RESUMO
Prostate cancer is the second most frequent cancer in men worldwide after lung cancer. Its diagnosis is based on the identification of the Gleason score that evaluates the abnormality of cells in glands through the analysis of the different Gleason patterns within tissue samples. The recent advancements in computational pathology, a domain aiming at developing algorithms to automatically analyze digitized histopathology images, lead to a large variety and availability of datasets and algorithms for Gleason grading and scoring. However, there is no clear consensus on which methods are best suited for each problem in relation to the characteristics of data and labels. This paper provides a systematic comparison on nine datasets with state-of-the-art training approaches for deep neural networks (including fully-supervised learning, weakly-supervised learning, semi-supervised learning, Additive-MIL, Attention-Based MIL, Dual-Stream MIL, TransMIL and CLAM) applied to Gleason grading and scoring tasks. The nine datasets are collected from pathology institutes and openly accessible repositories. The results show that the best methods for Gleason grading and Gleason scoring tasks are fully supervised learning and CLAM, respectively, guiding researchers to the best practice to adopt depending on the task to solve and the labels that are available.
Assuntos
Aprendizado Profundo , Gradação de Tumores , Neoplasias da Próstata , Humanos , Neoplasias da Próstata/patologia , Neoplasias da Próstata/diagnóstico por imagem , Masculino , Algoritmos , Interpretação de Imagem Assistida por Computador/métodosRESUMO
Rheumatoid Arthritis (RA) has a mortality rate 1,3 to 3 times higher than the general population, with cardiovascular mortality accounting for 40-50% of cases. Currently, cardiovascular disease is considered an extraarticular manifestation of RA (OR: 1,5-4,0). Ultrasound measurement of the intima-media thickness (IMT) of the common carotid artery and the presence of atherosclerotic plaques (AP) is a non-invasive method and a surrogate marker of subclinical arteriosclerosis. OBJECTIVE: To determine if subclinical arteriosclerosis findings through carotid ultrasound can serve as a good predictor of cardiovascular events (CVE) development in a cohort of RA patients over a 10-year period. METHODOLOGY: A cohort of RA patients seen in the Rheumatology outpatient clinic of a hospital in Castilla La Mancha in 2013 was evaluated. A prospective evaluation for the development of CVE over the following 10 years was conducted, and its correlation with previous ultrasound findings of IMT and AP was analyzed. RESULTS: Eight (24%) patients experienced a CVE. Three (9%) had heart failure, three (9%) had a stroke, and two (6%) experienced acute myocardial infarction. RA patients who developed a CVE had a higher IMT (0,97 +/- 0.08â¯mm) compared to the RA patients without CV complications (0,74 +/- 0.15â¯mm) (pâ¯=â¯0,003). The presence of IMTâ¯≥â¯0.9â¯mm and AP had a relative risk of 12,25 (pâ¯=â¯0,012) and 18,66 (pâ¯=â¯0,003), respectively, for the development of a CVE. CONCLUSIONS: Carotid ultrasound in RA patients may allow for early detection of subclinical atherosclerosis before the development of CVE, with IMTâ¯≥â¯0.9â¯mm being the most closely associated finding with CVE, unaffected by age.
Assuntos
Artrite Reumatoide , Aterosclerose , Doenças Cardiovasculares , Humanos , Espessura Intima-Media Carotídea , Fatores de Risco , Artrite Reumatoide/complicações , Artrite Reumatoide/diagnóstico por imagem , Aterosclerose/complicações , Doenças Cardiovasculares/diagnóstico por imagem , Doenças Cardiovasculares/etiologiaRESUMO
OBJECTIVES: To investigate the association between hospitalization for COVID-19 and oral changes, and to evaluate whether oral changes can indicate a higher risk of disease progression to death. MATERIALS AND METHODS: This case-control study analyzed patients hospitalized (university hospital), including those in intensive care unit and clinical wards. The study group comprised 69 COVID-19 positive patients (PCR-test), while the control group included 43 COVID-19 negative patients. A dentist performed oral evaluations, and salivary samples were collected for calcium, phosphatase, and pH analysis. Sociodemographic data, hospitalization information, and hematological test results were collected from electronic-medical records. The presence of oral changes was assessed using chi-square tests, and the predicted risk of death was analyzed using binary logistic regression. RESULTS: COVID-19 positive patients had a significantly higher prevalence of oral changes compared to COVID-19 negative patients. The presence of any oral changes in COVID-19 positive patients indicated a 13-fold higher risk of mortality. "Bleeding ulcers," "pressure ulcers," and "angular cheilitis" were significantly associated with hospitalization for COVID-19. CONCLUSION: There may be an association between hospitalization for COVID-19 and the development of oral changes, including bleeding ulcers, pressure ulcers. and angular cheilitis. These oral changes may serve as potential indicator for disease progression an increased risk of death. CLINICAL RELEVANCE: COVID-19 hospitalized patients have a higher prevalence of oral changes, which indicate an increased risk of mortality. Oral medicine staff should be included in multidisciplinary teams to detect and treat these oral changes promptly.
Assuntos
COVID-19 , Manifestações Bucais , Úlceras Orais , Queilite , COVID-19/complicações , COVID-19/epidemiologia , Hospitalização , Úlceras Orais/epidemiologia , Pacientes Internados , Prevalência , Progressão da Doença , Estudos de Casos e Controles , Brasil/epidemiologia , Humanos , Masculino , Feminino , Pessoa de Meia-Idade , Idoso , Idoso de 80 Anos ou maisRESUMO
BACKGROUND: The distinction between in situ melanoma (MIS) and invasive melanoma is challenging even for expert dermatologists. The use of pretrained convolutional neural networks (CNNs) as ancillary decision systems needs further research. AIM: To develop, validate and compare three deep transfer learning (DTL) algorithms to predict MIS vs. invasive melanoma and melanoma with a Breslow thickness (BT) of < 0.8â mm vs. ≥ 0.8â mm. METHODS: A dataset of 1315 dermoscopic images of histopathologically confirmed melanomas was created from Virgen del Rocio University Hospital and open repositories of the International Skin Imaging Collaboration archive and Polesie S et al. (Dermatol Pract Concept 2021; 11:e2021079). The images were labelled as MIS or invasive melanoma and < 0.8â mm or ≥ 0.8â mm of BT. We conducted three trainings, and overall means for receiver operating characteristic (ROC) curves, sensitivity, specificity, positive and negative predictive value, and balanced diagnostic accuracy outcomes were evaluated on the test set with ResNetV2, EfficientNetB6 and InceptionV3. The results of 10 dermatologists were compared with the algorithms. Grad-CAM gradient maps were generated, highlighting relevant areas considered by the CNNs within the images. RESULTS: EfficientNetB6 achieved the highest diagnostic accuracy for the comparison between MIS vs. invasive melanoma (61%) and BT < 0.8â mm vs. ≥ 0.8â mm (75%). For the BT comparison, ResNetV2 with an area under the ROC curve of 0.76 and InceptionV3 with an area under the ROC curve of 0.75, outperformed the results obtained by the dermatologist group with an area under the ROC curve of 0.70. CONCLUSION: EfficientNetB6 recorded the best prediction results, outperforming the dermatologists for the comparison of 0.8â mm of BT. DTL could be an ancillary aid to support dermatologists' decisions in the near future.
Assuntos
Melanoma , Neoplasias Cutâneas , Humanos , Dermatologistas , Dermoscopia/métodos , Neoplasias Cutâneas/diagnóstico , Melanoma/diagnóstico , Algoritmos , Aprendizado de Máquina , Melanoma Maligno CutâneoRESUMO
Computational pathology targets the automatic analysis of Whole Slide Images (WSI). WSIs are high-resolution digitized histopathology images, stained with chemical reagents to highlight specific tissue structures and scanned via whole slide scanners. The application of different parameters during WSI acquisition may lead to stain color heterogeneity, especially considering samples collected from several medical centers. Dealing with stain color heterogeneity often limits the robustness of methods developed to analyze WSIs, in particular Convolutional Neural Networks (CNN), the state-of-the-art algorithm for most computational pathology tasks. Stain color heterogeneity is still an unsolved problem, although several methods have been developed to alleviate it, such as Hue-Saturation-Contrast (HSC) color augmentation and stain augmentation methods. The goal of this paper is to present Data-Driven Color Augmentation (DDCA), a method to improve the efficiency of color augmentation methods by increasing the reliability of the samples used for training computational pathology models. During CNN training, a database including over 2 million H&E color variations collected from private and public datasets is used as a reference to discard augmented data with color distributions that do not correspond to realistic data. DDCA is applied to HSC color augmentation, stain augmentation and H&E-adversarial networks in colon and prostate cancer classification tasks. DDCA is then compared with 11 state-of-the-art baseline methods to handle color heterogeneity, showing that it can substantially improve classification performance on unseen data including heterogeneous color variations.
Assuntos
Recém-Nascido Prematuro , Técnicas de Reprodução Assistida , Feminino , Humanos , Idoso , Recém-Nascido , Porto Rico , HermenêuticaRESUMO
OBJECTIVE: The s100b inflammatory protein is involved in schizophrenia pathophysiology. We aim at studying the evolution of the s100b serum levels in acutely relapsed paranoid schizophrenia patients at three different time points (admission, discharge and 3 months after hospital discharge 3MAHD). METHODS: Twenty-three paranoid schizophrenia inpatients meeting DSM-IV criteria participated in the research. Twenty-three healthy subjects matched by age, gender and season acted as the control group. Psychopathology was measured with the Positive and Negative Syndrome Scale (PANSS). Serum s100b levels were determined at 12:00 and 24:00 h with an enzyme-linked immunoassay kit. RESULTS: Patients had significant higher serum s100b levels at admission and discharge (12:00 h) than the group of healthy subjects. At admission and discharge, s100b serum levels at 24 h had decreased compared to the 24:00 h s100b levels of the healthy subjects. At 3MAHD patients and healthy subjects had similar levels of serum s100b protein. Positive and negative PANSS scores decreased significantly between admission and discharge. Positive and negative PANSS scores decreased between discharge and 3MAHD, but these changes had no statistical significance. CONCLUSIONS: Our study confirms that the acute inflammatory response produced in acutely relapsed patients is reversed after 3 month of hospital discharge. The variations of serum s100b concentrations when the patients suffer from an acute relapse may be a useful predictor of disease evolution.
Assuntos
Hospitalização , Esquizofrenia Paranoide , Humanos , Esquizofrenia Paranoide/diagnóstico , Subunidade beta da Proteína Ligante de Cálcio S100 , Biomarcadores , InflamaçãoRESUMO
Introduction: Circulatory system diseases are one of the leading causes of death. Periodontal health is a component of oral and general health that contributes to quality of life. Objective: To analyze periodontal conditions and quality of life in patients with circulatory diseases. Material and method: Cross-sectional study, with patients with circulatory diseases (n=125). The clinical periodontal examination was performed considering: biofilm, bleeding on probing, suppuration, probing depth, gingival recession, and clinical attachment level. Quality of life was assessed with the Medical Outcomes Study 36 - Item Short-Form Health Survey (SF-36) and Oral Health Impact Profile - short form (OHIP-14). Data were analyzed with parametric and nonparametric tests (α=5%). Result: Elderly and men had a greater severity of periodontal disease. The Elderly had lower values in the domains of "Functional capacity" and "Physical aspects" (SF-36). The domain "Pain" had lower values in subjects with health/gingivitis. OHIP-14 did not show differences in the comparisons. There was no correlation between quality of life and the severity of periodontal disease. Conclusion: Among patients with circulatory diseases, the elderly and men had greater severity of periodontal disease. The elderly had a poorer quality of life (functional capacity and physical aspects). There was no correlation between the severity of periodontal disease and quality of life.
Introdução: As doenças do aparelho circulatório representam uma das principais causas de mortes. A saúde periodontal é um componente da saúde bucal e saúde geral que contribui para a qualidade de vida. Objetivo: Analisar as condições periodontais e a qualidade de vida em pacientes com doenças circulatórias. Material e método: Estudo transversal, com pacientes com doenças circulatórias (n=125). O exame clínico periodontal foi realizado considerando: biofilme, sangramento à sondagem, supuração, profundidade de sondagem, recessão gengival e nível de inserção clínica. A qualidade de vida foi avaliada com o Medical Outcomes Study 36 - Item Short-Form Health Survey (SF-36) e Oral Health Impact Profile - short form (OHIP-14). Os dados foram analisados com testes paramétricos e não paramétricos (α=5%). Resultado Idosos e os homens tiveram maior severidade da doença periodontal. Idosos apresentaram menores valores nos domínios "Capacidade funcional" e "Aspectos físicos" (SF-36). O domínio "Dor" teve menor valor nos indivíduos com saúde/gengivite. OHIP-14 não apresentou diferenças nas comparações. Não houve correlação entre qualidade de vida a severidade da doença periodontal. Conclusão Entre os pacientes com doenças circulatórias, os idosos e os homens tiveram maior severidade da doença periodontal. Idosos tiveram pior qualidade de vida (capacidade funcional e aspectos físicos). Não houve correlação entre a severidade da doença periodontal e a qualidade de vida.
Assuntos
Doenças Periodontais , Qualidade de Vida , Doenças Cardiovasculares , Sistema Cardiovascular , Saúde Bucal , Placa Dentária , Análise de Variância , Estatísticas não ParamétricasRESUMO
BACKGROUND: Chronic graft versus host disease (cGVHD) simulating eosinophilic fasciitis (EF) is an underdiagnosed and challenging complication due to the lack of knowledge about its pathogenesis, refractoriness to traditional immunosuppressive agents and their negative impact on the physical function and quality of life. The aim of this study is to describe the clinical-biological characteristics and response to treatment of a case series and to provide a comprehensive literature review on cGVHD related EF involvement. METHODS: Prospective observational study to describe the clinical and diagnostic evaluation characteristics of patients with EF-like follow-up as part of our multidisciplinary cGVHD consultations. In addition, the literature on joint and/or fascial musculoskeletal manifestations due to cGVHD was comprehensively reviewed. RESULTS: 118 patients were evaluated in multidisciplinary cGVHD consultations, 39 of whom (33%) developed fasciitis. Notably, 11 patients had isolated joint contractures without sclerotic skin. After a median of three lines of treatment, the vast majority of patients achieved some degree of response. 94 potentially eligible articles were identified by the search strategy, with 17 of them, the majority isolated case reports, making the final selection. The validated staging scales used for the assessment were the Joint and Fascial Score and the Photographic Range of Motion. CONCLUSION: Fascial/articular involvement needs to be recognized and evaluated early. To our knowledge, our cohort is the second largest series to have been reported. Literature addressing fascial/joints complications related to cGVHD is scarce. The search for new biomarkers, the use of advanced imaging techniques and multidisciplinary approach may help improve the prognosis of patients with cGVHD.
Assuntos
Fasciite , Doença Enxerto-Hospedeiro , Transplante de Células-Tronco Hematopoéticas , Eosinofilia , Fasciite/diagnóstico , Fasciite/etiologia , Doença Enxerto-Hospedeiro/etiologia , Doença Enxerto-Hospedeiro/patologia , Doença Enxerto-Hospedeiro/terapia , Humanos , Estudos Observacionais como Assunto , Qualidade de VidaRESUMO
Crohn's disease and ulcerative colitis are characterized by chronic inflammatory processes and an imbalanced immune response along the gastrointestinal (GI) tract. Pharmacological treatments have been widely used, although their long-term application has adverse side effects. On the other hand, milks fermented with specific lactic acid bacteria (LAB) have been shown to be useful as alternative or complementary aids. Many metabolites such as peptides, exopolysaccharides, and short-chain fatty acids are produced during milk fermentation. These components have been shown to change the pH of the gastrointestinal lumen, aid intestine mucosal recovery, modulate the microbiota, and reduce the inflammatory response (innate and adaptive immune system), both in vitro and in vivo. Therefore, the objective of the present review is to describe how these bioactive compounds from fermented milk by specific LAB can decrease the deleterious symptoms of inflammatory bowel disease.
Assuntos
Doenças Inflamatórias Intestinais , Leite , Animais , Ácidos Graxos Voláteis/metabolismo , Fermentação , Humanos , Doenças Inflamatórias Intestinais/tratamento farmacológico , Doenças Inflamatórias Intestinais/metabolismo , Leite/metabolismo , Leite/microbiologia , PeptídeosRESUMO
This study aimed to assess the potential antidepressant- and anxiolytic-like effects of huauzontle fermented by Lactiplantibacillus plantarum Lp22. The possible association between oxidative stress/inflammation biomarkers and unconditional behavioural tests was also evaluated. Red light-induced stress mice C57Bl/6 (n = 5 per group) received orally either fermented or unfermented huauzontle, diazepam or fluoxetine. A non-stressed group which received saline solution was also included. Then, anxiety-related and depression-related behaviour tests were performed; after that, blood and tissues samples were collected to determine oxidative stress/inflammation biomarkers. The mice receiving both fermented and unfermented huauzontle spent more time (94 s) in open arms in the elevated plus maze test p < 0.05; besides, travelled longer distance (p < 0.05) and increased by more than 50% the exploration time for the open field, as well as the time spent in the illuminated zone (197 s) in the light/dark test. Furthermore, reduced immobility time in the tail suspension and forced swim tests (23.1 and 15.85, respectively), and anhedonia was no detected in the sucrose preference test. The oxidative stress index was lower in the liver of fermented huauzontle-treated mice, while enhanced levels of IL-10, MCP-1 and BDNF in plasma, and lipoxygenase (LOX) activity in the hippocampus were found. Finally, PCA revealed a positive correlation among LOX and BDNF and parameters determined in the anxiety tests, as between catalase activity and immobility time in the depression test. These findings indicate the novel potential therapeutic applications of fermented huauzontle on depression and anxiety-like behaviours possibly mediated by antioxidant and anti-inflammatory mechanisms.
RESUMO
BACKGROUND: Posaconazole is an antifungal agent extensively used as a prophylaxis for invasive fungal infections (IFIs) in allogeneic stem cell transplant (SCT) recipients. Low posaconazole concentrations have been associated with reduced clinical response. The aim of this study was to develop a population pharmacokinetic (popPK) model of a posaconazole tablet formulation in allogeneic SCT adult recipients for supporting model-informed precision dosing (MIPD). MATERIALS AND METHOD: Prospective observational study performed in adult allogeneic SCT recipients receiving posaconazole as prophylaxis for IFIs and followed up by a therapeutic drug monitoring (TDM) program. Posaconazole plasma concentrations were quantified using an ultra-high-performance liquid chromatography (UPLC) with UV detector. A popPK model was developed using NONMEM v.7.4.0. Deterministic and stochastic simulations were carried out with the final model to evaluate the differences across physiological variables with impact on drug exposure. RESULTS: A one-compartment model with sequential absorption (zero and first order) and first order elimination described adequately 55 posaconazole concentrations from 36 patients. Higher doses of posaconazole were found to be required by males and patients with lower values of total serum proteins. A nomogram to estimate the posaconazole daily dose based on pharmacokinetic/pharmacodynamic (PKPD) criterion for males and females for different values of total proteins was developed. CONCLUSIONS: Gender and total serum proteins have been identified as covariates influencing posaconazole CL/F in adult allogeneic SCT recipients receiving the delay-released tablet formulation. Additional studies are required to better characterize the absorption of posaconazole and implications on dosage recommendations together with potential safety concerns.
Assuntos
Transplante de Células-Tronco Hematopoéticas , Triazóis , Antifúngicos/uso terapêutico , Feminino , Humanos , Masculino , ComprimidosRESUMO
Abstract Background: Chronic graft versus host disease (cGVHD) simulating eosinophilic fasciitis (EF) is an underdiagnosed and challenging complication due to the lack of knowledge about its pathogenesis, refractoriness to traditional immunosuppressive agents and their negative impact on the physical function and quality of life. The aim of this study is to describe the clinical-biological characteristics and response to treatment of a case series and to provide a comprehensive literature review on cGVHD related EF involvement. Methods: Prospective observational study to describe the clinical and diagnostic evaluation characteristics of patients with EF-like follow-up as part of our multidisciplinary cGVHD consultations. In addition, the literature on joint and/or fascial musculoskeletal manifestations due to cGVHD was comprehensively reviewed. Results: 118 patients were evaluated in multidisciplinary cGVHD consultations, 39 of whom (33%) developed fasciitis. Notably, 11 patients had isolated joint contractures without sclerotic skin. After a median of three lines of treatment, the vast majority of patients achieved some degree of response. 94 potentially eligible articles were identified by the search strategy, with 17 of them, the majority isolated case reports, making the final selection. The validated staging scales used for the assessment were the Joint and Fascial Score and the Photographic Range of Motion. Conclusion: Fascial/articular involvement needs to be recognized and evaluated early. To our knowledge, our cohort is the second largest series to have been reported. Literature addressing fascial/joints complications related to cGVHD is scarce. The search for new biomarkers, the use of advanced imaging techniques and multidisciplinary approach may help improve the prognosis of patients with cGVHD.
RESUMO
This study aims to analyze the chemical and microbial composition and characterize volatile compounds from the artisanal and commercial Tejuino beverage. For this, eight samples are analyzed (four artisanal and four commercial). The chemical and microbiological quality is determined by standard methods, and volatile compounds are determined by solid-phase microextraction. Overall, the physicochemical composition and microbiological quality are higher for artisanal Tejuino (p < 0.05). The pH values were 3.20 and 3.62, and 0.76 and 0.46 meq of lactic acid for artisanal and commercial Tejuino, respectively. With volatile compounds analyzed, esters, benzenes, and aldehydes were predominant; meanwhile, ethanol was a volatile compound with the highest concentration for all samples. Saccharomyces cerevisiae and Limosilactobacillus fermentum were identified in artisanal Tejuino; yeasts of the Pichia genera and Lactiplantibacillus plantarum, for commercial Tejuino, and Enterococcus genus were identified in both samples. The characterization of both types of Tejuino allows us to update the information available on this important Mexican beverage. In addition, the isolation of lactic acid bacteria, as representative bacteria of both drinks, offers an area of opportunity to know the potential functionality of these bacteria in traditional fermented products.
RESUMO
Healthy subjects present higher summer than winter S100B protein concentrations. There is no available information regarding if schizophrenia patients present the same pattern. The aim of this research is to study if patients with schizophrenia present seasonal changes in serum S100B concentrations.
Assuntos
Pacientes Internados , Esquizofrenia Paranoide , Humanos , Subunidade beta da Proteína Ligante de Cálcio S100 , Estações do AnoRESUMO
Monitoring animals' behavior living in wild or semi-wild environments is a very interesting subject for biologists who work with them. The difficulty and cost of implanting electronic devices in this kind of animals suggest that these devices must be robust and have low power consumption to increase their battery life as much as possible. Designing a custom smart device that can detect multiple animal behaviors and that meets the mentioned restrictions presents a major challenge that is addressed in this work. We propose an edge-computing solution, which embeds an ANN in a microcontroller that collects data from an IMU sensor to detect three different horse gaits. All the computation is performed in the microcontroller to reduce the amount of data transmitted via wireless radio, since sending information is one of the most power-consuming tasks in this type of devices. Multiples ANNs were implemented and deployed in different microcontroller architectures in order to find the best balance between energy consumption and computing performance. The results show that the embedded networks obtain up to 97.96% ± 1.42% accuracy, achieving an energy efficiency of 450 Mops/s/watt.
Assuntos
Algoritmos , Animais Selvagens , Animais , Comportamento Animal , Fontes de Energia Elétrica , Redes Neurais de ComputaçãoRESUMO
RESUMEN La pandemia por COVID-19 originado por el Coronavirus 2 causante de síndrome respiratorio agudo severo (SARS-CoV-2) es causante de una crisis de salud pública a nivel global. Muchos reportes indican resultados desalentadores en pacientes con cáncer respecto a la población general. Por ello, los expertos en el manejo de neoplasias oncohematológicas del Instituto Nacional de Enfermedades Neoplásicas, hospitales nacionales y una clínica privada de Lima Metropolitana han desarrollado recomendaciones obtenidas por consenso para continuar con el manejo de pacientes con neoplasias oncohematológicas en forma segura ante la coyuntura de pandemia.
ABSTRACT The ongoing COVID-19 pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a global public health crisis. Many reports indicate disappointing results in cancer patients compared to the general population. Therefore, experts in the management of oncohematological malignancies from the National Institute of Neoplastic Diseases, national hospitals and a private clinic in Metropolitan Lima have developed recommendations obtained by consensus to continue with the management of patients with oncohematological neoplasms safely in the face of the pandemic.
RESUMO
Prostate cancer (PCa) is the second most frequently diagnosed cancer among men worldwide, with almost 1.3 million new cases and 360,000 deaths in 2018. As it has been estimated, its mortality will double by 2040, mostly in countries with limited resources. These numbers suggest that recent trends in deep learning-based computer-aided diagnosis could play an important role, serving as screening methods for PCa detection. These algorithms have already been used with histopathological images in many works, in which authors tend to focus on achieving high accuracy results for classifying between malignant and normal cases. These results are commonly obtained by training very deep and complex convolutional neural networks, which require high computing power and resources not only in this process, but also in the inference step. As the number of cases rises in regions with limited resources, reducing prediction time becomes more important. In this work, we measured the performance of current state-of-the-art models for PCa detection with a novel benchmark and compared the results with PROMETEO, a custom architecture that we proposed. The results of the comprehensive comparison show that using dedicated models for specific applications could be of great importance in the future.
Assuntos
Aprendizado Profundo , Detecção Precoce de Câncer , Neoplasias da Próstata , Algoritmos , Humanos , Masculino , Redes Neurais de Computação , Antígeno Prostático Específico , Neoplasias da Próstata/diagnósticoRESUMO
BACKGROUND: Previous studies have demonstrated the acrylamide-removing properties of probiotic monocultures; however, potential advantages of consortia over monocultures in reducing the dietary exposure to acrylamide have not been proven. Hence this work aims to assess the acrylamide (AA)-binding properties of bacterial consortia, consisting of either probiotic strains and / or representative bacteria of duodenal microbiota, exposed to simulated gastrointestinal conditions (SGC). The AA binding capacity of ten probiotic strains (PS) and six duodenal strains (NDS) was evaluated under different conditions; then, three different consortia (PS, NDS, and PS + NDS) were assessed under SGC. RESULTS: Among individual PS, Bacillus coagulans GBI-30, Lactobacillus fermentum J23, L. pentosus J37 and J24, and L. casei Shirota, exhibited the highest AA-binding capacity (80-87%), while Bifidobacterium catenulatun ATCC27676, Streptococcus salivarius subsp. thermophilus ATCC19258, and S. gallolyticus ATCC9809 were the best (ca. 68%) NDS monocultures. Probiotic strain consortia showed higher (P < 0.05) AA binding capacity (> 90%) than monoculture bacteria. Conversely, individual NDS cultures displayed higher (P < 0.05) binding capacity than NDS consortia (60%). A significant reduction (P < 0.05) in AA removal capacity was observed when consortia were exposed to SGC, PS consortia being the most effective (> 60% removal). CONCLUSION: These results suggest that consortia of specific PS could play an important role in reducing the intestinal availability of acrylamide. © 2021 Society of Chemical Industry.