GSGilberto Szarf

Gilberto Szarf

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Gilberto Szarf has a total of 653 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

    Posicionamento sobre o Uso dos Mapas Multiparamétricos na Caracterização Tecidual Avançada pela Ressonância Magnética Cardiovascular – 2026

    by Isabela Bispo Santos da Silva Costa, Jorge Andion Torreão, Henrique Simão Trad, Roberto Sasdelli Neto, Gustavo Lemos Palandre, Gabriel Cordeiro Camargo, Rodrigo de Jesus Louzeiro Melo, Luíz Flávio Galvão Gonçalves, Fábio de Morais Medeiros, Gilberto Szarf, et al.

    1. Introdução As doenças cardiovasculares continuam representando a principal causa de morbimortalidade no Brasil, impondo um impacto expressivo tanto em termos de saúde pública quanto em custos socioeconômicos. Entre essas condições, as cardiomiopatias – tanto de origem isquêmica quanto não isquêmica – assumem papel de destaque pela sua elevada prevalência, diversidade etiológica e complexidade diagnóstica. O reconhecimento precoce e acurado dessas doenças é fundamental para direcionar intervenções terapêuticas adequadas, melhorar o prognóstico e reduzir eventos adversos. Posicionamento sobre o Uso [...] 1. Introdução As doenças cardiovasculares continuam representando a principal causa de morbimortalidade no Brasil, impondo um impacto expressivo tanto em termos de saúde pública quanto em custos socioeconômicos. Entre essas condições, as cardiomiopatias – tanto de origem isquêmica quanto não isquêmica – assumem papel de destaque pela sua elevada prevalência, diversidade etiológica e complexidade diagnóstica. O reconhecimento precoce e acurado dessas doenças é fundamental para direcionar intervenções terapêuticas adequadas, melhorar o prognóstico e reduzir eventos adversos. Posicionamento sobre o Uso

  • Open Access

    30-year-old man presenting with chest pain and odynophagia for 2 days

    by Lucca Duda Tavares, Mariana Machado Dantas, Débora Cestari Bacal, Lígia Lopes Balsalobre Trevizan, Gilberto Szarf, José Mariani, Fernando Bacal, Sandrigo Mangini

  • Cardiac involvement of Rosai–Dorfman disease responsive to trametinib

    by Henrique Martins dos Santos Filho, Eduardo Kaiser Ururahy Nunes Fonseca, Gilberto Szarf, Roberto Sasdelli Neto, Walther Yoshiharu Ishikawa

  • Open Access

    Computed tomography imaging features of major pulmonary and extrapulmonary complications of fibrotic lung diseases

    by Pedro Paulo Teixeira e Silva Torres, Paula Terra Martins Almeida Amaral, Alan Diniz Hummel, Eduardo Kaiser Ururahy Nunes Fonseca, Rodrigo Caruso Chate, Gilberto Szarf, Flávia Castro Velasco, Dante Luiz Escuissato, Marcelo Fouad Rabahi, Edson Marchiori

    Patients diagnosed with fibrosing interstitial lung disease are at risk of several complications, some of which may be life-threatening. Computed tomography imaging plays an important role in diagnosing these overlapping conditions. This article summarizes the computed tomography imaging features of the most common conditions associated with fibrosing interstitial lung diseases, categorized by involvement of the lung parenchyma or extra-pulmonary compartments. Some steps may help to recognize such complications, such as having knowledge of the underlying fibrotic lung disease, being

  • Open Access

    AI-Assisted Chest X-Ray Interpretation in Resource-Limited Settings: LuAna Stepped-Wedge Trial Protocol (Preprint)

    by Maria Carolina Bueno da Silva, Paula Bresciani M. de Andrade, Henrique Min Ho Lee, Pedro Vinicius Alves Silva Silva, Ana Cristina Ferreira, Cintia Pereira Kuss, Maria Gabriela de Almeida Rodrigues, Guilherme Alberto Sousa Ribeiro, Thiago Fellipe Ortiz Camargo, William Yang Chen Fan, et al.

    Background: Artificial intelligence (AI) has the potential to transform chest radiography interpretation by enhancing diagnostic accuracy, identifying subtle findings, reducing errors, and helping prioritize patient care. Although chest radiography remains a cost-effective and widely used imaging tool, its effectiveness is limited by overlapping anatomy and variability in clinical expertise. Integrating AI can help overcome some of these challenges, especially in resource-constrained settings. However, robust validation in real-world clinical contexts is essential before widespread implementation. This study protocol evaluates whether AI assistance improves general practitioners’ ability to detect radiographic findings on chest radiography in adults with respiratory complaints or those undergoing treatment for respiratory diseases compared with unaided interpretation. Potential benefits include increased diagnostic safety, higher physician confidence, more efficient workflows, and expanded access to expert support in underserved areas. Objective: This study aims to evaluate whether AI assistance enhances physicians’ ability to detect key radiographic abnormalities, including consolidation or pulmonary opacity, pneumothorax, atelectasis, pleural effusion, and cardiomegaly. The primary outcome is the difference in physicians’ diagnostic accuracy (per examination) when assisted by the AI tool compared with usual practice, using expert radiologist consensus as the reference value. Methods: This study is a protocol for a multicenter, stepped-wedge, cluster-randomized clinical trial following the CONSORT-AI (Consolidated Standards of Reporting Trials-Artificial Intelligence) extension and SPIRIT-AI (Standard Protocol Items: Recommendations for Interventional Trials-Artificial Intelligence) guidelines. The intervention involves the diagnostic support solution for chest radiography, Lung Analysis (LuAna), an AI-powered chest X-ray interpretation tool developed in partnership with the Brazilian Ministry of Health. Across 9 cities in Brazil, clusters will transition monthly from unaided chest X-ray interpretation by general practitioners to AI-assisted interpretation, with performance benchmarked against thoracic radiologists. The stepped-wedge design ensures that all clusters receive the intervention, reflecting real-world coordination, enhancing acceptability, improving statistical power, and strengthening causal inference through repeated measures. Diagnostic performance will be compared with a reference standard established by thoracic radiologists. Results: This project was funded in October 2024 (following ethics approval by the institutional review board). Data collection commenced in January 2026 and is projected to be completed by September 2026, marking the end of the trial period. As of November 2025, 3 centers were fully prepared for enrollment initiation. The LuAna clinical trial is currently ongoing, with data analysis (including statistical analyses) forecasted to be finalized by November 2026. Results are expected to be published by January 2027. Conclusions: This intervention is expected to enhance clinical decision-making by supporting earlier treatment initiation and more appropriate diagnostic pathways for patients with respiratory symptoms while maintaining a favorable safety profile and high physician usability. Findings from this trial will provide real-world evidence on the clinical utility of AI-assisted chest radiography. If effective, LuAna may leverage its scalability and equity advantages to become a replicable model for integrating AI into routine imaging workflows worldwide, especially in regions with limited access to specialist care.

  • Open Access

    AI- Assisted Chest X-Ray Interpretation in Resource-Limited Settings: LuAna Stepped-Wedge Trial Protocol (Preprint)

    by Maria Carolina Bueno da Silva, Paula Bresciani M. de Andrade, Henrique Min Ho Lee, Pedro Vinicius Alves Silva Silva, Ana Cristina Ferreira, Cintia Pereira Kuss, Maria Gabriela de Almeida Rodrigues, Guilherme Alberto Sousa Ribeiro, Thiago Fellipe Ortiz Camargo, William Yang Chen Fan, et al.

    BACKGROUND Artificial intelligence (AI) has the potential to transform chest radiography (CXR) interpretation by enhancing diagnostic accuracy, identifying subtle findings, reducing errors, and helping prioritize patient care. Although CXR remains a cost-effective and widely used imaging tool, its effectiveness is limited by overlapping anatomy and variability in clinical expertise. Integrating AI can help overcome some of these challenges, especially in resource-constrained settings. However, robust validation in real-world clinical contexts is essential before widespread implementation. This study protocol evaluates whether AI assistance improves general practitioners' ability to detect radiographic findings on CXR in adults with respiratory complaints or undergoing treatment for respiratory diseases, compared to unaided interpretation. Potential benefits include increased diagnostic safety, higher physician confidence, more efficient workflows, and expanded access to expert support in underserved areas. OBJECTIVE This project aims to evaluate whether AI assistance enhances physicians’ ability to detect key radiographic abnormalities— including consolidation or pulmonary opacity, pneumothorax, atelectasis, pleural effusion, and cardiomegaly. The primary outcome is the difference in physicians’ diagnostic accuracy (per examination) when assisted by the AI tool compared with usual practice, using the expert radiologist consensus as the reference value. METHODS This is a protocol for a multicenter, stepped-wedge, cluster-randomized clinical trial following the CONSORT-AI extension and SPIRIT-AI guidelines. The intervention involves the diagnostic support Solution for CXR - Lung Analysis (LuAna), an AI-powered chest X-ray interpretation tool developed in partnership with the Brazilian Ministry of Health. Across nine cities in Brazil, clusters will transition monthly from unaided chest X-ray interpretation by general practitioners to AI-assisted interpretation, with performance benchmarked against thoracic radiologists. The stepped-wedge design ensures all clusters receive the intervention, reflecting real-world coordination, enhancing acceptability, improving power, and strengthening causal inference through repeated measures. Diagnostic performance will be compared to a reference standard established by thoracic radiologists. RESULTS Thirteen research centers across Brazil will participate, covering all five regions and diverse healthcare settings, from primary care to specialized tuberculosis centers. Next steps involve finalizing regulatory approvals and starting participant enrolment once all sites are fully activated. CONCLUSIONS This intervention is expected to enhance clinical decision-making by supporting earlier treatment initiation and more appropriate diagnostic pathways for patients with respiratory symptoms, while maintaining a favorable safety profile and high physician usability. The findings from this trial will provide real-world evidence on the clinical utility of AI-assisted chest radiography. If effective, LuAna may leverage its scalability and equity advantages to become a replicable model for integrating AI into routine imaging workflows worldwide, especially in regions with limited access to specialist care. CLINICALTRIAL NCT06686251, Registered on 2024-11-13.

  • Open Access

    Implementation of positron emission tomography radiopharmaceuticals labeled with Gallium-68 in a hospital radiopharmacy: experience with more than 7,500 patients

    by Lilian Yuri Itaya Yamaga, Marycel Rosa Felisa Figols de Barboza, Luciana Malavolta, Leonardo Lima Fuscaldi, Jorge Mejía Cabeza, Solange Amorim Nogueira, Gilberto Szarf, Marcelo Livorsi da Cunha, Taise Vitor, Guilherme de Carvalho Campos, et al.

    Highlights ■ In-house 68Ga radiopharmaceutical synthesis under GMP in a tertiary hospital radiopharmacy. ■ Over 7,500 patients underwent imaging with [68Ga]Ga- DOTATATE and [68Ga]Ga-PSMA-11. ■ This is the first radiopharmacy in Brazil to introduce [68Ga] Ga-DOTA-UBI, and [68Ga]Ga-FAPi-46. ABSTRACT In this study, we report the implementation of in-house synthesis of gallium-68 (68Ga)-labeled radiopharmaceuticals for positron emission tomography (PET) at a tertiary hospital radiopharmacy. [68Ga]Ga-DOTATATE and [68Ga]Ga-PSMA-11 were introduced into clinical practice in January 2014 and October 2015, respectively, following the

  • Open Access

    Pulmonary nodules near fissures in hepatocellular carcinoma: assessing their clinical significance

    by Leonardo Chaves Machado, Gilberto Szarf, Gustavo Borges da Silva Teles, Patrícia Yokoo, Rodrigo Caruso Chate, Eduardo Kaiser Ururahy Nunes Fonseca

    Highlights ■ Typical perifissural nodules remain radiologically stable with no metastatic potential. ■ Only two nodules progressed to metastasis (one atypical perifissural nodule and one non-perifissural nodule). ■ Atypical and non-perifissural nodules require closer follow-up. ■ Annual growth rate was a strong marker of malignancy. ABSTRACT Objective: To evaluate whether pulmonary nodules adjacent to fissures, particularly perifissural nodules detected on chest computed tomography in patients with hepatocellular carcinoma, can be reliably classified as benign based on their radiological characteristics. Methods: