An emergency room influx and trauma cases prediction in a Brazilian ophthalmological hospital by an ophthalmologist without code experience / Predição do influxo de atendimentos e casos de trauma em um serviço de urgência oftalmológico brasileiro realizado por um oftalmologista sem experiência em programação

ABSTRACT Purpose: The emergency medical service is a fundamental part of healthcare, albeit crowded emergency rooms lead to delayed and low-quality assistance in actual urgent cases. Machine-learning algorithms can provide a smart and effective estimation of emergency patients' volume, which was previously restricted to artificial intelligence (AI) experts in coding and computer science but is now feasible by anyone without any coding experience through auto machine learning. This study aimed to create a machine-learning model designed by an ophthalmologist without any coding experience using AutoML to predict the influx in the emergency department and trauma cases. Methods: A dataset of 356,611 visits at Hospital da Universidade Federal de São Paulo from January 01, 2014 to December 31, 2019 was included in the model training, which included visits/day and the international classification disease code. The training and prediction were made with the Amazon Forecast by 2 ophthalmologists with no prior coding experience. Results: The forecast period predicted a mean emergency patient volume of 216.27/day in p90, 180.75/day in p50, and 140.35/day in p10, and a mean of 7.42 trauma cases/ day in p90, 3.99/day in p50, and 0.56/day in p10. In January of 2020, there were a total of 6,604 patient visits and a mean of 206.37 patients/day, which is 13.5% less than the p50 prediction. This period involved a total of 199 trauma cases and a mean of 6.21 cases/day, which is 55.77% more traumas than that by the p50 prediction. Conclusions: The development of models was previously restricted to data scientists' experts in coding and computer science, but transfer learning autoML has enabled AI development by any person with no code experience mandatory. This study model showed a close value to the actual 2020 January visits, and the only factors that may have influenced the results between the two approaches are holidays and dataset size. This is the first study to apply AutoML in hospital visits forecast, showing a close prediction of the actual hospital influx.

RESUMO Objetivo: Esse estudo tem como objetivo criar um modelo de Machine Learning por um oftalmologista sem experiência em programação utilizando auto Machine Learning predizendo influxo de pacientes em serviço de emergência e casos de trauma. Métodos: Um dataset de 366,610 visitas em Hospital Universitário da Universidade Federal de São Paulo de 01 de janeiro de 2014 até 31 de dezembro de 2019 foi incluído no treinamento do modelo, incluindo visitas/dia e código internacional de doenças. O treinamento e predição foram realizados com o Amazon Forecast por dois oftalmologistas sem experiência com programação. Resultados: O período de previsão estimou um volume de 206,37 pacientes/dia em p90, 180,75 em p50, 140,35 em p10 e média de 7,42 casos de trauma/dia em p90, 3,99 em p50 e 0,56 em p10. Janeiro de 2020 teve um total de 6.604 pacientes e média de 206,37 pacientes/dia, 13,5% menos do que a predição em p50. O período teve um total de 199 casos de trauma e média de 6,21 casos/dia, 55,77% mais casos do que a predição em p50. Conclusão: O desenvolvimento de modelos era restrito a cientistas de dados com experiencia em programação, porém a transferência de ensino com a tecnologia de auto Machine Learning permite o desenvolvimento de algoritmos por qualquer pessoa sem experiencia em programação. Esse estudo mostra um modelo com valores preditos próximos ao que ocorreram em janeiro de 2020. Fatores que podem ter influenciados no resultado foram feriados e tamanho do banco de dados. Esse é o primeiro estudo que aplicada auto Machine Learning em predição de visitas hospitalares com resultados próximos aos que ocorreram.

Application of data mining algorithms in research on traditional Chinese medicine formula / 中国药房

In recent years， data mining algorithms have been widely employed in scientific research within the field of traditional Chinese medicine （TCM）. The data mining algorithms are used to effectively handle and analyze the complex data in TCM formulas， providing a rational explanation for the mechanism of action. This method has proven particularly useful in uncovering patterns of compatibility and frequent combinations of herbs in TCM， thereby enhancing the reliability and accuracy of clinical diagnosis， target screening， and the study of new drugs. This paper reviews and analyzes 147 papers on TCM formula research that utilize data mining algorithms. The results indicate that data mining algorithms play a unique advantage in six sub- areas， including the study on the mechanism of action in TCM formula， the dose-efficacy of TCM formulas， the identification of core drugs pairs/groups， mining the relationships among “formulas-drug-symptom”， the discovery of new formulas， and mining the compatibility law. Notably， association rules and clustering algorithms are the most representative.

Algoritmo de Indicação de Equipamento Coletor para Estomias / Algorithm for Indication of Collector Equipment for Stomas / Algoritmo para la Indicación de Equipos Colectores para Estomas

Objetivo:Relatar a experiência de uma equipe de enfermeiros estomaterapeutas na construção de um algoritmo para a indicação de equipamento coletor para estomias de eliminação. Método: Relato de experiência, do período de janeiro de 2018 a setembro de 2019, sobre o processo de construção de um algoritmo para indicação de equipamento coletor para estomias de eliminação. Resultados: A partir de determinadas características clínicas (parâmetros de avaliação) e da categorização dos equipamentos coletores (solução), foi desenvolvido um algoritmo para indicação de equipamento coletor para estomias de eliminação. Conclusão: Espera-se que esse instrumento possa auxiliar os enfermeiros na sua prática profissional quanto à escolha do equipamento coletor e na construção de protocolos clínicos.

Objective:To report the experience of a team of enterostomal therapists in the construction of an algorithm for the indication of collecting equipment for elimination stomas. Method: Experience report, from January 2018 to September 2019, on the process of building an algorithm to indicate collecting equipment for elimination stomas. Results: Based on certain clinical characteristics (assessment parameters) and the categorization of collecting equipment (solution), an algorithm was developed to indicate collecting equipment for elimination stomas. Conclusion: It is expected that this instrument can help nurses in their professional practice regarding the choice of collecting equipment and the construction of clinical protocols.

Objetivo:Relatar la experiencia de un equipo de enfermeros estomaterapeutas en la construcción de un algoritmo para la indicación de equipos recolectores para estomas de eliminación. Método: Informe de experiencia, de enero de 2018 a septiembre de 2019, sobre el proceso de construcción de un algoritmo para indicar equipos colectores para estomas de eliminación. Resultado: A partir de ciertas características clínicas (parámetros de evaluación) y la categorización de los equipos colectores (solución), se desarrolló un algoritmo para indicar equipos colectores para estomas de eliminación. Conclusión: Se espera que este instrumento pueda ayudar a los enfermeros en su práctica profesional en cuanto a la elección de equipos de recolección y la construcción de protocolos clínicos.

El hombre perfecto / I am your person

La fantasía que impera en este film plantea la ilusión de encontrar un ser complementario que se adapte a nuestras preferencias y nos haga plenos. "Mi algoritmo está diseñado para hacerte feliz" dice el humanoide. Ilusión de que alguien tendría la posibilidad de ser complementario, de saber exactamente lo que el otro requiere. Estamos en las antípodas de la famosa fórmula de Lacan:" (Le Séminaire, Encore, 1975) "No hay relación sexual" (o sea, no hay complementariedad). No habría resto, el sujeto no estaría atravesado por la castración simbólica. La IA compite con Zeus. La fantasía del Uno, organismo previo a la separación del andrógino por parte de Zeus, se podría materializar con la IA

The fantasy that prevails in this film, raises the illusion of finding a complementary being that adapts to our preferences and makes us full. "My algorithm is designed to make you happy," says the humanoid. Illusion that someone would have the possibility of being complementary, of knowing exactly what the other requires. We are at the antipodes of Lacan's famous formula: "(Le Séminaire, Encore, 1975) "There is no sexual intercourse" (that is, there is no complementarity). There would be no rest, the subject would not be pierced by symbolic castration. AI competes with Zeus. The fantasy of the One, an organism prior to the separation of the androgynous by Zeus, could materialize with AI.

Detection of alloimmunization in Glanzmann Thrombasthenia and Bernard-Soulier Syndrome: Data from a Brazilian Center

ABSTRACT Introduction: The Glanzmann Thrombasthenia (GT) and Bernard-Soulier Syndrome (BSS) are rare hereditary disorders of platelet function. Their treatment often requires platelet transfusion, which can lead to the development of alloantibodies. Objective: In this study, we aim to develop a strategy for alloantibody detection and to describe the frequency of alloimmunization in a patient population from a single center in southeastern Brazil. Methods: Samples from patients with GT or BSS were tested using the Platelet Immunofluorescence Test (PIFT). If a positive result was obtained, a confirmatory step using the Monoclonal Antibody Immobilization of Platelet Antigens (MAIPA) and Luminex bead-based platelet assay (PAKLx) was executed. Main results: Among 11 patients with GT, we detected the presence of alloantibodies in 5 using PIFT, with confirmation through MAIPA and PAKLx in 2 (1 anti-HLA and 1 anti-HPA), resulting in a frequency of 18.1%. Among 4 patients with BSS, PIFT was positive in 3, with confirmation by MAIPA and PAKLx in 1 (anti-HLA), showing a frequency of 25%. The two patients with anti-HLA antibodies exhibited a panel reactive antibody (PRA-HLA) testing greater than 97%. Conclusion: Our study highlights the importance of identifying platelet alloimmunization in this patient population. The proposed algorithm for platelet alloantibodies detection allows resource optimization.

Experiencia en la introducción de resultados científicos desde la formación doctoral / Experience in the introduction of scientific results from the doctoral training i: en

Durante estos años, condicionados por los efectos de una pandemia y la situación económica global, la incorporación oportuna de los resultados científico-técnicos es necesidad y responsabilidad de la comunidad científica. En este trabajo se expone una experiencia en la introducción de resultados científicos desde la formación doctoral, dirigida al área de la atención inicial al paciente con traumatismo maxilofacial. La importancia de esta práctica radica en los aportes social, científico y profesional y en la formación de recursos humanos para lograr la transformación y el mejoramiento de la realidad.

During these years, conditioned by the effects of a pandemic and the global economic situation, the opportune incorporation of the scientific technical results is necessity and responsibility of scientific community. An experience in the introduction of scientific results from the doctoral training, directed to the area of initial care to the patient with maxillofacial traumatism, is presented in this work. The importance of this practice resides in the social, scientific, professional contributions and in the formation of human resources to achieve the transformation and improvement of reality.

Application of data mining to extract knowledge about the occurrence of fistulas after palatoplasty / Aplicação de mineração de dados para extração de conhecimento sobre ocorrência de fístulas após palatoplastia

Introduction: Data mining techniques expand access to important information for the decision-making process during health care. The objective the study proposes using data mining techniques to identify variables (surgical treatment protocols, patient characteristics, post-surgical complications) associated with fistulas after primary palatoplasty in patients with unilateral transforamen incisor cleft (UTIC). Method: A data set of 222 patients with UTIC without syndromes, operated by four surgeons with Furlow's or von Langenbeck's primary palatoplasty techniques, was analyzed for this study. Two models for detecting the outcome of surgery were induced using data mining techniques (Decision Tree and Apriori). Results: Five rules were selected from a decision tree pointing to some variables as predictors of fistulas associated with primary palatoplasty: infection, cough, hypernasality, and surgeon. Analysis of the model indicates that it correctly classifies 95.9% of occurrences between the absence and presence of fistulas. The second model indicates that the absence of post-surgical complications (infection and fever) and normal speech results (absent hypernasality, without suggestive of velopharyngeal dysfunction) are related to the absence of fistulas. Regarding surgical procedures, the Furlow technique and the Vomer flap were more frequent in patients with fistulas. Conclusion: Data mining techniques, as applied in the present study, pointed to infection and cough, hypernasality, and surgeon and surgical techniques as predictors of fistulas related to primary palatoplasty.

Introdução: As técnicas de mineração de dados ampliam o acesso a informações importantes para o processo de tomada de decisão durante os cuidados com a saúde. O objetivo do estudo propõe a utilização de técnicas de mineração de dados para identificar variáveis (protocolos de tratamento cirúrgico, características do paciente, intercorrências pós-cirúrgicas) associadas à ocorrência de fístulas após palatoplastia primária em pacientes com fissura transforame incisivo unilateral (FTIU). Método: Um conjunto de dados de 222 pacientes com FTIU sem síndromes, operados por quatro cirurgiões com as técnicas de palatoplastia primária de Furlow ou von Langenbeck, foi analisado para este estudo. Dois modelos para detecção do resultado da cirurgia foram induzidos usando técnicas de mineração de dados (Árvore de Decisão e Apriori). Resultados: Cinco regras foram selecionadas de uma árvore de decisão apontando para algumas variáveis como preditivas de fístulas associadas à palatoplastia primária: infecção, tosse, hipernasalidade, cirurgião. A análise do modelo indica que ele classifica corretamente 95,9% das ocorrências entre ausência e presença de fístulas. O segundo modelo indica que a ausência de intercorrências pós-cirúrgicas (infecção e febre) e resultado de fala normal (hipernasalidade ausente, sem sugestivo de disfunção velofaríngea) estão relacionados à ausência de fístulas. Em relação aos procedimentos cirúrgicos, o uso da técnica de Furlow e retalho de Vomer foram mais frequentes nos pacientes com fístulas. Conclusão: Técnicas de mineração de dados, conforme aplicadas no presente estudo, apontaram para infecção e tosse, presença de hipernasalidade, cirurgião e técnica cirúrgica como preditores de fístulas relacionadas à palatoplastia primária.

Recent research on machine learning in the diagnosis and treatment of necrotizing enterocolitis in neonates / 中国当代儿科杂志

Necrotizing enterocolitis (NEC), with the main manifestations of bloody stool, abdominal distension, and vomiting, is one of the leading causes of death in neonates, and early identification and diagnosis are crucial for the prognosis of NEC. The emergence and development of machine learning has provided the potential for early, rapid, and accurate identification of this disease. This article summarizes the algorithms of machine learning recently used in NEC, analyzes the high-risk predictive factors revealed by these algorithms, evaluates the ability and characteristics of machine learning in the etiology, definition, and diagnosis of NEC, and discusses the challenges and prospects for the future application of machine learning in NEC.

Machine learning in medicine: what clinicians should know

With the advent of artificial intelligence (AI), machines are increasingly being used to complete complicated tasks, yielding remarkable results. Machine learning (ML) is the most relevant subset of AI in medicine, which will soon become an integral part of our everyday practice. Therefore, physicians should acquaint themselves with ML and AI, and their role as an enabler rather than a competitor. Herein, we introduce basic concepts and terms used in AI and ML, and aim to demystify commonly used AI/ML algorithms such as learning methods including neural networks/deep learning, decision tree and application domain in computer vision and natural language processing through specific examples. We discuss how machines are already being used to augment the physician's decision-making process, and postulate the potential impact of ML on medical practice and medical research based on its current capabilities and known limitations. Moreover, we discuss the feasibility of full machine autonomy in medicine.

Algoritmos para prevenir lesão por pressão em paciente com COVID-19 em prona / Algoritmos para prevenir la úlcera por presión en pacientes con COVID-19 en prona / Algorithms to prevent pressure injury in COVID-19 patients in prone position

Resumo Objetivo Elaborar e validar o conteúdo de dois algoritmos para orientar profissionais da linha de frente na prevenção e no tratamento da lesão por pressão em paciente com COVID-19 em posição prona. Métodos Estudo realizado entre setembro e novembro de 2021. Para a construção dos algoritmos, realizou-se revisão da literatura junto às bases de dados MEDLINE®, SciELO e Lilacs. Foram pesquisados artigos publicados entre 2011 e 2021. A validação dos algoritmos foi feita por 59 profissionais da saúde (enfermeiros, fisioterapeutas e médicos), que trabalhavam na linha de frente da COVID-19, utilizando-se a técnica Delphi. Para a análise de dados, foi adotado o Índice de Validade de Conteúdo e o coeficiente alfa de Cronbach. Resultados No primeiro ciclo de avaliação, os itens dos algoritmos foram considerados pelos juízes como "parcialmente adequados a totalmente adequados", e o Índice de Validade de Conteúdo variou entre 0,87 e 0,92. O coeficiente alfa de Cronbach variou entre 0,95 e 0,96, indicando excelente consistência interna do questionário de avaliação utilizado pelos juízes. Após implementados os ajustes sugeridos pelos juízes, os algoritmos foram reenviados para o segundo ciclo de avaliação, no qual todos os itens foram julgados como "adequado" e "totalmente adequado", resultando em um Índice de Validade do Conteúdo de 1,0. Conclusão Os algoritmos para orientar profissionais da saúde na prevenção e no tratamento da lesão por pressão em pacientes com COVID-19 em posição prona foram avaliados por enfermeiros, fisioterapeutas e médicos que estavam na linha de frente de combate à COVID-19, que chegaram a um consenso quanto ao conteúdo no segundo ciclo de avaliação.

Resumen Objetivo Elaborar y validar el contenido de dos algoritmos para orientar profesionales de la línea de frente sobre la prevención y tratamiento de la úlcera por presión en pacientes con COVID-19 en posición prona. Métodos Estudio realizado entre septiembre y noviembre de 2021. Para la elaboración de los algoritmos, se realizó revisión de la literatura en las bases de datos MEDLINE®, SciELO y Lilacs. Se buscaron artículos publicados entre 2011 y 2021. La validación de los algoritmos fue realizada por 59 profesionales de la salud (enfermeros, fisioterapeutas y médicos), que trabajaban en la línea de frente del COVID-19, utilizando el método Delphi. Para el análisis de datos se adoptó el Índice de Validez de Contenido y el coeficiente alfa de Cronbach. Resultados En el primer ciclo de evaluación, los ítems de los algoritmos fueron considerados por los jueces como "parcialmente adecuados a totalmente adecuados", y el Índice de Validez de Contenido varió entre 0,87 y 0,92. El coeficiente alfa de Cronbach varió entre 0,95 y 0,96, lo que indica una excelente consistencia interna del cuestionario de evaluación utilizado por los jueces. Después de implementar las mejoras sugeridas por los jueces, se reenviaron los algoritmos para el segundo ciclo de evaluación, en el cual todos los ítems fueron calificados como "adecuado" y "totalmente adecuado", con un resultado del Índice de Validez de Contenido de 1,0. Conclusión Los algoritmos para orientar profesionales de la salud sobre la prevención y el tratamiento de la úlcera por presión en pacientes con COVID-19 en posición prona fueron evaluados por enfermeros, fisioterapeutas y médicos que estaban en la línea de frente de combate al COVID-19 y llegaron a un consenso respecto al contenido en el segundo ciclo de evaluación.

Abstract Objective To develop and validate the content of two algorithms to guide frontline professionals in the prevention and treatment of pressure injuries in COVID-19 patients in prone position. Methods Study conducted between September and November 2021. A literature review was performed in MEDLINE®, SciELO and Lilacs databases to build the algorithms. Articles published between 2011 and 2021 were searched. The validation of algorithms was performed by 59 health professionals (nurses, physical therapists and physicians) who worked on the frontline of COVID-19. The Delphi technique was used, and Content Validity Index and Cronbach's alpha coefficient were adopted for data analysis. Results In the first evaluation cycle, the items of algorithms were considered as "partially adequate to totally adequate" by the judges, and the Content Validity Index ranged between 0.87 and 0.92. Cronbach's alpha coefficient ranged between 0.95 and 0.96, indicating excellent internal consistency of the evaluation questionnaire used by the judges. After implementing the adjustments suggested by judges, the algorithms were sent to a second evaluation cycle, in which all items were judged as "adequate" and "totally adequate", resulting in a Content Validity Index of 1.0. Conclusion Algorithms to guide healthcare professionals in the prevention and treatment of pressure injury in COVID-19 patients in prone position were evaluated by nurses, physical therapists and physicians working on the frontline of COVID-19. They achieved consensus on content in the second evaluation cycle.

Association of human papillomavirus (HPV), p16, p53 and p63 expression with non-bilharzia-associated squamous cell carcinoma of the bladder and algorithm construction for histopathological grading prediction

ABSTRACT Objective To investigate the expression of human papillomavirus (HPV), p16, p53, and p63 in non-schistosomiasis-related squamous cell carcinoma of the bladder and to develop an accurate and automated tool to predict histological classification based on clinicopathological features. Methods Twenty-eight patients with primary bladder pure squamous cell carcinoma who underwent cystectomy or transurethral resection of bladder tumor (TURBT) for bladder cancer between January 2011 and July 2017 were evaluated. Clinical data and follow-up information were obtained from medical records. Formalin-fixed, paraffin-embedded surgical specimens were used for immunohistochemical staining for p16, p53, and p63. Human papillomavirus detection was evaluated by PCR. Statistical analysis was performed, and statistical significance was set at p<0.05. Finally, decision trees were built to classify patients' prognostic features. Leave-one-out cross-validation was used to test the generalizability of the model. Results Neither direct HPV detection nor its indirect marker (p16 protein) was identified in most cases. The absence of p16 was correlated with less aggressive histological grading (p=0.040). The positive p16 staining detection found only in pT1 and pT2 cases in our sample suggests a possible role for this tumor suppressor protein in the initial stages of bladder squamous cell carcinoma. The decision trees constructed described the relationship between clinical features, such as hematuria/dysuria, the level of tumor invasion, HPV status, lymphovascular invasion, gender, age, compromised lymph nodes, and tumor degree differentiation, with high classification accuracy. Conclusion The algorithm classifier approach established decision pathways for semi-automatic tumor histological classification, laying the foundation for tailored semi-automated decision support systems for pathologists.

Epipericardial fat necrosis as a differential diagnosis of acute chest pain: a case report and algorithm proposal for diagnostic approach

ABSTRACT Chest pain is a frequent, potentially life-threatening condition in the emergency department and requires immediate investigation and treatment. This case report highlights a rare differential diagnosis of pleuritic chest pain: epipericardial fat necrosis. A 29-year-old man presented with normal clinical evaluation, electrocardiography, point-of-care ultrasound, and unremarkable laboratory tests. The initial hypothesis was acute pleuritis. Chest radiography revealed peri-cardiac nonspecific findings, and computed tomography revealed epicardial fat necrosis. Despite the rarity of this condition, accurate diagnosis allows for better practices. An algorithm for a diagnostic approach is proposed.

Hiperkalemia crónica, diagnóstico y manejo. Consenso colombiano / Chronic hyperkalemia, diagnosis and management. Colombian consensus

Resumen Objetivo: Generar recomendaciones para el diagnóstico, el manejo y el seguimiento de la hiperkalemia crónica. Método: Este consenso fue realizado por nefrólogos y cardiólogos siguiendo la metodología GRADE. Resultados: La hiperkalemia crónica puede definirse como una condición bioquímica, con o sin manifestaciones clínicas, caracterizada por una elevación recurrente de las concentraciones séricas de potasio que puede requerir una intervención farmacológica, no farmacológica o ambas. Puede clasificarse en leve (K+ 5,0 a < 5,5 mEq/l), moderada (K+ 5,5 a 6,0 mEq/l) o grave (K+ > 6,0 mEq/l). Su incidencia y prevalencia no han sido claramente determinadas. Se consideran factores de riesgo la enfermedad renal crónica, la insuficiencia cardiaca crónica, la diabetes mellitus, la edad ≥ 65 años, la hipertensión arterial y el tratamiento con inhibidores del sistema renina-angiotensina-aldosterona (iSRAA), entre otros. No hay consenso sobre el manejo de la hiperkalemia crónica. Se sugiere identificar y eliminar o controlar los factores de riesgo, brindar asesoramiento sobre la ingesta de potasio y, para quien esté indicado, optimizar la terapia con iSRAA, administrar aglutinantes orales del potasio y corregir la acidosis metabólica. Conclusiones: Se recomienda prestar atención al diagnóstico, el manejo y el seguimiento de la hiperkalemia crónica, en especial en los pacientes con factores de riesgo.

Abstract Objective: Generate recommendations for the diagnosis, management, and follow-up of chronic hyperkalemia. Method: This consensus was made by nephrologists and cardiologists following the GRADE methodology. Results: Chronic hyperkalemia can be defined as a biochemical condition with or without clinical manifestations characterized by a recurrent elevation of serum potassium levels that may require pharmacological and or non-pharmacological intervention. It can be classified as mild (K+ 5.0 to < 5.5 mEq/L), moderate (K+ 5.5 to 6.0 mEq/L) or severe (K+ > 6.0 mEq/L). Its incidence and prevalence have yet to be determined. Risk factors: chronic kidney disease, chronic heart failure, diabetes mellitus, age ≥ 65 years, hypertension, and drugs that inhibit the renin angiotensin aldosterone system (RAASi), among others. There is no consensus for the management of chronic hyperkalemia. The suggested pattern for patients is to identify and eliminate or control risk factors, provide advice on potassium intake and, for whom it is indicated, optimize RAASi therapy, administer oral potassium binders and correct metabolic acidosis. Conclusions: The recommendation is to pay attention to the diagnosis, management, and follow-up of chronic hyperkalemia, especially in patients with risk factors.

Significance Interpretation, Risk Assessment, and Prevention Paths of Intelligent Medicine / 中国医学伦理学

Artificial intelligence (AI) has shown a tendency to replace human brain intelligence, and its rise has led to significant social changes. AI has pioneered the historical process of human-machine co-evolution. In the embedding of AI into medicine, it can be seen that there are a wide range of medical application scenarios. Intelligent medicine based on AI technology is of great significance, achieving spatial transmutation under spatial connection and value release under data mining. Intelligent medicine occupies a "high-level position" in value evaluation. Intelligent medicine has become a risk complex due to its tendency towards de-subjectivity of doctors, questioning the effectiveness of the intelligent algorithms, increasing difficulty in protecting information security and individual privacy, and indifference to medical humanistic care. To prevent the risks of intelligent medicine, it is necessary to reasonably define the relationship between AI and doctors, effectively prevent the risks of intelligent algorithms, increase the protection efforts of patient data security and personal privacy, thickly cultivate humanistic care in AI, and promote the improvement of the scientific and technological ethical governance system through ethical review.

Comparison of effects of absolute scatter correction and relative scatter correction on image quality in PET reconstruction / 中华核医学与分子影像杂志

Objective:To study the impact of different scattering correction algorithms in the reconstruction of PET/CT images on image artifacts and the precision of quantitative parameters.Methods:The phantom as described in the National Electrical Manufacturers Association (NEMA) NU2 standard was filled with 18F. The background activity was fixed, and the activity of the solution in the spheres was adjusted to obtain several configurations, including the normal ratio group (4.08∶1) and the extreme ratio group (200∶1). The surface contamination group with the same ratio as the extreme ratio group contained a small radioactive source with different doses of 18F (74, 37, 3.7 and 0.37 MBq) placed at the surface of the phantom. PET/CT images of 30 patients (21 males, 9 females, age: (44.5±10.2) years) from Peking University Cancer Hospital & Institute between July 2012 and December 2021 were retrospectively analyzed, including 10 with normal images ( 18F-FDG) and 20 with abnormal images (10 with dislocation during acquisition, 10 with surface contamination). The images were reconstructed with relative and absolute scattering correction. The phantom was evaluated using the target to background ratio (TBR) and the artifact classification. CV as well as the artifact classification were used to compare the clinical image quality. Mann-Whitney U test and χ2 test were used to analyze data. Results:In the normal ratio group and the extreme ratio group, the TBRs of phantom images reconstructed with relative correction were significantly higher than those with absolute correction (normal ratio group: 3.30(1.94, 4.53) vs 2.72(1.56, 3.56); z=-2.20, P=0.028; extreme ratio group: 105.47(45.62, 162.82) vs 101.36(43.96, 155.57); z=-1.99, P=0.046). In the surface contamination group, with the increase of the activity of the small source, the artifact became more obvious, and the artifact classification score of absolute correction was significantly better than that of relative correction (1.5(1.0, 2.0) vs 2.5(2.0, 3.0); z=-2.00, P=0.046). In the 10 normal 18F-FDG PET/CT patients, the CVliver of the relative correction (9.67%(8.00%, 11.00%)) was significantly lower than that of absolute correction (11.00%(9.00%, 12.00%); z=-2.57, P=0.010), indicating the higher image quality of images with relative correction. In abnormal images, the image quality of absolute correction was significantly higher than that of relative correction with fewer and less severe artifacts (dislocation cases: 9/10 vs 4/10; χ2=5.50, P=0.019; surface contamination cases: 9/10 vs 4/10; χ2=5.50, P=0.019). Conclusions:The relative scattering correction is suitable for normal situations in clinical PET acquisition. However, with dislocation or surface contamination, the absolute scattering correction helps to reduce the artifacts and improve the image quality.

Exaustão algorítmica: influenciadores digitais, trabalho de plataforma e saúde mental / Algorithmic exhaustion: digital influencers, platform work, and mental health / Agotamiento algorítmico: influencers digitales, trabajo en plataformas y salud mental

Este trabalho tem como objetivo articular as noções de tecnologia, trabalho, saúde e influenciadores digitais e reivindicar essa articulação como objeto de investigação para o campo da comunicação. Especificamente, busca-se entender as particularidades do esgotamento vivido por influenciadores digitais, a partir de revisão bibliográfica e exposição de exemplos (coletados a partir de observação espontânea). Como resultado, propõe-se a noção de 'exaustão algorítmica', uma sensação relatada por influenciadores digitais relacionada aos 'problemas psicológicos' vivenciados por eles e gerados pelo ritmo de trabalho que vem sendo ditado pelo que reconhecem como 'o algoritmo'. A 'exaustão'caracteriza-se por um sentimento permanente de insatisfação, desânimo e esgotamento, ausência de criatividade, medo de penalidades das plataformas e de 'não dar conta'.

This work aims to articulate the notions of technology, work, health and digital influencers and claim this articulation as an object of investigation for the field of communication. Specifically, we seek to understand the particularities of the exhaustion experienced by digital influencers from a literature review and exposition of examples (obtained by spontaneous observation). As a result, the notion of 'algorithmic exhaustion' is proposed, a sensation reported by digital influencers that one is going through 'psychological problems' generated by the pace of work dictated by what they recognize as 'the algorithm'. 'Exhaustion' is characterized by a permanent feeling of dissatisfaction, discouragement and exhaustion, lack of creativity, fear of platform penalties and 'not getting it done'.

Este trabajo tiene como objetivo articular las nociones de tecnología, trabajo, salud e influencers digitales y reivindicar esa articulación como objeto de investigación para el campo de la comunicación. En concreto, buscamos comprender las particularidades del agotamiento que experimentan los influencers digitales a partir de una revisión bibliográfica y exposición de ejemplos (obtenidos por observación espontánea). Como resultado, se propone la noción de 'agotamiento algorítmico', sensación reportada por influencersdigitales de que se está pasando por 'problemas psicológicos' generados por el ritmo de trabajo dictado por lo que reconocen como 'el algoritmo'. El agotamiento se caracteriza por un sentimiento permanente de insatisfacción, desánimo y agotamiento, falta de creatividad, miedo a las sanciones de la plataforma y al 'no hacerlo'.

Predição de choque séptico e hipovolêmico em pacientes de unidade de terapia intensiva com o uso de machine learning / Prediction of septic and hypovolemic shock in intensive care unit patients using machine learning

RESUMO Objetivo: Criar e validar um modelo de predição de choque séptico ou hipovolêmico a partir de variáveis de fácil obtenção coletadas na admissão de pacientes internados em uma unidade de terapia intensiva. Métodos: Estudo de modelagem preditiva com dados de coorte concorrente realizada em um hospital do interior do nordeste brasileiro. Foram incluídos pacientes com 18 anos ou mais sem uso de droga vasoativa no dia da admissão e que foram internados entre novembro de 2020 e julho de 2021. Foram testados os algoritmos de classificação do tipo Decision Tree, Random Forest, AdaBoost, Gradient Boosting e XGBoost para a construção do modelo. O método de validação utilizado foi o k-fold cross validation. As métricas de avaliação utilizadas foram recall, precisão e área sob a curva Receiver Operating Characteristic. Resultados: Foram utilizados 720 pacientes para criação e validação do modelo. Os modelos apresentaram alta capacidade preditiva com área sob a curva Receiver Operating Characteristic de 0,979; 0,999; 0,980; 0,998 e 1,00 para os algoritmos de Decision Tree, Random Forest, AdaBoost, Gradient Boosting e XGBoost, respectivamente. Conclusão: O modelo preditivo criado e validado apresentou elevada capacidade de predição do choque séptico e hipovolêmico desde o momento da admissão de pacientes na unidade de terapia intensiva.

ABSTRACT Objective: To create and validate a model for predicting septic or hypovolemic shock from easily obtainable variables collected from patients at admission to an intensive care unit. Methods: A predictive modeling study with concurrent cohort data was conducted in a hospital in the interior of northeastern Brazil. Patients aged 18 years or older who were not using vasoactive drugs on the day of admission and were hospitalized from November 2020 to July 2021 were included. The Decision Tree, Random Forest, AdaBoost, Gradient Boosting and XGBoost classification algorithms were tested for use in building the model. The validation method used was k-fold cross validation. The evaluation metrics used were recall, precision and area under the Receiver Operating Characteristic curve. Results: A total of 720 patients were used to create and validate the model. The models showed high predictive capacity with areas under the Receiver Operating Characteristic curve of 0.979; 0.999; 0.980; 0.998 and 1.00 for the Decision Tree, Random Forest, AdaBoost, Gradient Boosting and XGBoost algorithms, respectively. Conclusion: The predictive model created and validated showed a high ability to predict septic and hypovolemic shock from the time of admission of patients to the intensive care unit.

Deep learning model-assisted detection of kidney stones on computed tomography

ABSTRACT Introduction: The aim of this study was to investigate the success of a deep learning model in detecting kidney stones in different planes according to stone size on unenhanced computed tomography (CT) images. Materials and Methods: This retrospective study included 455 patients who underwent CT scanning for kidney stones between January 2016 and January 2020; of them, 405 were diagnosed with kidney stones and 50 were not. Patients with renal stones of 0-1 cm, 1-2 cm, and >2 cm in size were classified into groups 1, 2, and 3, respectively. Two radiologists reviewed 2,959 CT images of 455 patients in three planes. Subsequently, these CT images were evaluated using a deep learning model. The accuracy rate, sensitivity, specificity, and positive and negative predictive values of the deep learning model were determined. Results: The training group accuracy rates of the deep learning model were 98.2%, 99.1%, and 97.3% in the axial plane; 99.1%, 98.2%, and 97.3% in the coronal plane; and 98.2%, 98.2%, and 98.2% in the sagittal plane, respectively. The testing group accuracy rates of the deep learning model were 78%, 68% and 70% in the axial plane; 63%, 72%, and 64% in the coronal plane; and 85%, 89%, and 93% in the sagittal plane, respectively. Conclusions: The use of deep learning algorithms for the detection of kidney stones is reliable and effective. Additionally, these algorithms can reduce the reporting time and cost of CT-dependent urolithiasis detection, leading to early diagnosis and management.

Una nueva perspectiva en rinosinusitis: EPOS 2020 / A new perspective on rhinosinusitis: EPOS 2020

Resumen EPOS 2020 (European Position Paper on Rhinosinusitis and Nasal Polyps 2020) es una guía clínica desarrollada por un grupo profesionales expertos en el área rinosinusal de la Sociedad Europea de Rinología, que corresponde a la última actualización de sus versiones anteriores (2005, 2007 y 2012). El objetivo principal del documento es entregar recomendaciones claras basadas en la mejor evidencia disponible y algoritmos de manejo concisos para las patologías de rinosinusitis aguda y crónica tanto en adultos como en pacientes pediátricos. Algunas de las novedades más importantes de esta guía, son: nueva clasificación de rinosinusitis crónica en primarias y secundarias, rinosinusitis crónica en pediatría, nuevos conceptos en cirugía sinusal, entre otros. También enfatiza la importancia de manejo multidisciplinario de la patología, incluyendo el autocuidado del paciente, inclusive promoviendo el uso de medicamentos de venta libre, antes del manejo médico en niveles escalonados de atención. El objetivo de esta revisión es dar a conocer de manera resumida el manejo de rinosinusitis aguda y crónica en adultos propuesta en esta guía.

Abstract EPOS 2020 (European Position Paper on Rhinosinusitis and Nasal Polyps 2020) is a clinical guide developed by a group of professional experts in the rhinosinusal area of the European Society of Rhinology, which corresponds to the latest update of its previous versions (2005, 2007 and 2012). The main objective of the document is to bring clear recommendations based on the best available evidence and concise management algorithms for the pathologies of acute and chronic rhinosinusitis in both adults and pediatric patients. Some of the most important novelties of this guide are: new classification of chronic rhinosinusitis in primary and secondary, chronic rhinosinusitis in pediatrics, new concepts in sinus surgery, among others. It also emphasizes the importance of multidisciplinary management of the pathology, including self-care of the patient, promoting the use of over-the-counter medications, before medical management at tiered levels of care. The objective of this review is to present in a summarized way the management of acute and chronic rhinosinusitis in adults proposed in this guide.

Inteligência artificial na fisiologia coronária: onde estamos? / Artificial intelligence in coronary physiology: where do we stand?

O uso da fisiologia coronariana invasiva na seleção de indivíduos para revascularização coronariana foi estabelecido nas orientações atuais para manejo da doença arterial coronariana estável. Em comparação com a angiografia isolada, a fisiologia coronariana provou melhorar os resultados clínicos e a relação custo-efetividade no processo de revascularização. Ensaios controlados randomizados, no entanto, questionaram a eficácia do teste de isquemia na seleção de indivíduos para revascularização. Após uma intervenção coronária percutânea com sucesso angiográfico, 20 a 40% dos pacientes apresentaram angina persistente ou recorrente. A inteligência artificial é definida como o uso de vários algoritmos e conceitos computacionais para realizar tarefas complexas de maneira eficiente. Pode ser classificada em dois tipos: abordagens não supervisionadas e supervisionadas. O aprendizado supervisionado é usado principalmente nas tarefas de regressão e classificação, e nele é realizado um mapeamento otimizado entre variáveis de saída e entrada pareadas para executar as tarefas. Em contraste com isso, o aprendizado não supervisionado funciona de maneira diferente. Nesse aprendizado, os dados das variáveis de saída não estão disponíveis, e outros clusters e relações entre os dados de entrada são descobertos, usando-se vários algoritmos. Para se adquirir uma representação mais abstrata dos dados, a tecnologia de aprendizado profundo que utiliza as redes neurais multicamadas domina o aprendizado artificial atualmente.

The use of invasive coronary physiology to select individuals for coronary revascularization has been established in current guidelines for the management of stable coronary artery disease. Compared to angiography alone, coronary physiology has been proven to improve clinical outcomes and cost-effectiveness in the revascularization process. Randomized controlled trials, however, have questioned the efficacy of ischemia testing in selecting individuals for revascularization. After an angiographically successful percutaneous coronary intervention, 20 to 40% of patients experienced persistent or recurrent angina. Artificial intelligence is defined as the usage of various algorithms and computational concepts to perform the complex tasks in an efficient manner. It can be classified into two types: unsupervised and supervised approaches. Supervised learning is majorly used for the regression and classification tasks, and in this optimized mapping between output variables and paired input is carried out to perform the tasks. In contrast to this, unsupervised learning works in the different manner. In unsupervised learning, output variables data is not available and further clusters and relations between input data are found out by using the various algorithms. To acquire more abstract representation of data, deep learning technology, which uses the multilayer neural networks, dominates the artificial learning nowadays.