Search

Introduction: Previous studies have reported differences in the management and outcome of women stroke patients in comparison with men. We aim to analyze sex and gender differences in the medical assistance, access to treatment and outcome of acute stroke patients in Catalonia. Patients and methods: Data were obtained from a prospective population-based registry of stroke code activations in Catalonia (CICAT) from January/2016 to December/2019. The registry includes demographic data, stroke severity, stroke subtype, reperfusion therapy, and time workflow. Centralized clinical outcome at 90 days was assessed in patients receiving reperfusion therapy. Results: A total of 23,371 stroke code activations were registered (54% men, 46% women). No differences in prehospital time metrics were observed. Women more frequently had a final diagnosis of stroke mimic, were older and had a previous worse functional situation. Among ischemic stroke patients, women had higher stroke severity and more frequently presented proximal large vessel occlusion. Women received more frequently reperfusion therapy (48.2% vs 43.1%, p Discussion and conclusion: We found some differences by sex in that acute stroke was more frequent in older women and the stroke severity was higher. We found no differences in medical assistance times, access to reperfusion treatment and early complications. Worse clinical outcome at 90 days in women was conditioned by stroke severity and older age, but not by sex itself.

A Síndrome Braquicefálica é o conjunto de alterações anatômicas e fisiológicas que dificultam a passagem de ar nas vias aéreas nos cães de focinho curto. Esses animais têm alterações primárias como estenose de narinas e prolongamento do palato mole, que podem levar às alterações secundárias ao longo do tempo. Cada animal pode apresentar essas alterações com diferentes graus de intensidade. Como sinais clínicos dessa síndrome, pode-se citar ronco acordado e durante o sono, tosse, engasgos, intolerância ao exercício físico, disfagia, espirro reverso, dispneia e cianose. Essas alterações provenientes da síndrome braquicefálica são progressivas e degenerativas. O relato tem como objetivo apresentar as técnicas cirúrgicas de estafilectomia com uso de bisturi ultrassônico e rinoplastia utilizadas para correção da angústia respiratória em canino da raça Pug, aliadas com o diagnóstico e tratamento de alterações respiratórias secundárias pelo exame de laringotraqueobroncoscopia.

© 2022 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY 4.0 licenseBackground: Traumatic brain injury (TBI) is increasingly recognised as being responsible for a substantial proportion of the global burden of disease. Neurosurgical interventions are an important aspect of care for patients with TBI, but there is little epidemiological data available on this patient population. We aimed to characterise differences in casemix, management, and mortality of patients receiving emergency neurosurgery for TBI across different levels of human development. Methods: We did a prospective observational cohort study of consecutive patients with TBI undergoing emergency neurosurgery, in a convenience sample of hospitals identified by open invitation, through international and regional scientific societies and meetings, individual contacts, and social media. Patients receiving emergency neurosurgery for TBI in each hospital's 30-day study period were all eligible for inclusion, with the exception of patients undergoing insertion of an intracranial pressure monitor only, ventriculostomy placement only, or a procedure for drainage of a chronic subdural haematoma. The primary outcome was mortality at 14 days postoperatively (or last point of observation if the patient was discharged before this time point). Countries were stratified according to their Human Development Index (HDI)—a composite of life expectancy, education, and income measures—into very high HDI, high HDI, medium HDI, and low HDI tiers. Mixed effects logistic regression was used to examine the effect of HDI on mortality while accounting for and quantifying between-hospital and between-country variation. Findings: Our study included 1635 records from 159 hospitals in 57 countries, collected between Nov 1, 2018, and Jan 31, 2020. 328 (20%) records were from countries in the very high HDI tier, 539 (33%) from countries in the high HDI tier, 614 (38%) from countries in the medium HDI tier, and 154 (9%) from countries in the low HDI tier. The median age was 35 years (IQR 24–51), with the oldest patients in the very high HDI tier (median 54 years, IQR 34–69) and the youngest in the low HDI tier (median 28 years, IQR 20–38). The most common procedures were elevation of a depressed skull fracture in the low HDI tier (69 [45%]), evacuation of a supratentorial extradural haematoma in the medium HDI tier (189 [31%]) and high HDI tier (173 [32%]), and evacuation of a supratentorial acute subdural haematoma in the very high HDI tier (155 [47%]). Median time from injury to surgery was 13 h (IQR 6–32). Overall mortality was 18% (299 of 1635). After adjustment for casemix, the odds of mortality were greater in the medium HDI tier (odds ratio [OR] 2·84, 95% CI 1·55–5·2) and high HDI tier (2·26, 1·23–4·15), but not the low HDI tier (1·66, 0·61–4·46), relative to the very high HDI tier. There was significant between-hospital variation in mortality (median OR 2·04, 95% CI 1·17–2·49). Interpretation: Patients receiving emergency neurosurgery for TBI differed considerably in their admission characteristics and management across human development settings. Level of human development was associated with mortality. Substantial opportunities to improve care globally were identified, including reducing delays to surgery. Between-hospital variation in mortality suggests changes at an institutional level could influence outcome and comparative effectiveness research could identify best practices. Funding: National Institute for Health Research Global Health Research Group.

Aspects report; Data collection instrument; In-depth interview instructions; and Informed consent document.

Investigators in neuroscience have turned to Big Data to address replication and reliability issues by increasing sample sizes, statistical power, and representativeness of data. These efforts unveil new questions about integrating data arising from distinct sources and instruments. We focus on the most frequently assessed cognitive domain - memory testing - and demonstrate a process for reliable data harmonization across three common measures. We aggregated global raw data from 53 studies totaling N = 10,505 individuals. A mega-analysis was conducted using empirical bayes harmonization to remove site effects, followed by linear models adjusting for common covariates. A continuous item response theory (IRT) model estimated each individual’s latent verbal learning ability while accounting for item difficulties. Harmonization significantly reduced inter-site variance while preserving covariate effects, and our conversion tool is freely available online. This demonstrates that large-scale data sharing and harmonization initiatives can address reproducibility and integration challenges across the behavioral sciences.TeaserWe present a global effort to devise harmonization procedures necessary to meaningfully leverage big data.

This retrospective cross-sectional, observational study aims to determine the correlation between demographic variables and nonsyndromic craniosynostosis.A cross-sectional study was carried out. The inclusion criteria were nonsyndromic craniosynostosis and the exclusion criteria were patients who did not have complete records.The sample included 49 patients and mainly women. Patients with more than 4 days in the intensive care unit were related with more complications. Intracranial Hypertension was seen predominantly in patients between 6 and 8 months P=0007.Some variables can modificate complications risk but future studies are needed to further investigate the influence of such variables on craniofacial care.

BACKGROUND Major depressive disorder (MDD) is the leading cause of global disability; however, the existing treatments do not always address cognitive dysfunction—a core feature of MDD. Immersive virtual reality (VR) has emerged as a promising modality to enhance the real-world effectiveness of cognitive remediation. OBJECTIVE This study aimed to develop the first prototype VR cognitive remediation program for MDD (“bWell-D”). This study gathered qualitative data from end users early in the design process to enhance its efficacy and feasibility in clinical settings. METHODS Semistructured end-user interviews were conducted remotely (n=15 patients and n=12 clinicians), assessing the participants’ perceptions and goals for a VR cognitive remediation program. Video samples of bWell-D were also shared to obtain feedback regarding the program. The interviews were transcribed, coded, and analyzed via thematic analysis. RESULTS End users showed an optimistic outlook toward VR as a treatment modality, and perceived it as a novel approach with the potential of having multiple applications. The participants expressed a need for an engaging VR treatment that included realistic and multisensorial settings and activities, as well as customizable features. Some skepticism regarding its effectiveness was also reported, especially when the real-world applications of the practiced skills were not made explicit, as well as some concerns regarding equipment accessibility. A home-based or hybrid (ie, home and clinic) treatment modality was preferred. CONCLUSIONS Patients and clinicians considered bWell-D interesting, acceptable, and potentially feasible, and provided suggestions to enhance its real-world applicability. The inclusion of end-user feedback is encouraged when developing future VR programs for clinical purposes.

International audience; Objectives: Persistent functional impairment is common in bipolar disorder (BD) and is influenced by a number of demographic, clinical, and cognitive features. The goal of this project was to estimate and compare the influence of key factors on community function in multiple cohorts of well-characterized samples of individuals with BD.Methods: Thirteen cohorts from 7 countries included n=5,882 individuals with BD across multiple sites. The statistical approach consisted of a systematic uniform application of analyses across sites. Each site performed a logistic regression analysis with empirically derived "higher versus lower function" as the dependent variable and selected clinical and demographic variables as predictors.Results: We found high rates of functional impairment, ranging from 41-75%. Lower community functioning was associated with depressive symptoms in 10 of 12 of the cohorts that included this variable in the analysis. Lower levels of education, a greater number of prior mood episodes, presence of a comorbid substance use disorder, and a greater total number of psychotropic medications were also associated with low functioning.Conclusions: The bipolar clinical research community is poised to work together to characterize the multi-dimensional contributors to impairment and address the barriers that impede patients' complete recovery. We must also identify the core features which enable many to thrive and live successfully with BD. A large-scale, worldwide, prospective longitudinal study focused squarely on BD and its heterogeneous presentations will serve as a platform for discovery and promote major advances toward optimizing outcomes for every individual with this illness.

Significant advances documenting the costs associated with insomnia have been achieved. However, those related to insomnia associated with mood disorders remain understudied, even though insomnia is more severe in the presence of comorbid conditions such as depression and anxiety. The aim of this study was to determine the direct and indirect costs of insomnia associated with depression and anxiety disorders (DAD) from the perspective of the patient in a private healthcare system. This was an observational study of chronic insomnia associated with DAD at a private Sleep Disorders Clinic in Mexico City between 2019 and 2020. Patients were followed for up to one year. Healthcare resource utilization data were collected through clinical records. Direct and indirect costs associated with insomnia treatment were estimated through micro-costing. The estimated economic burden was projected to 5 years adjusting for inflation and discounting future costs. A deterministic sensitivity analysis was performed. The median cost of the first year of insomnia treatment associated with DAD was US$3537.57 per patient. The work productivity loss represented the highest economic burden (63.84%) followed by direct medical costs (28.32%), and the direct non-medical costs (7.85%). The estimated annual economic burden for patients treated in the private healthcare system in Mexico was US$293 million. The costs of insomnia associated with DAD at a private clinic in México were found to be high. The burden of the costs faced by these patients is substantial relative to the median income of the population. The economic costs at an individual and societal levels are substantial.The online version contains supplementary material available at 10.1007/s41105-022-00412-6.

Objetivo: Classificar conteúdo textual sobre medicamentos presentes em mídia social. Método: Foi realizado um estudo de caso na mídia social Twitter. Um dicionário foi elaborado para identificar as mensagens publicadas na mídia social com conteúdo textual sobre medicamentos. O dicionário foi construído com informações extraídas do anuário estatístico de medicamentos para o ano de 2017. As informações que compõem o dicionário estão relacionadas as substâncias mais comercializadas e mais faturadas para o período. As mensagens identificadas pelo dicionário referem 185 dias do ano de 2017, publicadas em território brasileiro e em língua portuguesa brasileira. Para o processamento do texto técnicas de mineração de dados e texto foram aplicadas. As mensagens identificadas foram avaliadas de forma quantitativa e qualitativa. Para a análise quantitativa foi empregada análises exploratórias, contagens e análise termo frequência. Para a análise qualitativa foi empregada uma análise de conteúdo e uma análise de tópicos. Resultados: O dicionário elaborado possui 1.006 termos, do tipo princípios ativos e marcas comerciais. No total 238.217 mensagens sobre medicamentos foram identificadas com o uso do dicionário. A análise exploratória indicou as categorias de medicamentos mais encontradas para o corpus. As categorias de medicamentos com maior número de representantes no corpus foram: medicamentos que atuam no sistema nervoso (1), substâncias de venda sob prescrição e isentos de prescrição. Em seguida medicamentos que atuam no trato alimentar e metabolismo (2), substâncias de venda sob prescrição e isentos de prescrição. Anti-infecciosos de uso sistêmico (3), substâncias de venda sob prescrição. A análise termo frequência indicou os termos do dicionário mais frequentes nas mensagens para esse corpus Dorflex (1), Dipirona (2), Paracetamol (3), Omeprazol (4) e Codeína (5). No total 6.213 mensagens foram classificadas por um profissional farmacêutico. A análise de conteúdo descreveu aspectos da linguagem, intento dos usuários e relações de uso sobre os medicamentos na mídia social. Uma parte das mensagens foi associada a conceitos da assistência farmacêutica brasileira. Mensagens com menções a reações adversas à medicamentos, suspeita de ineficácia, erros de medicação foram encontradas. A análise de tópicos foi realizada para os medicamentos Dorflex, Dipirona, Paracetamol, Omeprazol, Codeína, Sais de ferro e Vacina de Febre Amarela. Na análise de tópicos aspectos comuns e incomuns entre as mensagens agrupadas sobre um mesmo termo foram descritas. Conclusão: O dicionário permitiu identificar mensagens sobre medicamentos na mídia social. A mídia social estudo de caso foi utilizada pelos consumidores como plataforma para relatar eventos com medicamentos. Algumas mensagens foram associadas a eventos de risco como as reações adversas, suspeita de ineficácia e erros de medicação. Outras mensagens caracterizaram novos aspectos do consumo de medicamentos, como questões comportamentais e afetivas. As mídias sociais podem ser utilizadas como nova fonte de dados em saúde. Há inúmeras possibilidades de aplicações para a assistência farmacêutica. Descritores: Medicamento, Rede social, Dicionário, Serviços de Assistência Farmacêutica Objective: To classify textual content about drugs present in social media. Method: A case study was carried out on the social media Twitter. A dictionary was created to identify messages published on social media with textual content about drugs. The dictionary was built with information extracted from the statistical yearbook of medicines for the year 2017. The information that makes up the dictionary is related to the most commercialized and most billed substances for the period. The messages identified by the dictionary refer to 185 days in 2017, published in Brazilian territory and in Brazilian Portuguese. For text processing data and text mining techniques were applied. The messages identified were evaluated quantitatively and qualitatively. For the quantitative analysis, exploratory analysis, counting and thermofrequency analysis were used. For the qualitative analysis, a content analysis and a topic analysis were used. Results: The dictionary elaborated has 1,006 terms, such as active principles and trademarks. A total of 238,217 messages about medication were identified using the dictionary. The exploratory analysis indicated the most commonly found medication categories for the corpus. The drug categories with the highest number of representatives in the corpus were: drugs that act on the nervous system (1), prescription and over-the-counter. Then drugs that act on the alimentary tract and metabolism (2), prescription and over-the-counter substances. Systemic anti-infectives (3), prescription drugs. The term frequency analysis indicated the most frequent dictionary terms in the messages for this corpus Dorflex (1), Dipyrone (2), Paracetamol (3), Omeprazole (4) and Codeine (5). A total of 6,213 messages were classified by a pharmaceutical professional. Content analysis described aspects of language, users' intent, and usage relationships about medications on social media. A part of the messages was associated with concepts of Brazilian pharmaceutical care. Messages with mentions of adverse drug reactions, suspected ineffectiveness, medication errors were found. Topical analysis was performed for Dorflex, Dipyrone, Paracetamol, Omeprazole, Codeine, Iron salts and Yellow Fever Vaccine. In the analysis of topics, common and uncommon aspects among the messages grouped under the same term were described. Conclusion: The dictionary made it possible to identify messages about drugs on social media. The social media case study was used by consumers as a platform to report drug events. Some messages were associated with risk eventsix such as adverse reactions, suspected ineffectiveness and medication errors. Other messages featured new aspects of medication consumption, such as behavioral and affective issues. Social media can be used as a new source of health data. There are numerous application possibilities for pharmaceutical care. Keywords: Drugs, Social Media, Dictionary, Pharmaceutical Care

O Teste de Inteligência Digital em Saúde (TIDs) foi desenvolvido para possibilitar uma autoavaliação das competências digitais necessárias para uma atuação profissional qualificada na área de saúde digital de estudantes e trabalhadores da saúde com base nos perfis de literacia digital disponíveis no modelo de inteligência digital proposto pelo DQ Institute de Singapura. Objetivo: O objetivo deste estudo foi realizar uma análise da literacia digital de estudantes e trabalhadores da saúde para identificar os respectivos perfis de competência digital e estabelecer uma escala de valores para sua mensuração. Métodos: Trata-se de um estudo transversal, descritivo e com enfoque quantitativo. Foi realizada revisão integrativa de literatura nas bases indexadas Pubmed, Eric e Portal de Periódicos Capes, identificados os estudos relevantes e selecionado o framework de inteligência digital como base para o desenvolvimento deste estudo. Para atingir o objetivo principal e os objetivos secundários foram realizadas atividades de tradução e adaptação transcultural, do inglês para a língua portuguesa, das competências do relatório de literacia digital do DQ Institute. As competências estavam organizadas em oito áreas de desenvolvimento e nomeadas da seguinte maneira: identidade digital, uso digital, proteção digital, segurança digital, inteligência emocional digital, comunicação digital, literacia digital e direito digital. Cada competência oferecia três perfis de conhecimento: cidadania digital, criatividade digital e competitividade digital, em conformidade com o framework de literacia digital, habilidades e prontidão do DQ Institute. Foi realizada a organização do texto traduzido em formato de tabela para facilitar a inserção das afirmações traduzidas em banco de dados. Uma vez que a base de dados do instrumento estava pronta, foi iniciado o desenvolvimento computacional do TIDs e do relatório de sinastria digital. Na sequência foram realizadas as etapas de avaliação por especialistas, ajustes e aplicação do teste em sua versão Alfa. Após sua aplicação foi realizada a análise dos resultados, discussão sobre os pontos fortes e fracos identificados, verificação dos ajustes necessários para o desenvolvimento e aplicação da versão Beta. Resultados: Uma plataforma digital foi criada no servidor de rede do Departamento de Informática em Saúde da Escola Paulista de Medicina (EPM) na Universidade Federal de São Paulo (UNIFESP) para armazenar o teste nas versões Alfa e Beta. Em sua primeira aplicação foram obtidos 292 respondentes válidos, alunos do curso de especialização em informática em saúde da UAB/UNIFESP, que responderam a 44 afirmações e tiveram acesso ao relatório com a apresentação do percentual global e específico por perfil, bem como a apresentação dos pontos fortes e fracos relacionados às competências de literacia digital. Na versão Beta foram obtidos 79 respondentes que tiveram a possibilidade de responder 56 afirmações relacionadas à saúde digital e tiveram acesso ao relatório de sinastria digital com apresentação de gráficos dinâmicos e um questionário de avaliação dos resultados obtidos. Foi realizada estatística descritiva quantitativa dos percentuais de atingimento global e específicos nas duas versões e a estatística descritiva qualitativa para apresentar a relação entre as variáveis do teste e as respostas obtidas e os agrupamentos por área de atuação. Conclusão: Foi possível atingir os objetivos definidos e realizar a análise da literacia digital dos estudantes e trabalhadores da saúde, estabelecer uma escala de valores e construir a plataforma para aplicar e armazenar os resultados do TIDs. The Digital Health Intelligence Test (TIDs) was developed to enable a self-assessment of the digital skills needed for a qualified professional performance in the digital health area of students and health workers based on the digital literacy profiles available in the digital intelligence model proposed by the DQ Institute of Singapore. Objective: The aim of this study was to perform an analysis of the digital literacy of students and health workers to identify their digital competency profiles and establish a scale of values for their measurement. Methods: This is a cross-sectional, descriptive study with a quantitative focus. An integrative literature review was performed in the indexed databases Pubmed, Eric and Portal of Capes Journals, identifying the relevant studies and selecting the digital intelligence framework as the basis for the development of this study. To achieve the main objective and secondary objectives, cross-cultural translation and adaptation activities were conducted, from English to Portuguese, of the competencies of the digital literacy report of the DQ Institute. The competencies were organized in eight areas of development and named as follows: digital identity, digital use, digital protection, digital security, digital emotional intelligence, digital communication, digital literacy, and digital law. Each competency offered three knowledge profiles: digital citizenship, digital creativity, and digital competitiveness, in accordance with the Digital Literacy framework, skills and readiness of the DQ Institute. The translated text was included in table format to facilitate the insertion of the statements translated into a database. Once the instrument database was ready, the computational development of TIDs and digital sinastria report was initiated. Next, the evaluation steps were performed by specialists, adjustments, and application of the test in its Alpha version. After its application, the results were performed, discussion about the identified strengths and weaknesses, verification of the necessary adjustments for the development and application of the Beta version. Results: A digital platform was created on the network server of the Department of Health Informatics of the Paulista School of Medicine (EPM) at the Federal University of São Paulo (UNIFESP) to store the test in alpha and beta versions. In its first application, 292 valid respondents were obtained, students of the specialization course in health informatics at UAB/UNIFESP, who responded to forty-four statements and had access to the report with the presentation of the overall and specific percentage by profile, as well as the presentation of strengths and weaknesses related to digital literacy skills. In the Beta version, seventy-nine valid respondents were obtained who had the possibility to answer fifty-six statements related to digital health and had access to the digital sinastria report with presentation of dynamic graphs and a questionnaire to evaluate the results obtained. Quantitative descriptive statistics of the percentages of global and specific achievement in both versions and qualitative descriptive statistics were performed to present the relationship between the test variables and the responses obtained and the groupings by area of activity. Conclusion: It was possible to achieve the defined objectives and perform the analysis of the digital literacy of students and health workers, establish a scale of values, and build the platform to apply and store the results of TIDs.

O câncer de próstata é o segundo câncer mais frequente em homens no mundo. A radioterapia apresenta um papel fundamental no tratamento do câncer de próstata. Para planejamento do tratamento inicial, o paciente é classificado em um grupo de risco, podendo ser baixo, intermediário ou alto risco. Entretanto, esses grupos, normalmente baseados em três dados principais coletados durante o diagnóstico (concentração do antígeno prostático específico pré-terapia, soma do escore de Gleason e estágio clínico do tumor), costumam apresentar alta heterogeneidade e, assim, pacientes com diferentes características podem ser classificados em um mesmo grupo de risco. O tratamento excessivo de tumores com pouca probabilidade de progressão e o subtratamento de tumores mais agressivos são consequências disso. As técnicas de Aprendizado de Máquina vêm sendo utilizadas com sucesso em diversas aplicações voltadas à oncologia, incluindo classificação e avaliação de risco. Com isso, o objetivo deste estudo é refinar a metodologia de determinação do grupo de risco de câncer de próstata em pacientes tratados com radioterapia, utilizando métodos de aprendizado não supervisionado. Foram utilizados dados de 485 pacientes com câncer de próstata tratados com radioterapia entre janeiro de 2010 e janeiro de 2017 no Hospital das Clínicas da Faculdade de Medicina de Ribeirão Preto. Oito diferentes algoritmos de agrupamento foram implementados na linguagem de computação Python, sendo eles: K-means, Mini Batch K-means, Affinity Propagation, Agglomerative Clustering, BIRCH, DBSCAN, OPTICS e Agrupamento Profundo com Autoencoder. Três estratégias foram propostas para obter um melhor resultado do refinamento dos grupos de risco e seus resultados foram analisados e comparados utilizado o método de Kaplan-Meier, juntamente com o teste log-rank, e o método da silhueta. Os algoritmos com os melhores desempenhos foram Mini Batch K-means e Autoencoder, apresentando curvas distintas no gráfico de Kaplan-Meier, os menores valores no teste log-rank e valores positivos de coeficiente de silhueta. Foram implementadas árvores de decisões a partir dos grupos resultantes destes algoritmos para tornar explícitos os indicadores mais relevantes e para apresentar conjuntos de regras que permitem entender a lógica de geração dos agrupamentos. Por fim, uma ferramenta foi desenvolvida para que o usuário consiga classificar novos pacientes nos grupos de risco definidos pelos melhores resultados das três estratégias. Conclui-se que este estudo apresenta novas regras de grupos de risco refinadas, baseadas em dados referentes ao tumor e ao paciente, além de oferecer uma maneira mais simples e direta para o médico especialista compreender a formação dos grupos de risco. Prostate cancer is the second most common cancer in men worldwide. Radiotherapy is fundamental for the treatment of prostate cancer. For initial treatment planning, the patient is classified into a risk group, which can be low, intermediate or high risk. However, these groups, usually based on three main data collected during diagnosis (pre-therapy prostate specific antigen concentration, sum of Gleason score and tumor clinical stage), tend to present great heterogeneity and, thus, patients with different characteristics may be classified in the same risk group. Overtreatment of tumors that are unlikely to progress and undertreatment of more aggressive tumors are consequences of this. Machine learning techniques have been used successfully in a variety of oncology applications, including risk assessment and classification. Therefore, the aim of this study is to refine the methodology for determining the risk group for prostate cancer in patients treated with radiotherapy, using unsupervised learning methods. Data from 485 patients with prostate cancer treated with radiotherapy between January 2010 and January 2017 at the Hospital das Clínicas of the Faculty of Medicine of Ribeirão Preto were used. Eight different clustering algorithms were implemented in the Python programming language. The algorithms are: K-means, Mini Batch K-means, Affinity Propagation, Agglomerative Clustering, BIRCH, DBSCAN, OPTICS and Deep Clustering using Autoencoder. Three strategies were proposed to obtain a better result from the refinement of risk groups and their results were analyzed and compared using the Kaplan-Meier method, with the log-rank test, and the silhouette method. The algorithms with the best performances were Mini Batch K-means and Autoencoder, presenting different curves in the Kaplan-Meier graph, the lowest values in the log-rank test and positive silhouette coefficient values. Decision trees were implemented from the groups resulting from these algorithms to make the most relevant indicators explicit and in order to present sets of rules that allow understanding the logic of generating the clusters. Finally, a tool was developed so that the user can classify new patients into the risk groups defined by the best results of the three strategies. It is concluded that this study presents new refined risk group rules, based on tumor and patient data, in addition to offering a friendlier way for the specialist to understand the risk groups.