Trends in scientific output on artificial intelligence and health in Latin America in Scopus

González-Argote, Javier; Alonso-Galbán, Patricia; Castillo, Adrián; Lepez, Carlos Oscar; Castillo‐González, William; Bonardi, Mabel Cecilia; Cano, Carlos Alberto Gómez

doi:10.4108/eetsis.vi.3231

Cited by 3 publications

(1 citation statement)

References 44 publications

(48 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [1], trends in scientific production on artificial intelligence and health in Latin America in Scopus was explored. The conclusion was that main topics were predictive models and the application of artificial intelligence for classifying, diagnosing and treating diseases .…”

Section: Introductionmentioning

confidence: 99%

Fast Lung Image Segmentation Using Lightweight VAEL-Unet

Hao,

Zhang,

2024

ICST Transactions on Scalable Information Systems

View full text Add to dashboard Cite

INTRODUCTION: A lightweght lung image segmentation model was explored. It was with fast speed and low resouces consumed while the accuracy was comparable to those SOAT models. OBJECTIVES: To improve the segmentation accuracy and computational efficiency of the model in extracting lung regions from chest X-ray images, a lightweight segmentation model enhanced with a visual attention mechanism called VAEL-Unet, was proposed. METHODS: Firstly, the bneck module from the MobileNetV3 network was employed to replace the convolutional and pooling operations at different positions in the U-Net encoder, enabling the model to extract deeper-level features while reducing complexity and parameters. Secondly, an attention module was introduced during feature fusion, where the processed feature maps were sequentially fused with the corresponding positions in the decoder to obtain the segmented image. RESULTS: On ChestXray, the accuracy of VAEL-Unet improves from 97.37% in the traditional U-Net network to 97.69%, while the F1-score increases by 0.67%, 0.77%, 0.61%, and 1.03% compared to U-Net, SegNet, ResUnet and DeepLabV3+ networks. respectively. On LUNA dataset. the F1-score demonstrates improvements of 0.51%, 0.48%, 0.22% and 0.46%, respectively, while the accuracy has increased from 97.78% in the traditional U-Net model to 98.08% in the VAEL-Unet model. The training time of the VAEL-Unet is much less compared to other models. The number of parameters of VAEL-Unet is only 1.1M, significantly less than 32M of U-Net, 29M of SegNet, 48M of Res-Unet, 5.8M of DeeplabV3+ and 41M of DeepLabV3Plus_ResNet50. CONCLUSION: These results indicate that VAEL-Unet’s segmentation performance is slightly better than other referenced models while its training time and parameters are much less.

show abstract

Section: Introductionmentioning

confidence: 99%

Fast Lung Image Segmentation Using Lightweight VAEL-Unet

Hao,

Zhang,

2024

ICST Transactions on Scalable Information Systems

View full text Add to dashboard Cite

show abstract

Bibliometric Analysis of the Worldwide Scholarly Output on Artificial Intelligence in Scopus

Auza-Santiváñez,

Carías Díaz,

Vedia Cruz

et al. 2023

Gamification and Augmented Reality

View full text Add to dashboard Cite

Introduction: the use of bibliometric analyses is useful to gain insight into the development, trends, and impact of scholarly output on artificial intelligence (AI) in several fields.Objective: to characterize the worldwide scholarly output on AI in Scopus in the period 2013-2022.Method: a descriptive observational bibliometric study was carried out. The study population consisted of the 776,961 documents identified using SciVal. The following variables were studied: number of documents (Ndoc), year of publication, annual variation rate (AVR) of the scholarly output, type of document, source, number of citations (Ncit), field-weighted citation impact (FWCI), author(s), author-level h-index, institution, country, type of collaboration, and keyphrases.Results: the scholarly output showed a steady quantitative increase during the period studied, with a positive AVR. Conference papers (68.5%) and articles (26.5%) were the main types of documents. Neurocomputing led the list of sources in both Ndoc (12,989) and Ncit (351,837). The highest FWCI (3.02) corresponded to Proceedings – IEEE International Conference on Robotics and Automation. China, the United States and India were the countries with the highest Ndoc by year of publication. Institutional collaboration was the most common (46.7%) type of collaboration. The most prominent keyphrases were: Robot, Artificial Intelligence, Deep Learning, Convolutional Neural Network and Robotics. Conclusions: the scholarly production analyzed is characterized by its constant quantitative growth and is mostly represented by conference papers. Productivity and impact indicators based on citations show remarkable results. The science produced was led by China, and scientific collaboration played a relevant role.

show abstract

Community participation in the Latin American context: Bibliometric Analysis

Hernández Estevez,

Gonzalez-Argote

2024

Southern perspective / Perspectiva austral

Self Cite

View full text Add to dashboard Cite

Introduction: community participation and outreach constitute one of the key work tools for each science. Despite the existence of previous studies on this topic, it is necessary to have an overview of the current state of knowledgeObjective: characterizes the scientific publications grouped in Scopus regarding community participation in the Latin American contextMethod: A bibliometric study was developed. 5,832 publications were analyzed as the research universe. Bibliometric indicators were applied from the Scopus database and the Scival tool.Results: Articles published in 2022 predominated with 988 investigations (16,94 %). The thematic areas of social sciences stood out (2,639 investigations; 45,25 %). The relationship between the different thematic areas was mostly heterogeneous, with greater participation from the medical sciences. Articles published in research article format stood out with 4,196 investigations, equivalent to 71,94 %. The largest number of articles was published in journals located in quartile 1 (Q1) with 2,407 investigations (41,27 %). The researches with authors from the United States (4,933 articles; 84,58 %) and Canada (885 works; 15,17 %) stood out.Conclusions: scientific production on community participation in the Latin American context was characterized by gradual growth. In turn, in correlation with a heterogeneous behavior of the branches of sciences involved in scientific activity. The center of the research responds to original articles located in the countries with the highest rate of scientific activity.

show abstract

Trends in scientific output on artificial intelligence and health in Latin America in Scopus

Cited by 3 publications

References 44 publications

Fast Lung Image Segmentation Using Lightweight VAEL-Unet

Fast Lung Image Segmentation Using Lightweight VAEL-Unet

Bibliometric Analysis of the Worldwide Scholarly Output on Artificial Intelligence in Scopus

Community participation in the Latin American context: Bibliometric Analysis

Contact Info

Product

Resources

About