ChatGPT's learning and reasoning capacity in anesthesiology

Introduction: Over the past few months, ChatGPT has raised a lot of interest given its ability to perform complex tasks through natural language and conversation. However, its use in clinical decision-making is limited and its application in the field of anesthesiology is unknown. Objective: To assess ChatGPT's basic and clinical reasoning and its learning ability in a performance test on general and specific anesthesia topics. Methods: A three-phase assessment was conducted. Basic knowledge of anesthesia was assessed in the first phase, followed by a review of difficult airway management and, finally, measurement of decision-making ability in ten clinical cases. The second and the third phases were conducted before and after feeding ChatGPT with the 2022 guidelines of the American Society of Anesthesiologists on difficult airway management. Results: On average, ChatGPT succeded 65% of the time in the first phase and 48% of the time in the second phase. Agreement in clinical cases was 20%, with 90% relevance and 10% error rate. After learning, ChatGPT improved in the second phase, and was correct 59% of the time, with agreement in clinical cases also increasing to 40%. Conclusions: ChatGPT showed acceptable accuracy in the basic knowledge test, high relevance in the management of specific difficult airway clinical cases, and the ability to improve after learning.

Introducción: En los últimos meses, ChatGPT ha suscitado un gran interés debido a su capacidad para realizar tareas complejas a través del lenguaje natural y la conversación. Sin embargo, su uso en la toma de decisiones clínicas es limitado y su aplicación en el campo de anestesiología es desconocido. Objetivo: Evaluar el razonamiento básico, clínico y la capacidad de aprendizaje de ChatGPT en una prueba de rendimiento sobre temas generales y específicos de anestesiología. Métodos: Se llevó a cabo una evaluación dividida en tres fases. Se valoraron conocimientos básicos de anestesiología en la primera fase, seguida de una revisión del manejo de vía aérea difícil y, finalmente, se midió la toma de decisiones en diez casos clínicos. La segunda y tercera fases se realizaron antes y después de alimentar a ChatGPT con las guías de la Sociedad Americana de Anestesiólogos del manejo de la vía aérea difícil del 2022. Resultados: ChatGPT obtuvo una tasa de acierto promedio del 65 % en la primera fase y del 48 % en la segunda fase. En los casos clínicos, obtuvo una concordancia del 20 %, una relevancia del 90 % y una tasa de error del 10 %. Posterior al aprendizaje, ChatGPT mejoró su tasa de acierto al 59 % en la segunda fase y aumentó la concordancia al 40 % en los casos clínicos. Conclusiones: ChatGPT demostró una precisión aceptable en la prueba de conocimientos básicos, una alta relevancia en el manejo de los casos clínicos específicos de vía aérea difícil y la capacidad de mejoría secundaria a un aprendizaje.

Artificial intelligence, applications and challenges in simulation-based education

The rapid advancement of Artificial Intelligence (AI) has taken the world by "surprise" due to the lack of regulation over this technological innovation which, while promising application opportunities in different fields of knowledge, including education, simultaneously generates concern, rejection and even fear. In the field of Health Sciences Education, clinical simulation has transformed educational practice; however, its formal insertion is still heterogeneous, and we are now facing a new technological revolution where AI has the potential to transform the way we conceive its application.

El rápido avance de la inteligencia artificial (IA) ha tomado al mundo por "sorpresa" debido a la falta de regulación sobre esta innovación tecnológica, que si bien promete oportunidades de aplicación en diferentes campos del conocimiento, incluido el educativo, también genera preocupación e incluso miedo y rechazo. En el campo de la Educación en Ciencias de la Salud la Simulación Clínica ha transformado la práctica educativa; sin embargo, aún es heterogénea su inserción formal, y ahora nos enfrentamos a una nueva revolución tecnológica, en la que las IA tienen el potencial de transformar la manera en que concebimos su aplicación.

Una revisión sistemática del uso de la inteligencia artificial en la educación / A systematic review of the use of artificial intelligence in education

Introducción. El uso de la inteligencia artificial (IA) en la educación ha sido objeto de una creciente atención en los últimos años. La IA se ha utilizado para mejorar la personalización del aprendizaje, la retroalimentación y la evaluación de los estudiantes. Sin embargo, también hay desafíos y limitaciones asociados. El objetivo de este trabajo fue identificar las principales tendencias y áreas de aplicación de la inteligencia artificial en la educación, así como analizar los beneficios y limitaciones de su uso en este ámbito. Métodos. Se llevó a cabo una revisión sistemática que exploró el empleo de la inteligencia artificial en el ámbito educativo. Esta revisión siguió una metodología de investigación basada en la búsqueda de literatura, compuesta por cinco etapas. La investigación se realizó utilizando Scopus como fuente de consulta primaria y se empleó la herramienta VOSviewer para analizar los resultados obtenidos. Resultados. Se encontraron numerosos estudios que investigan el uso de la IA en la educación. Los resultados sugieren que la IA puede mejorar significativamente la personalización del aprendizaje, proporcionando recomendaciones de actividades y retroalimentación adaptadas a las necesidades individuales de cada estudiante. Conclusiones. A pesar de las ventajas del uso de la IA en la educación, también hay desafíos y limitaciones que deben abordarse, como la calidad de los datos utilizados por la IA, la necesidad de capacitación para educadores y estudiantes, y las preocupaciones sobre la privacidad y la seguridad de los datos de los estudiantes. Es importante seguir evaluando los efectos del uso de la IA en la educación para garantizar su uso efectivo y responsable.

Introduction. The use of artificial intelligence (AI) in education has been the subject of increasing attention in recent years. AI has been used to improve personalized learning, feedback, and student assessment. However, there are also challenges and limitations. The aim of this study was to identify the main trends and areas of application of artificial intelligence in education, as well as to analyze the benefits and limitations of its use in this field. Methods. A systematic review was carried out on the use of artificial intelligence in education, using a literature search research methodology with five stages, based on the Scopus query and the tool for analyzing results with VOSviewer. Results. Numerous studies investigating the use of AI in education were found. The results suggest that AI can significantly improve personalized learning by providing activity recommendations and feedback tailored to the individual needs of each student. Conclusions. Despite the advantages of using AI in education, there are also challenges and limitations that need to be addressed, such as the quality of data used by AI, the need for training for educators and students, and concerns about the privacy and security of student data. It is important to continue evaluating the effects of AI use in education to ensure its effective and responsible use.

Artificial intelligence to predict bed bath time in Intensive Care Units / Inteligencia artificial para predecir la hora del baño en cama en Unidades de Cuidados Intensivos / Inteligência artificial para predição do tempo de banho no leito em Unidades de Terapia Intensiva

ABSTRACT Objectives: to assess the predictive performance of different artificial intelligence algorithms to estimate bed bath execution time in critically ill patients. Methods: a methodological study, which used artificial intelligence algorithms to predict bed bath time in critically ill patients. The results of multiple regression models, multilayer perceptron neural networks and radial basis function, decision tree and random forest were analyzed. Results: among the models assessed, the neural network model with a radial basis function, containing 13 neurons in the hidden layer, presented the best predictive performance to estimate the bed bath execution time. In data validation, the squared correlation between the predicted values and the original values was 62.3%. Conclusions: the neural network model with radial basis function showed better predictive performance to estimate bed bath execution time in critically ill patients.

RESUMEN Objetivos: evaluar el rendimiento predictivo de diferentes algoritmos de inteligencia artificial para estimar el tiempo de ejecución del baño en cama en pacientes críticos. Métodos: estudio metodológico, que utilizó algoritmos de inteligencia artificial para predecir el tiempo de baño en cama en pacientes críticos. Se analizaron los resultados de modelos de regresión múltiple, redes neuronales perceptrón multicapa y función de base radial, árbol de decisión y random forest. Resultados: entre los modelos evaluados, el modelo de red neuronal con función de base radial, que contiene 13 neuronas en la capa oculta, presentó el mejor desempeño predictivo para estimar el tiempo de ejecución del baño en cama. En la validación de datos, la correlación al cuadrado entre los valores predichos y los valores originales fue del 62,3%. Conclusiones: el modelo de red neuronal con función de base radial mostró mejor rendimiento predictivo para estimar el tiempo de ejecución del baño en cama en pacientes críticos.

RESUMO Objetivos: avaliar a performance preditiva de diferentes algoritmos de inteligência artificial para estimar o tempo de execução do banho no leito em pacientes críticos. Métodos: estudo metodológico, que utilizou algoritmos de inteligência artificial para predizer o tempo de banho no leito em pacientes críticos. Foram analisados os resultados dos modelos de regressão múltipla, redes neurais perceptron multicamadas e função de base radial, árvore de decisão e random forest. Resultados: entre os modelos avaliados, o modelo de rede neural com função de base radial, contendo 13 neurônios na camada oculta, apresentou melhor performance preditiva para estimar o tempo de execução do banho no leito. Na validação dos dados, o quadrado da correlação entre os valores preditos e os valores originais foi de 62,3%. Conclusões: o modelo de rede neural com função de base radial apresentou melhor performance preditiva para estimar o tempo de execução do banho no leito em pacientes críticos.

Diagnóstico da retinopatia diabética por inteligência artificial por meio de smartphone / Diagnosis of diabetic retinopathy by artificial intelligence using smartphone

RESUMO Objetivo: Obter imagens de fundoscopia por meio de equipamento portátil e de baixo custo e, usando inteligência artificial, avaliar a presença de retinopatia diabética. Métodos: Por meio de um smartphone acoplado a um dispositivo com lente de 20D, foram obtidas imagens de fundo de olhos de pacientes diabéticos; usando a inteligência artificial, a presença de retinopatia diabética foi classificada por algoritmo binário. Resultados: Foram avaliadas 97 imagens da fundoscopia ocular (45 normais e 52 com retinopatia diabética). Com auxílio da inteligência artificial, houve acurácia diagnóstica em torno de 70 a 100% na classificação da presença de retinopatia diabética. Conclusão: A abordagem usando dispositivo portátil de baixo custo apresentou eficácia satisfatória na triagem de pacientes diabéticos com ou sem retinopatia diabética, sendo útil para locais sem condições de infraestrutura.

ABSTRACT Introduction: To obtain fundoscopy images through portable and low-cost equipment using artificial intelligence to assess the presence of DR. Methods: Fundus images of diabetic patients' eyes were obtained by using a smartphone coupled to a device with a 20D lens. By using artificial intelligence (AI), the presence of DR was classified by a binary algorithm. Results: 97 ocular fundoscopy images were evaluated (45 normal and 52 with DR). Through AI diagnostic accuracy around was 70% to 100% in the classification of the presence of DR. Conclusion: The approach using a low-cost portable device showed satisfactory efficacy in the screening of diabetic patients with or without diabetic retinopathy, being useful for places without infrastructure conditions.

A literacia em saúde no ChatGPT: explorando o potencial de uso de inteligência artificial para a elaboração de textos acadêmicos / Health literacy in ChatGPT: exploring the potential of the use of artificial intelligence to produce academic text

Resumo O presente estudo buscou conhecer as principais características das respostas geradas pela ferramenta ChatGPT a consultas sobre um tema emergente na literatura acadêmica de língua portuguesa - a literacia em saúde -, assim como discutir de que forma tais evidências podem contribuir para uma melhor compreensão sobre os limites e os desafios relacionados ao uso de Inteligência Artificial (IA) para a construção do conhecimento acadêmico. Trata-se de um estudo descritivo e exploratório, baseado em consultas ao ChatGPT, a partir de cinco perguntas disparadoras, feitas em sequência, nas línguas portuguesa (Brasil) e inglesa, com níveis de complexidade linguística crescentes. A análise dos resultados evidenciou uma ampla perspectiva para o uso de tecnologias baseadas em IA, como o ChatGPT, uma ferramenta disponibilizada de forma ampla e irrestrita, com uma interface intuitiva e simples, que se mostrou capaz de gerar textos coerentes, estruturados, em linguagem natural. Considerando o fenômeno do produtivismo acadêmico, associado a uma tendência crescente de má conduta profissional, sobretudo o plágio, coloca-se necessidade de um olhar ainda mais cuidadoso sobre o processo de produção e divulgação do conhecimento científico mediado por tecnologias de IA.

Abstract The aim of this study was to identify and analyze the main constituent elements of text generated by ChatGPT in response to questions on an emerging topic in the academic literature in Portuguese - health literacy - and discuss how the evidence produced can contribute to improving our understanding of the limits and challenges of using artificial intelligence (AI) in academic writing. We conducted an exploratory descriptive study based on responses to five consecutive questions in Portuguese and English with increasing levels of complexity put to ChatGPT. Our findings reveal the potential of the use of widely available, unrestricted access AI-based technologies like ChatGPT for academic writing. Featuring a simple and intuitive interface, the tool generated structured and coherent text using natural-like language. Considering that academic productivism is associated with a growing trend in professional misconduct, especially plagiarism, there is a need too take a careful look at academic writing and scientific knowledge dissemination processes mediated by AI technologies.

The function of competitive intelligence in South African insurance post-COVID-19 outbreak

Background: Competitive intelligence (CI) involves monitoring competitors and providing organizations with actionable and meaningful intelligence. Some studies have focused on the role of CI in other industries post-COVID-19 pandemic. Objectives: This article aims to examine the impact of COVID-19 on the South African insurance sector and how the integration of CI and related technologies can sustain the South African insurance sector post-COVID-19 epidemic. Method: Qualitative research with an exploratory-driven approach was used to examine the impact of the COVID-19 pandemic on the South African insurance sector. Qualitative secondary data analyses were conducted to measure insurance claims and death benefits paid during the COVID-19 pandemic. Results: The research findings showed that the COVID-19 pandemic significantly impacted the South African insurance industry, leading to a reassessment of pricing, products, and risk management. COVID-19 caused disparities in death benefits and claims between provinces; not everyone was insured. Despite challenges, South African insurers remained well-capitalised and attentive to policyholders. Integrating CI and analytical technologies could enhance the flexibility of prevention, risk management, and product design. Conclusion: COVID-19 requires digital transformation and CI for South African insurers' competitiveness. Integrating artificial intelligence (AI), big data (BD), and CI enhances value, efficiency, and risk assessments. Contribution: This study highlights the importance of integrating CI strategies and related technologies into South African insurance firms' operations to aid in their recovery from the COVID-19 crisis. It addresses a research gap and adds to academic knowledge in this area.

Application progress of artificial intelligence in the diagnosis of esophageal cancer / 中国胸心血管外科临床杂志

@#Esophageal cancer is an aggressive malignancy with high morbidity and poor prognosis. Symptoms of early esophageal cancer are insidious and difficult to detect, while advanced esophageal obstruction, lesion infiltration and metastasis seriously affect patients’ quality of life. Early detection and treatment can help to increase the survival chance of patients. Recently, artificial intelligence (AI) has shown remarkable success in diagnosis of esophageal cancer, highlighting the great potential of new AI-assisted diagnostic modalities. This paper aims to review recent progress of AI in the diagnosis of esophageal cancer and to prospect its clinical application.

Research progress of artificial intelligence in pathological subtypes classification and gene expression analysis of lung adenocarcinoma / 中国胸心血管外科临床杂志

@#Lung adenocarcinoma is a prevalent histological subtype of non-small cell lung cancer with different morphologic and molecular features that are critical for prognosis and treatment planning. In recent years, with the development of artificial intelligence technology, its application in the study of pathological subtypes and gene expression of lung adenocarcinoma has gained widespread attention. This paper reviews the research progress of machine learning and deep learning in pathological subtypes classification and gene expression analysis of lung adenocarcinoma, and some problems and challenges at the present stage are summarized and the future directions of artificial intelligence in lung adenocarcinoma research are foreseen.

Methods for Identifying Seeds of Chinese Medicinal Materials： A Review / 中国实验方剂学杂志

Seeds are the source for the production of Chinese medicinal materials. The seed authenticity and quality of directly affect the effectiveness and safety of Chinese medicinal materials. The seed quality is faced with the problems such as mixed sources， existence of adulterants and seeds stocked for years， low maturity， and low purity. To ensure the high-quality and sustainable development of the Chinese medicinal material industry， it is urgent to standardize the seed market and identify and evaluate the quality of the seeds circulating in the market. Seed identification methods include visual inspection， microscopic observation， micro-character identification， chemical fingerprinting， molecular identification， electronic nose， X-ray diffraction， electrochemical fingerprinting， spectral imaging， and artificial intelligence. These methods have different application scopes and unique advantages and disadvantages. According to the different species of Chinese herbal medicines and different requirements of testing sites， suitable methods can be selected to achieve rapid and accurate identification with low costs. In the future， the seed identification methods should be developed based on emerging technologies with interdisciplinary knowledge， and intelligent， nondestructive， and single-grain detection methods are needed for the modern Chinese medicinal material industry. This paper introduces the seed identification technologies currently applied in research and production， compares the principles， applicability， advantages， and disadvantages of different technologies， and provides an outlook on the future development of seed identification technologies， aiming to provide a reference for the identification and quality evaluation of seeds of Chinese medicinal material.

The Development of Chinese Herbal Formulae for Non-severe COVID-19 Based on Artificial Intelligence Technology and Investigation of Its Action Mechanisms / 中医杂志

ObjectiveTo develop traditional Chinese medicine （TCM） formulae for the treatment of nonsevere coronavirus disease 2019 （COVID-19） and to explore its anti-inflammatory mechanism. MethodsThe dysregulated signaling pathways were determined in macrophages from bronchoalveolar lavage fluid of COVID-19 patients and in lung epithelial cells infected with SARS-CoV-2 in vitro based on transcriptome analysis. A total of 102 TCM formulae for the clinical treatment of nonsevere COVID-19 were collected through literature. The pathway-reversing rates of these formulae in macrophages and lung epithelial cells were evaluated based on signature signaling pathways， and the basic formula was determined in conjunction with TCM theory. The commonly used Chinese materia medica for nonsevere COVID-19 were summarized from the 102 TCM formulae as abovementioned. And together with the screening results from the Pharmacopoeia of the People's Republic of China， a “Chinese materia medica pool” was esta-blished for the development of TCM formulae for COVID-19. The regulatory effects of each herb on signaling pathways were obtained based on targeted transcriptome analysis. Oriented at reversing dysregulated signaling pathways of COVID-19， the calculation was carried out， and the artificial intelligent methods for compositing formulae， that are exhaustive method and parallel computing， were used to obtain candidate compound formulas. Finally， with reference to professional experience， an innovative formula for the treatment of nonsevere COVID-19 was developed. The ethanol extract of the formula was evaluated for its anti-inflammatory effects by detecting the mRNA expression of interleukin 1b （Il1b）， C-X-C motif chemokine ligand 2 （Cxcl2）， C-X-C motif chemokine ligand 10 （Cxcl10）， C-C motif chemokine ligand 2 （Ccl2）， nitric oxide synthase 2 （Nos2）， and prostaglandin-endoperoxide synthase 2 （Ptgs2） using reverse transcription-quantitative polymerase chain reaction （RT-qPCR） in RAW264.7 cells treated with lipopolysaccharide （LPS）. ResultsIn macrophages and lung epithelial cells， 34 dysregulated signaling pathways associated with COVID-19 were identified respectively. The effects of the 102 formulae for clinical treatment of nonsevere COVID-19 were evaluated based on the dysregulated signaling pathways and targeted transcriptome， and the result showed that Yinqiao Powder and Pingwei Powder （银翘散合平胃散， YQPWP） ranked first， reversing 91.18% of the dysregulated signaling pathways in macrophages and 100% of the dysregulated signaling pathways in lung epithelial cells. Additionally， YQPWP had the function of scattering wind and clearing heat， resolving toxins and removing dampness in accordance with the pathogenesis of wind-heat with dampness in COVID-19. It was selected as the basic formula， and was further modified and optimized to develop an innovative fomula Qiaobang Zhupi Yin （翘蒡术皮饮， QBZPY） based on expert experience and artificial intelligence in composing formulae. QBZPY can reverse all the dysregulated signaling pathways associated with COVID-19 in macrophages and lung epithelial cells， with the reversing rates of 100%. The chief medicinal of QBZPY， including Lianqiao （Fructus Forsythiae）， Xixiancao （Herba Siegesbeckiae） and Niubangzi （Fructus Arctii）， can down-regulate multiple signaling pathways related with virus infection， immune response， and epithelial damage. RT-qPCR results indicated that compared with the model group， the QBZPY group down-regulated the mRNA expression of Il1b， tumor necrosis factor （Tnf）， Cxcl2， Cxcl10， Ccl2， Nos2 and Ptgs2 induced by LPS in RAW264.7 cells （P＜0.05 or P＜0.01）. ConclusionBased on targeted transcriptome analysis， expert experience in TCM and artificial intelligence， QBZPY has been developed for the treatment of nonsevere COVID-19. The ethanol extract of QBZPY has been found to inhibit mRNA expression of several pro-inflammatory genes in a cellular inflammation model.

Artificial intelligence ChatGPT in ophthalmology and scientific writing / 国际眼科杂志(Guoji Yanke Zazhi)

ChatGPT is a large language models(LLMs)that uses deep learning techniques to produce human-like responses to natural language inputs. It belongs to the family of generative pre-training transformer(GPT)models currently publicly available developed by OpenAI in November 2022. ChatGPT is capable of capturing the nuances and intricacies of human language, generating appropriate and contextually relevant responses. It can assist medical professionals in various tasks, such as research, diagnosis, patient monitoring, and medical education, from identifying research programs to assisting in clinical and laboratory diagnosis, to know new developments in their fields and scientific writing. ChatGPT has also attracted increasing attention and widely used in ophthalmology. However, the use of ChatGPT and other artificial intelligence tools in such tasks comes now with several limitations, ethical and legal concerns, such as credibility, plagiarism, copyright infringement, and biases. Future research can focus on developing new methods to mitigate these limitations while harnessing the benefits of ChatGPT in medicine and related aspects.

Ai based motion analysis software for sport and physical therapy assessment / Software de análisis de movimiento basado en ia para la evaluación del deporte y la fisioterapia / Software de análise de movimento baseado em ia para avaliação de esporte e fisioterapia

ABSTRACT Introduction: As the World Health Organization declared the novel coronavirus as a pandemic in March 2020, physical therapy is more difficult to execute, and social distancing is mandatory in the healthcare sector. Objective: In physical therapy, an online video analysis software that provides real-time graphic and numerical information about the patient's movement executions without direct personal contact would mean a significant improvement in eHealth treatment. Methods: We have developed a software layer on top of OpenPose human body position estimation software that can extract the time series of angles of arbitrary body parts using the output coordinates from OpenPose processing the data recorded by two cameras simultaneously. To validate the procedure of determining the joint angles using the Openpose software we have used the Kinovea software. Results: The comparison of the determined maximal knee angle in our and the Kinovea software, which is widely used in biomechanical measurements, was not significantly different (2.03±1.06°, p<0.05) Conclusion: This indicates, that the developed software can calculate the appropriate joint angles with the accuracy that physiotherapy treatments require. As, to our knowledge no such software yet exists, with the help of this software development, therapists could control and correct the exercises in real-time, and also from a distance, and physical therapy effectiveness could be increased. Level of Evidence II; Experimental, comparative.

RESUMEN Introducción: Como la Organización Mundial de la Salud declaró el nuevo coronavirus como una pandemia en marzo de 2020, la fisioterapia es más difícil de ejecutar, el distanciamiento social es obligatorio en el sector de la salud. Objetivo: En la práctica de fisioterapia un software de análisis de vídeo online que proporcione información gráfica y numérica en tiempo real sobre las ejecuciones de movimiento del paciente sin contacto personal directo supondría una mejora significativa en el tratamiento de la eSalud. Métodos: Fue desarrollado una capa de software sobre el software de estimación de posición del cuerpo humano OpenPose que puede extraer la serie temporal de ángulos de partes arbitrarias del cuerpo utilizando las coordenadas de salida de OpenPose procesando los datos registrados por dos cámaras simultáneamente. Para validar el procedimiento de determinación de los ángulos articulares mediante el software Openpose fue utilizado el software Kinovea. Resultados: La comparación del ángulo máximo de rodilla determinado en nuestro software y Kinovea, que es ampliamente utilizado en mediciones biomecánicas, no fue significativamente diferente (2,03±1,06°, p<0,05). Conclusión: Esto indica que el software desarrollado puede calcular los ángulos articulares adecuados con la precisión que requieren los tratamientos de fisioterapia. Dado que aún no existe dicho software, con la ayuda de este desarrollo de software, los terapeutas podrían controlar y corregir los ejercicios en tiempo real, y también a distancia, y se podría aumentar la eficacia de la fisioterapia. Nivel de Evidencia II; Experimental, comparativo.

RESUMO Introdução: Como a Organização Mundial da Saúde declarou o novo coronavírus como pandemia em março de 2020, a fisioterapia é mais difícil de executar, o distanciamento social é obrigatório no setor de saúde. Objetivo: Na prática da fisioterapia, um software de análise de vídeo online que fornece informações gráficas e numéricas em tempo real sobre as execuções de movimento do paciente sem contato pessoal direto significaria uma melhora significativa no tratamento eHealth. Métodos: Desenvolveu-se uma camada de software em cima do software de estimativa de posição do corpo humano OpenPose que pode extrair as séries temporais de ângulos de partes do corpo arbitrárias usando as coordenadas de saída do OpenPose processando os dados gravados por duas câmeras simultaneamente. Para validar o procedimento de determinação dos ângulos articulares utilizando o software Openpose utilizou-se o software Kinovea. Resultados: A comparação do ângulo máximo do joelho determinado em nosso e no software Kinovea, amplamente utilizado em medidas biomecânicas, não foi significativamente diferente (2,03±1,06°, p<0,05) Conclusão: Isso indica que o software desenvolvido pode calcular os ângulos articulares adequados com a precisão que os tratamentos de fisioterapia exigem. Como esse software ainda não existe, com a ajuda do desenvolvimento desse software, os terapeutas puderam controlar e corrigir os exercícios em tempo real, e também à distância, aumentando a eficácia da fisioterapia. Nível de Evidência II; Experimental, comparativo.

Impacto de la inteligencia artificial en las métricas terapéuticas del ataque cerebrovascular durante la pandemia COVID-19 / Impact of artificial intelligence on therapeutic metrics of cerebrovascular attack during the COVID-19 pandemic

Resumen Introducción : El inicio de la pandemia COVID-19, obligó a implementar cambios en el sistema de aten ción de los servicios de emergencia. Coincidentemente, en nuestra institución, implementamos el software de inteligencia artificial (IA), RAPID.AI, para el análisis de imágenes en el ataque cerebrovascular isquémico (ACVi). Nuestro objetivo fue evaluar el impacto del uso de la IA junto a los cambios en el triage durante la pandemia por COVID-19 en pacientes con ACVi por oclusión de gran vaso cerebral (OGVC). Métodos : Se crearon 2 grupos de pacientes con ACVi por OGVC tratados con terapia de reperfusión endovenosa más endovascular o terapia endovascu lar directa. Grupo 1: pacientes de enero 2019 a junio 2020; Grupo 2: pacientes de julio 2020 a diciembre de 2021, estudiados con RAPID.AI. Se analizaron datos clínicos, y métricas temporales. Se compararon según hora de arribo de 08:00 a 20:00 h (diurno) vs. 20:01 a 7:59 h (nocturno). Resultados : El grupo 1 comprendió 153 pacientes y el grupo 2 133. En el grupo 2 la métrica puerta-imagen y adquisición de la imagen fueron menores, con menor tiempo puerta-inicio de imagen y puerta-recanalización; los pacientes en horario nocturno presentaron mayor NIHSS y tiempos inicio-ingreso con menor proporción de independencia funcional a 90 días. Conclusiones : El uso de la IA para el análisis de imá genes junto a un menor tiempo puerta-fin de imagen, permitió acortar el intervalo hasta la punción inguinal. En el análisis por horarios durante la pandemia, los pacientes ingresados en horario diurno presentaron métricas puerta-imagen, tiempo de imagen y puerta-recanalización significativamente menores.

Abstract Introduction : The start of the COVID-19 pandemic forced the implementation of changes in the emergency services care system. Concomitantly, at our institution, we implemented the artificial intelligence (AI) software, RAPID.AI, for image analysis in ischemic stroke (IS). Our objective was to evaluate the impact of the use of AI together with the changes in the triage during the COVID-19 pandemic in patients with stroke due to large vessel occlusion (LVO). Methods : We included patients with IS due to LVO treated with intravenous reperfusion therapy plus en dovascular or direct endovascular therapy. Results : Two groups were created. Group 1: patients from January 2019 to June 2020; Group 2: patients from July 2020 to December 2021, studied with RAPID.AI. Clini cal data and temporal metrics were analyzed. They were compared according to arrival time from 08:00 to 20:00 (daytime) vs 20:01 to 7:59 (night). Results: We included 286 patients, 153 in group 1 and 133 in group 2. In group 2, door-image metric and image duration were lower, with shorter door-image onset and door-recanalization times; patients who arrived at night had higher NIHSS and longer time from onset-to-door with lower propor tion of functional independence at 90 days (mRS ≤ 2). Conclusions : The use of AI for image analysis along with a shorter door to end of image time allowed to reduce the interval to groin puncture. In the analysis by hours during the pandemic, patients admitted in daytime hours had significantly lower door to image, image time acquisition, and door to recanalization metrics.

Consejos para escribir un artículo científico. Parte 1: la práctica de la escritura / Tips for Writing a Scientific Paper: The Practice of Writing

La escritura de artículos académicos es una competencia necesaria para la difusión del conocimiento científico y para el desarrollo profesional de quienes trabajan en diversas disciplinas. Sin embargo, a pesar de su importancia, esta habilidad compleja no suele ser enseñada en forma sistemática, lo que puede operar como una barrera para que los investigadores comuniquen los resultados de sus trabajos. En esta primera entrega, sintetizamos los principales consejos que han brindado expertos en la temática, añadiendo algunos de nuestra experiencia personal que consideramos útiles para facilitar el proceso de la escritura académica y el desarrollo de esta competencia en un contexto colaborativo. En una segunda entrega profundizaremos respecto de la problemática de la escritura de las diferentes secciones de un artículo científico y se ofrecerán consejos para optimizarla y volverla lo más eficaz posible. (AU)

Academic writing is essential for scientific knowledge dissemination and the professional development of those working in various disciplines. Yet, however important this complex skill is, it is not usually taught systematically, a fact that can act as a barrier for researchers to communicate the results of their work. In this first part, we synthesize the main tips provided by experts in the field, adding some of our personal experiences that they consider relevant to facilitate the process of academic writing and develop this skill in a collaborative context. In a second article, we will go deeper into the problem of writing the different sections of a scientific article and offer advice on ways to optimize it and make it as effective as possible. (AU)

ChatGPT: oportunidades y riesgos en la asistencia, docencia e investigación médica / ChatGPT: opportunities and risks in the fields of medical care, teaching, and research

Resumen ChatGPT es un asistente virtual con inteligencia artificial que utiliza lenguaje natural para comunicarse, es decir, mantiene conversaciones como las que se tendrían con otro humano. Puede aplicarse en educación a todos los niveles, que incluye la educación médica, tanto para la formación, la investigación, la escritura de artículos científicos, la atención clínica y la medicina personalizada. Puede modificar la interacción entre médicos y pacientes para mejorar los estándares de calidad de la atención médica y la seguridad, por ejemplo, al sugerir medidas preventivas en un paciente que en ocasiones no son consideradas por el médico por múltiples causas. Los usos potenciales del ChatGPT en la educación médica, como una herramienta de ayuda en la redacción de artículos científicos, un asistente en la atención para pacientes y médicos para una práctica más personalizada, son algunas de las aplicaciones que se analizan en este artículo. Los aspectos éticos, originalidad, contenido inapropiado o incorrecto, citas incorrectas, ciberseguridad, alucinaciones y plagio son ejemplos de las situaciones a tomar en cuenta al usar las herramientas basadas en inteligencia artificial en medicina.

Abstract ChatGPT is a virtual assistant with artificial intelligence (AI) that uses natural language to communicate, i.e., it holds conversations as those that would take place with another human being. It can be applied at all educational levels, including medical education, where it can impact medical training, research, the writing of scientific articles, clinical care, and personalized medicine. It can modify interactions between physicians and patients and thus improve the standards of healthcare quality and safety, for example, by suggesting preventive measures in a patient that sometimes are not considered by the physician for multiple reasons. ChatGPT potential uses in medical education, as a tool to support the writing of scientific articles, as a medical care assistant for patients and doctors for a more personalized medical approach, are some of the applications discussed in this article. Ethical aspects, originality, inappropriate or incorrect content, incorrect citations, cybersecurity, hallucinations, and plagiarism are some examples of situations to be considered when using AI-based tools in medicine.

Chatbots, generative AI, and scholarly manuscripts: WAME recommendations on chatbots and generative artificial intelligence in relation to scholarly publications

This statement revises our earlier "WAME Recommendations on ChatGPT and Chatbots in Relation to Scholarly Publications" (January 20, 2023). The revision reflects the proliferation of chatbots and their expanding use in scholarly publishing over the last few months, as well as emerging concerns regarding lack of authenticity of content when using chatbots. These recommendations are intended to inform editors and help them develop policies for the use of chatbots in papers published in their journals. They aim to help authors and reviewers understand how best to attribute the use of chatbots in their work and to address the need for all journal editors to have access to manuscript screening tools. In this rapidly evolving field, we will continue to modify these recommendations as the software and its applications develop.

Esta declaración revisa las anteriores "Recomendaciones de WAME sobre ChatGPT y Chatbots en Relation to Scholarly Publications" (20 de enero de 2023). La revisión refleja la proliferación de chatbots y su creciente uso en las publicaciones académicas en los últimos meses, así como la preocupación por la falta de autenticidad de los contenidos cuando se utilizan chatbots. Estas recomendaciones pretenden informar a los editores y ayudarles a desarrollar políticas para el uso de chatbots en los artículos sometidos en sus revistas. Su objetivo es ayudar a autores y revisores a entender cuál es la mejor manera de atribuir el uso de chatbots en su trabajo y a la necesidad de que todos los editores de revistas tengan acceso a herramientas de selección de manuscritos. En este campo en rápida evolución, seguiremos modificando estas recomendaciones a medida que se desarrollen el software y sus aplicaciones.

The use of artificial intelligence to improve the scientific writing of non-native english speakers

SUMMARY OBJECTIVE: Scientific writing in English is a daunting task for non-native English speakers. The challenges of writing in a foreign language are evident in the scientific literature where texts by non-native English-speaking scientists tend to be less clear and succinct, contain grammatical errors, and are often rejected by prestigious journals. METHODS: We conducted a non-systematic review of the most recent literature using the terms "Artificial Intelligence," "Scientific Writing," and "Non-English Speaking" to create a narrative review. RESULTS: Artificial intelligence can be a solution to improve scientific writing, especially for non-native English-speaking scientists. Artificial intelligence can assist in the search for pertinent scientific papers, generate summaries, and help with the writing of different sections of the manuscript, including the abstract, introduction, methods, results, and discussion. Artificial intelligence-based programs can correct grammatical errors and improve writing style, both of which are particularly helpful for non-native English speakers. Two artificial intelligence programs that can help with the search for pertinent scientific papers on the internet are Elicit and ResearchRabbit. Scispace Copilot can be used to summarize the retrieved reference. The artificial intelligence software programs such as Grammarly and Paperpal can correct grammatical and spelling errors, while ChatGPT can also restructure sentences and paragraphs, reword text, and suggest appropriate words and phrases. CONCLUSION: Overall, artificial intelligence can be an effective tool to improve the clarity, style, and coherence of scientific writing, helping non-native English-speaking scientists to communicate their research more effectively.

Application of artificial intelligence in predicting malignancy risk in breast masses on ultrasound / Aplicação de inteligência artificial na predição de risco de malignidade em nódulos mamários na ultrassonografia

Abstract Objective: To evaluate the results obtained with an artificial intelligence-based software for predicting the risk of malignancy in breast masses from ultrasound images. Materials and Methods: This was a retrospective, single-center study evaluating 555 breast masses submitted to percutaneous biopsy at a cancer referral center. Ultrasonographic findings were classified in accordance with the BI-RADS lexicon. The images were analyzed by using Koios DS Breast software and classified as benign, probably benign, low to intermediate suspicion, high suspicion, or probably malignant. The histological classification was considered the reference standard. Results: The mean age of the patients was 51 years, and the mean mass size was 16 mm. The radiologist evaluation had a sensitivity and specificity of 99.1% and 34.0%, respectively, compared with 98.2% and 39.0%, respectively, for the software evaluation. The positive predictive value for malignancy for the BI-RADS categories was similar between the radiologist and software evaluations. Two false-negative results were identified in the radiologist evaluation, the masses in question being classified as suspicious by the software, whereas four false-negative results were identified in the software evaluation, the masses in question being classified as suspicious by the radiologist. Conclusion: In our sample, the performance of artificial intelligence-based software was comparable to that of a radiologist.

Resumo Objetivo: O objetivo deste trabalho foi avaliar os resultados de um software baseado em algoritmo de inteligência artificial para predição do risco de malignidade em nódulos mamários. Materiais e Métodos: Estudo retrospectivo e unicêntrico que avaliou 555 nódulos mamários submetidos a biópsia percutânea em um centro de referência oncológico. Os achados ultrassonográficos foram classificados de acordo com o léxico do BI-RADS. As imagens foram analisadas pelo software Koios DS Breast e divididas em benigna ou provavelmente benigna, suspeita baixa ou intermediária, suspeita alta ou provavelmente maligna. O resultado histopatológico foi considerado como padrão ouro. Resultados: A média de idade das pacientes foi de 51 anos e o tamanho médio dos nódulos foi de 16 mm. A sensibilidade e a especificidade foram de 99,1% e 34,0% para o radiologista e 98,2% e 39,0% para o software, respectivamente. O valor preditivo positivo para malignidade para as categorias BIRADS foi semelhante para o radiologista e para o software. Foram identificados dois resultados falso-negativos na avaliação pelo radiologista que foram classificados como suspeitos pelo software, e quatro resultados falso-negativos na avaliação pelo software que foram classificados como suspeitos pelo radiologista. Conclusão: Na nossa amostra, o software de inteligência artificial demonstrou resultados comparáveis à avaliação pelo radiologista.

Validation of a deep learning algorithm for bone age estimation among patients in the city of São Paulo, Brazil / Validação de algoritmo de aprendizado profundo para detecção da idade óssea em pacientes de São Paulo, Brasil

Abstract Objective: To validate a deep learning (DL) model for bone age estimation in individuals in the city of São Paulo, comparing it with the Greulich and Pyle method. Materials and Methods: This was a cross-sectional study of hand and wrist radiographs obtained for the determination of bone age. The manual analysis was performed by an experienced radiologist. The model used was based on a convolutional neural network that placed third in the 2017 Radiological Society of North America challenge. The mean absolute error (MAE) and the root-mean-square error (RMSE) were calculated for the model versus the radiologist, with comparisons by sex, race, and age. Results: The sample comprised 714 examinations. There was a correlation between the two methods, with a coefficient of determination of 0.94. The MAE of the predictions was 7.68 months, and the RMSE was 10.27 months. There were no statistically significant differences between sexes or among races (p > 0.05). The algorithm overestimated bone age in younger individuals (p = 0.001). Conclusion: Our DL algorithm demonstrated potential for estimating bone age in individuals in the city of São Paulo, regardless of sex and race. However, improvements are needed, particularly in relation to its use in younger patients.

Resumo Objetivo: Validar em indivíduos paulistas um modelo de aprendizado profundo (deep learning - DL) para estimativa da idade óssea, comparando-o com o método de Greulich e Pyle. Materiais e Métodos: Estudo transversal com radiografias de mão e punho para idade óssea. A análise manual foi feita por um radiologista experiente. Foi usado um modelo baseado em uma rede neural convolucional que ficou em terceiro lugar no desafio de 2017 da Radiological Society of North America. Calcularam-se o erro médio absoluto (mean absolute error - MAE) e a raiz do erro médio quadrado (root mean-square error - RMSE) do modelo contra o radiologista, com comparações entre sexo, etnia e idade. Resultados: A amostra compreendia 714 exames. Houve correlação entre ambos os métodos com coeficiente de determinação de 0,94. O MAE das predições foi 7,68 meses e a RMSE foi 10,27 meses. Não houve diferenças estatisticamente significantes entre sexos ou raças (p > 0,05). O algoritmo superestimou a idade óssea nos mais jovens (p = 0,001). Conclusão: O nosso algoritmo de DL demonstrou potencial para estimar a idade óssea em indivíduos paulistas, independentemente do sexo e da raça. Entretanto, há necessidade de aprimoramentos, particularmente em pacientes mais jovens.