizing otitis externa. Abstract: This case report presents a peculiar case of necrotizing otitis externa (NOE) with skull base involvement which proved diagnostically challenging. The initial patient presentation and the imaging performed on the 78-year-old patient suggested a neoplastic rhinopharyngeal lesion and only after several unsuccessful biopsies the patient was transferred to our unit. Upon re-evaluation of the clinical picture, a clinical hypothesis of NOE with skull base erosion was made and confirmed by identifying Pseudomonas aeruginosa in biopsy specimens of skull base bone and external auditory canal skin. Upon diagnosis confirmation, the patient was treated with culture-oriented long-term antibiotics with complete resolution of the disease. Given the complex clinical presentation, we chose to submit a posteriori this NOE case to two large language models (LLM) to test their ability to handle difficult differential diagnoses. LLMs are easily approachable artificial intelligence tools that enable human-like interaction with the user relying upon large information databases for analyzing queries. The LLMs of choice were ChatGPT-3 and ChatGPT-4 and they were requested to analyze the case being provided with only objective clinical and imaging data.

Recycling alginate composites for thermal insulation.

We present a new method for the total functional recycling of alginate-based composite materials made via ionotropic gelation. The original material, an alginate/fiberglass foam with thermal insulation characteristics, was produced following a patented process in which fiberglass waste is embedded into the polyanionic gel matrix, and the resulting compound is then freeze-dried. The functional recycling is carried out by disassembling the ionic matrix - which is initially formed by the interaction between a cation (e.g. calcium) and the negatively charged alginate backbone - with the use of a chelator (Ethylenediaminetetraacetic acid disodium salt) with a high affinity for the cations, thus obtaining a homogeneous solution. An ionotropic gel can then be re-formed upon deactivation of the chelating activity under mild acid conditions. We managed to maintain or improve the thermal, mechanical and acoustic performances of the original material and we successfully tested the possibility of multiple recycling cycles.