Comparison of artificial intelligence large language model chatbots in answering frequently asked questions in anaesthesia.

Background: Patients are increasingly using artificial intelligence (AI) chatbots to seek answers to medical queries. Methods: Ten frequently asked questions in anaesthesia were posed to three AI chatbots: ChatGPT4 (OpenAI), Bard (Google), and Bing Chat (Microsoft). Each chatbot's answers were evaluated in a randomised, blinded order by five residency programme directors from 15 medical institutions in the USA. Three medical content quality categories (accuracy, comprehensiveness, safety) and three communication quality categories (understandability, empathy/respect, and ethics) were scored between 1 and 5 (1 representing worst, 5 representing best). Results: ChatGPT4 and Bard outperformed Bing Chat (median [inter-quartile range] scores: 4 [3-4], 4 [3-4], and 3 [2-4], respectively; P<0.001 with all metrics combined). All AI chatbots performed poorly in accuracy (score of ≥4 by 58%, 48%, and 36% of experts for ChatGPT4, Bard, and Bing Chat, respectively), comprehensiveness (score ≥4 by 42%, 30%, and 12% of experts for ChatGPT4, Bard, and Bing Chat, respectively), and safety (score ≥4 by 50%, 40%, and 28% of experts for ChatGPT4, Bard, and Bing Chat, respectively). Notably, answers from ChatGPT4, Bard, and Bing Chat differed statistically in comprehensiveness (ChatGPT4, 3 [2-4] vs Bing Chat, 2 [2-3], P<0.001; and Bard 3 [2-4] vs Bing Chat, 2 [2-3], P=0.002). All large language model chatbots performed well with no statistical difference for understandability (P=0.24), empathy (P=0.032), and ethics (P=0.465). Conclusions: In answering anaesthesia patient frequently asked questions, the chatbots perform well on communication metrics but are suboptimal for medical content metrics. Overall, ChatGPT4 and Bard were comparable to each other, both outperforming Bing Chat.

Radiometric studies of Mycobacterium tuberculosis

A atividade metabólica do M. tuberculosis sob diversas condiçöes experimentais foi estudada utilizando um sistema radiométrico automático, capaz de quantificar o 14CO2 produzido pela oxidaçäo de substâncias marcadas com Carbono-14. As experiências realizadas incluíram: a) vias metabólicas; b) determinaçäo dos tempos de detecçäo para inoculaçöes de diversas magnitudes; c) efeito da filtraçäo sobre a reprodutibilidade dos resultados; d) influência de meio hostil; e) determinaçäo das concentraçöes inibitórias mínimas para hidrazida, estreptomicina, etambutol e rifampicina; f) tempo de duplicaçäo para o M. tuberculosis e M. bovis. Estes microorganismos metabolizaram até 14CO2 o 14C-formato, (U-14C) acetato, (u-14C) glicerol, (1-14C) ácido palmítico, (1-14C) ácido láurico, (u-14C) L-ácido málico, (U-14C) D-glicose e (1-14C) D-glicose, mas näo (1-14C) L-glicose, (u-14C) glicina ou (U-14C) piruvato. Usando 14C-formato, (1-14C) ácido palmítico ou (1-14C) ácido láurico foi possível detectar 10 bacilos/frasco em 24-48 horas e até 10 bacilos/frasco em 16-20 dias. Resultados reprodutíveis foram obtidos sem filtrar a suspensäo de bactérias, desde que cultivadas em meio 7H9 líquido com 0,05% de polissorbato 80 e homogenizadas antes da experiência. Drogas que bloqueiam a síntese protêica apresentaram concentraçäo inibitória mínima menor com o método radiométrico do que com o convencional. O tempo médio de duplicaçäo para o M. bovis e várias cepas de M. tuberculosis com diversas substâncias marcadas foi 9 + ou - 1 horas