LKLucas Kallas-Silva

Lucas Kallas-Silva

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Lucas Kallas-Silva has a total of 71 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Multi-Task Deep Learning Model for the Automated Segmentation of Neuroimages in Real-World CT Scans

    by Larissa Moraes, Luigi Lago, Priscila Satomi Acamine, Paulo Estevam, Lucas Kallás Silva, Rafael Loureiro, Joselisa Paiva, Gilberto Szarf, Edson Amaro, Birajara Machado

    Background and Objective: Segmenting head structures in medical images is challenging due to anatomical heterogeneity and structural complexity. Multi-task learning can improve performance through shared representations and may help address these challenges. However, architectures such as U-Net can struggle with distorted anatomy, while 3D shape-learning approaches often increase computational cost, motivating efficient alternatives.Methods: We propose DeepCTE3D, a multi-task model based on an adapted U-Net architecture, called И-Net, for segmenting the intracranial space, lateral ventricles, and eyeballs from 3D CT images. The model was trained and evaluated on 559 real-world CT scans from newborns to adults, including cases with severe anatomical distortions such as enlarged ventricles, skull deformities, and abnormal eyeballs. We hypothesized that segmentation complexity and morphological heterogeneity affect learning performance and generalization within a multi-task framework.Results: DeepCTE3D achieved Dice similarity coefficients of 98.13% for the intracranial space, 80.76% for the lateral ventricles, and 87.73% for the eyeballs. Because of their greater morphological heterogeneity and segmentation complexity, the lateral ventricles served as a reference task. Compared with single-task models, the multi-task approach improved training convergence while maintaining comparable or superior performance, particularly enhancing generalization for challenging structures.Conclusions: DeepCTE3D effectively segmented complex head structures across diverse CT datasets, including anatomically distorted cases. The findings support the hypothesis that accounting for segmentation complexity and morphological heterogeneity improves model generalization, especially for highly variable structures. These results highlight the potential of multi-task learning to extend segmentation capabilities to additional head structures and support CT imaging applications, particularly in low-resource settings.

  • Open Access

    Comparative Effectiveness Of Interventions To Improve Adherence To Antiretroviral Therapy Among People Living With HIV: A Systematic Review, Network Meta-Analysis, And Component Network Meta-Analysis

    by Nadia Rehman, Gordon H. Guyatt, Jessica Gu, Lucas Kallás Silva, Mou Jin Jin, Raha Sodagari, Michelle Li, Juana Xiong, Jay Gajjar, Dominik Mertz, et al.

    Background: Optimal adherence to antiretroviral therapy (ART) is crucial for sustained viral suppression. We compared the effectiveness of adherence interventions and their constituent components among people living with HIV (PLHIV).Methods: We searched PubMed, Embase, CINAHL, PsycINFO, Web of Science, and CENTRAL from Jan 1, 1995, to Dec 31, 2024, for randomised controlled trials evaluating interventions to improve ART adherence among PLHIV. Pairs of reviewers independently screened studies, extracted data, and assessed risk of bias using ROBUST-RCT. We used frequentist random-effects network meta-analysis (NMA) and component network meta-analysis (cNMA) to evaluate ART adherence, viral load suppression, and quality of life. We ranked intervention categories by effect estimates and the certainty of evidence relative to the standard of care (SoC).Findings: Of 4853 records identified, 251 trials involving 54 450 participants were eligible. Eight intervention categories improved ART adherence compared with SoC, including four with moderate/high-certainty evidence: mHealth (OR 1·66, 95% CI 1·02–2·71), behavioural skills training/medication adherence training (OR 1·59, 95% CI 1·14–2·21), economic interventions (OR 1·56, 95% CI 1·02–2·40), and cognitive behavioural therapy (OR 1·53, 95% CI 1·23–1·92). Four intervention categories improved viral load suppression compared with SoC, including three with moderate/high-certainty evidence: supporter (OR 1·43, 95% CI 1·09–1·89), health service delivery (OR 1·33, 95% CI 1·01–1·78), and reminders (OR 1·37, 95% CI 1·13–1·66). Across outcomes, no intervention demonstrated convincing superiority over other active interventions. cNMA found that behavioural skills training/medication adherence training, cognitive behavioural therapy, mHealth, supporter, and psychological components improved ART adherence, whereas peer-led delivery was associated with decreased viral load suppression. Multicomponent interventions frequently demonstrated larger effects than individual components.Interpretation: Several interventions improved outcomes. The effectiveness of adherence-support interventions depends on component content and combinations. These findings can inform the design of new context-specific adherence-support programmes across diverse settings.

  • Open Access

    Comparative Effectiveness of Interventions to Improve Adherence to Antiretroviral Therapy Among People Living with HIV: A Systematic Review, Network Meta-analysis, and Component Network Meta-analysis

    by Nadia Rehman, Gordon H. Guyatt, Jessica Gu, Lucas Kallás Silva, Mou Jin Jin, Raha Sodagari, Michelle Li, Juana Xiong, Jay Gajjar, Dominik Mertz, et al.

    Background: Optimal adherence to antiretroviral therapy (ART) is crucial for sustained viral suppression. We compared the effectiveness of adherence interventions and their constituent components among people living with HIV (PLHIV). Methods: We searched PubMed, Embase, CINAHL, PsycINFO, Web of Science, and CENTRAL from Jan 1, 1995, to Dec 31, 2024, for randomised controlled trials evaluating interventions to improve ART adherence among PLHIV. Pairs of reviewers independently screened studies, extracted data, and assessed risk of bias using ROBUST-RCT. We used frequentist random-effects network meta-analysis (NMA) and component network meta-analysis (cNMA) to evaluate ART adherence, viral load suppression, and quality of life. We ranked intervention categories by effect estimates and the certainty of evidence relative to the standard of care (SoC). Findings: Of 4853 records identified, 251 trials involving 54 450 participants were eligible. Eight intervention categories improved ART adherence compared with SoC, including four with moderate/high-certainty evidence: mHealth (OR 1·66, 95% CI 1·02–2·71), behavioural skills training/medication adherence training (OR 1·59, 95% CI 1·14–2·21), economic interventions (OR 1·56, 95% CI 1·02–2·40), and cognitive behavioural therapy (OR 1·53, 95% CI 1·23–1·92). Four intervention categories improved viral load suppression compared with SoC, including three with moderate/high-certainty evidence: supporter (OR 1·43, 95% CI 1·09–1·89), health service delivery (OR 1·33, 95% CI 1·01–1·78), and reminders (OR 1·37, 95% CI 1·13–1·66). Across outcomes, no intervention demonstrated convincing superiority over other active interventions. cNMA found that behavioural skills training/medication adherence training, cognitive behavioural therapy, mHealth, supporter, and psychological components improved ART adherence, whereas peer-led delivery was associated with decreased viral load suppression. Multicomponent interventions frequently demonstrated larger effects than individual components. Interpretation: Several interventions improved outcomes. The effectiveness of adherence-support interventions depends on component content and combinations. These findings can inform the design of new context-specific adherence-support programmes across diverse settings.

  • Open Access

    False-positive test results: prevalence, time trends, and factors associated with false-positive serological markers in blood donation screening

    by Lucas Kallas-Silva, Leandro Dinalli dos Santos, Valeria de Freitas Dutra, Jose Mauro Kutner, Carolina Bonet-Bub

  • Open Access

    Medullary thyroid carcinoma and RET molecular testing: A virtual tumor board on clinical decision‐making

    by Lucas Kallas‐Silva, Victor Alexandre dos Santos Valsecchi, Ana O. Hoff, Evelin Cavalcante Farias, Rui Monteiro de Barros Maciel, Rachel Riera, Laura Sterian Ward, Lucas Leite Cunha

  • Open Access

    An automated measurement of head circumference using CT scans: An application in children with head abnormalities

    by Priscila Satomi Acamine, Rafael Maffei Loureiro, Lucas dos Anjos Longas, Fabio Augusto Ribeiro Dalpra, Luigi Villanova Machado de Barros Lago, Larissa Vasconcellos de Moraes, Paulo Cesar Filho Estevam, Luiz Otávio Vittorelli, Lucas Silva Kallás, Ana Paula Antunes Pascalicchio Bertozzi, et al.


  • Myths and misinformation associated with vaccine incompleteness: A survey study

    by Lucas Kallas-Silva, Marcia Thereza Couto, Maria Eduarda Muniz Soares, Sofia Natalia Ferreira-Silva, Vivian I. Avelino-Silva


  • Open Access

    Mirror, mirror, on the wall, who’s the fairest of them all?

    by Uri Adrian Prync Flato, Ricardo Ferreira Mendes de Oliveira, Lucas Kallas-Silva, Maria Fernanda Dias Azevedo

  • Open Access

    Interactive Serious Game to Teach Basic Life Support Among Schoolchildren in Brazil: Design and Rationale

    by Uri Adrian Prync Flato, Emilio José Beffa dos Santos, Isabella Bispo Diaz T Martins, Vinicius Gazin Rossignoli, Thais Dias Midega, Lucas Kallas-Silva, Ricardo Ferreira Mendes de Oliveira, Adriana Do Socorro Lima Figueiredo Flato, Mario Vicente Guimarães, Hélio Penna Guimarães

    Cardiovascular diseases are leading causes of death and morbidity worldwide. CPR and early defibrillation significantly enhance survival rates. Serious Games (SG) improve learning through entertainment. Current strategies target Cardiopulmonary resuscitation for communities and schoolchildren, but none have been validated for children in low-to-middle-income settings. The SG Children Save Hearts, developed in Brazil, teaches the five resuscitation steps according to International Liaison Committee on Resuscitation 2020 guidelines and requires formal usability assessment. The study aimed to evaluate the usability of SG Children Save Hearts among IT and healthcare professionals using the System Usability Scale (SUS). The usability test was conducted in August 2022 in the university's IT department. The game was developed targeting schoolchildren aged 7 to 17. Categorical variables as absolute and relative frequencies, while continuous variables were presented as median with interquartile range (IQR). Normality was assessed using the Shapiro-Wilk test. Comparisons between IT and healthcare professionals were made using the independent t-test for normal distributions or the Mann-Whitney U test for non-normal distributions. We included 17 volunteers with a mean age of 22 years (IQR 20-26). All participants played the game and completed a 10-question survey on its usability using a Likert-type scale. The final grade was converted to a 0 to 100 scale, with a grade above 70 considered acceptable for a minimum viable product. The mean SUS score was 75 (IQR 72.5-87.5). Healthcare professionals gave higher grades to all five domains compared to IT professionals. The average time spent playing the game was 3.2 minutes. Novel technologies have shown promising results for CPR teaching using active teaching methods, but face challenges in developing countries, such as language barriers, device acquisition, cultural differences, and technical support. To our knowledge, this is the first SG developed in portuguese for brazilian schoolchidren. Despite some usability issues, the SG Children Save Hearts is considered adequate for teaching CPR to schoolchildren in Brazil.


  • Open Access

    Design and rationale of an interactive serious game to teach Basic Life Support among schoolchildren in Brazil. (Preprint)

    by Uri Adrian Prync Flato, Emilio José Beffa dos Santos, Isabella Bispo Diaz Toledo Martini, Vinicius Rossignoli, Thais Dias Midega, Lucas Kallas-Silva, Ricardo Ferreira Mendes de Oliveira, Adriana Do Socorro Lima Figueiredo, Mario Vicente Guimarães, Hélio Penna Guimarães

    BACKGROUND Cardiovascular diseases are among the leading causes of death and morbidity in the world including in Brazil. Cardiopulmonary resuscitation (CPR) and early defibrillation increase the chances of survival. Serious Games (SG) are defined as learning and training tools that can be used to enhance the learning processes. It can be presented in several ways, using the game's characteristics as an entertainment factor to achieve educational objectives and specific skills. Current strategies focus on teaching people in the community and schoolchildren how to conduct CPR. The SG Children Save Hearts was developed to teach the five steps of resuscitation according to the guidelines of the International Liaison Committee on Resuscitation on Basic Life Support. Before being applied to schoolchildren for efficacy evaluation, it should go through a formative (Qualitative Usability Testing) assessment by game developers and health professionals. This parameter refers to the ease with which users can consume, learn, and interact with the system. Therefore, usability validation is a critical step in ensuring the effectiveness of the educational resources before introducing them to a specific population. OBJECTIVE This study aims to discover and fix usability issues using a System Usability Scale (SUS). METHODS The Serious game Children Save Hearts was developed on the Smile Game Builder platform (open source) and implemented on Windows 7 to 11 operating systems, targeting schoolchildren from 7 to 17 years old. The script and storytelling are based on the ILCOR 2020 guidelines. The game includes using a joystick and simple commands to simplify the user experience, allowing greater focus on the content. After finishing the SG, participants were invited to answer a survey of 10 questions regarding its usability on a Likert-type scale. A grade higher than 70 is considered acceptable to proceed to a minimum viable product. A sample size of 17 users was needed based on an estimated probability of encountering an error in the interface expressed as 10% to find 85% of the problems. RESULTS The SG Children Save Hearts was applied to 17 volunteers, with a mean age of 22 years, and 47% were male. Regarding professional training, 8 (47%) were bachelor's in information technology, and 9 (53%) were nurses and doctors (health professionals). All participants played the game and answered the questionnaire at the end. The mean SUS score was 75 (Table 1). The mean score was 3.0. Questions 2 and 4 had the lowest score (2.0), and questions 7 and 9 had the highest. The game performed better on efficiency and error minimization criteria with healthcare professionals (p<0,05), with worse performance on memorization criteria. The average time spent in the game was 3.2 minutes, which was approximately 2 minutes to complete the questionnaire. CONCLUSIONS The SUS is a validated tool for usability tests and was selected for its practicality, ease of realization, and having been cross-culturally validated into Portuguese, that is, with discriminatory power of acceptability of implementation in Brazil. Active teaching methods were required to improve the chance of survival and translate accessible knowledge into practice. Several cardiac awareness programs, such as Kids Save Lives and World Restart a Heart Day (WRAH), are currently underway to teach schoolchildren that CPR is vital to our future. Although the game has points to improve in its usability, it is considered adequate to be used in the teaching process of schoolchildren in cardiopulmonary resuscitation in Brazil.