Nesta versão da nossa ferramenta, vamos usar o Semantic Scholars: uma ferramenta que busca nas principais bases de dados!
Como é uma ferramenta em desenvolvimento, vem mais truques no futuro. Como exemplo, seria legal poder fazer perguntas aos resultados!
No momento, estamos limitados aos 10 resultados principais.
Exemplo #1
Vamos supor que eu queira estudar inteligência artificial para classificar cobras.
Goal: I want to apply transfer learning to snake classification
A busca pode ser feita em português, somente pode limitar seus resultados.
O mais interessante: o Semantic Scholars, buscador base que estamos usando, não deixa de sugerir preprints. Preprints são excelentes para acelerar o compartilhamento de resultados.
Example #2
Thought: I was wondering whether it is possible to apply transfer learning to snake classification
Detalhe
Você recebe um resumo de tudo!
SnakeFace: a transfer learning based app for snake classification is a paper that demonstrates the ease of building state-of-the-art machine learning models without coding expertise. The authors present three test-of-concept models using deep learning and transfer learning to classify snake species based on images. The results show high accuracy in identifying different species, highlighting the potential for this method in biology. Another paper, Snake Species Classification using Transfer Learning Technique, proposes a deep learning model that uses transfer learning to build a snake species classifier. It combines snake photographic images with geographic location information and achieves classification using a gradient boost classifier. A deep-learning based ensembling approach for snake species classification is described in Deep Learning and Gradient Boosting Ensembles for Classification of Snake Species. The authors employ state-of-the-art models, apply transfer learning, and use geographic metadata information to train a gradient boosting ensemble classifier. The fourth paper, Snake Detection and Classification using Deep Learning, explores machine (deep) learning networks for snake detection and classification in a mobile environment. It compares different Convolutional Neural Networks (CNNs) for transfer learning and discusses the performance differences between offline and online classification. Other papers include A CNN Based Model for Venomous and Non-venomous Snake Classification, Life-Saving APP: Snake Classification 'Venomous and Non-venomous' Using fast.ai Based on Indian Species, Impact of Pretrained Networks For Snake Species Classification, A fine-grained classification method based on self-attention Siamese network, A Fine-Tuned BERT-Based Transfer Learning Approach for Text Classification, and Transfer learning for medical image classification: a literature review. These papers cover various aspects of snake classification using transfer learning, deep learning models, and different techniques to improve accuracy and efficiency. (Pires, 2021; Desingu et al., 2021; Mirunalini et al., 2021; Yang & Sinnott, 2021; Ilma et al., 2021; Lakshmi et al., 2021; Krishnan, 2021; Can et al., 2021; Qasim, Bangyal, & Almazroi, 2021; Kim et al., 2021)
I would like to study mean stack in healthcare
The paper titled "Galaxy and MEAN Stack to Create a User-Friendly Workflow for the Rational Optimization of Cancer Chemotherapy" by Pires et al. (2021) describes the development of an online platform that uses MEAN stack and Galaxy pipeline to identify specific targets for cancer therapy. The platform translates RNA-seq data into protein targets for chemotherapy, and the identified targets are stored in MongoDB. The platform can be accessed through a user-friendly interface compatible with mobile devices. In the paper titled "Healthcare professionals' perceptions of risk in the context of genetic testing for the prediction of chronic disease: a qualitative metasynthesis" by Falahee et al. (2021), the authors conducted a qualitative metasynthesis of healthcare professionals' perceptions of genetic risk in the context of predictive genetic testing for chronic disease. The study revealed that healthcare professionals had reservations about the utility of predictive testing and perceived patients' understanding of risk information to be limited and influenced by sociocultural factors. They also expressed concerns about the psychosocial impact of genetic risk information and emphasized the need for multidisciplinary support. The paper titled "Stack-Size Sensitive On-Chip Memory Backup for Self-Powered Nonvolatile Processors" by Zhao et al. (2021) proposes a method to reduce the amount of data that needs to be backed up during a power failure in wearable devices. The authors analyze the application program and identify efficient backup positions based on the variation of stack size along program execution. The evaluation results show a significant reduction in nonvolatile stack size with minimal storage overhead, improving energy utilization and program forward progress compared to instant backup. In the pilot randomized clinical trial titled "Potential reversal of epigenetic age using a diet and lifestyle intervention" by Fitzgerald et al. (2021), the researchers assessed the effects of a diet and lifestyle intervention on biological aging. The intervention included diet, sleep, exercise, relaxation guidance, and supplementation. The results showed a decrease in epigenetic age in the treatment group compared to controls. Changes in blood biomarkers were also observed. However, larger-scale and longer duration clinical trials are needed to confirm these findings. The study titled "A multi-institutional analysis of a general pelvis continuous Hounsfield unit synthetic CT software for radiotherapy" by Yu et al. (2021) aimed to validate a synthetic computed tomography (sCT) software for MRI-only workflow in radiotherapy. The software generated sCTs with continuous Hounsfield units and large field-of-view coverage for general pelvis anatomy. The evaluation showed that the sCTs were realistic and enabled MRI-only planning. The study highlights the potential of this software in improving the workflow of radiotherapy for general pelvis anatomy. In the paper titled "Clinical and virologic factors associated with outcomes of COVID-19 before and after vaccination among Veterans" by Lee et al. (2021), the authors conducted a retrospective analysis to characterize the clinical, demographic, and vaccination factors affecting COVID-19 outcomes in Veterans. They also analyzed the viral epidemiology in the study population. The results showed that age and chronic kidney disease were significant factors associated with hospitalization, regardless of vaccination status. The Delta variant of SARS-CoV-2 was found to be predominant in post-vaccination infections. The paper titled "Actinfo: Information Platform for Physical Activity" by Carreira (2021) describes the development of Actinfo, an information platform for the management of physical activity data. The platform uses MEAN stack technologies and provides visualizations of relevant statistics from physical activity studies. It also allows for the comparison of physical activity data from different studies. The data in Actinfo is modelled after the FHIR standard for healthcare information exchange, ensuring interoperability with clinical data. A comparative study validated the accuracy of the data processing tools implemented in Actinfo. The paper titled "Genetic Knowledge Among Participants in the Coriell Personalized Medicine Collaborative" by Schmidlen et al. (year) investigated the genetic knowledge of participants in the Coriell Personalized Medicine Collaborative. The study aimed to assess participants' understanding of genetics and genetic testing. The findings revealed varying levels of genetic knowledge among the participants, highlighting the need for improved education and communication about genetics in personalized medicine. The study titled "Pituitary Magnetic Resonance Imaging in the Postoperative Follow-Up of Patients with Acromegaly, Less Is More!" by Fernandes et al. (year) examined the use of magnetic resonance imaging (MRI) in the postoperative follow-up of patients with acromegaly. The results showed that routine serial MRI imaging may not be necessary for patients receiving pituitary-targeted therapies. Instead, close monitoring of biochemical markers and selective use of MRI may be more appropriate for detecting tumor growth and assessing treatment response. The feasibility study titled "Predictors of right ventricular remodeling in reperfused inferior myocardial infarctions" by Garg et al. (year) aimed to investigate the predictors of right ventricular (RV) remodeling in patients with reperfused inferior myocardial infarction. The study utilized cardiac magnetic resonance imaging (CMR) and voxel feature tracking to analyze RV function. The results showed that certain CMR-derived parameters, such as peak longitudinal strain (PLS) and time to peak longitudinal strain rate (TTP LSR) of the RV and right atrium (RA), correlated with RV ejection fraction and RV remodeling. Further studies are needed to explore the clinical implications of these findings.
Comments