BMBirajara Soares Machado

Birajara Soares Machado

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Birajara Soares Machado has a total of 42 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

    Attitudes, knowledge, opinions, and expectations of medical students towards medical artificial intelligence solutions: a cross-sectional survey study

    by Bruna Giavina-Bianchi, Mara Giavina-Bianchi, Nelson Wolosker, Edson Amaro, Birajara Soares Machado

    INTRODUCTION Artificial intelligence (AI) technologies are widely used in everyday applications worldwide, with significant societal and economic changes.() Artificial intelligence can learn from large sets of data, including medical texts, images, and patient records. This makes medical AI solutions applicable to various areas including diagnosis,() therapy,() clinical decision support,()personalized medicine,()clinical research,() drug development,() administrative tasks,()scientific or layman writing,(,) and virtual healthcare assistance.() There is a growing focus on how prepared the healthcare workforce is, regarding this new situation,()and it has


  • Open Access

    Medical Expectations of Physicians on AI Solutions in Daily Practice: Cross-Sectional Survey Study

    by Mara Giavina-Bianchi, Edson Amaro Jr, Birajara Soares Machado

    Background: The use of artificial intelligence (AI) in medicine has been a trending subject in the past few years. Although not frequently used in daily practice yet, it brings along many expectations, doubts, and fears for physicians. Surveys can be used to help understand this situation. Objective: This study aimed to explore the degree of knowledge, expectations, and fears on possible AI use by physicians in daily practice, according to sex and time since graduation. Methods: An electronic survey was sent to physicians of a large hospital in Brazil, from August to September 2022. Results: A total of 164 physicians responded to our survey. Overall, 54.3% (89/164) of physicians considered themselves to have an intermediate knowledge of AI, and 78.5% (128/163) believed that AI should be regulated by a governmental agency. If AI solutions were reliable, fast, and available, 77.9% (127/163) intended to frequently or always use AI for diagnosis (143/164, 87.2%), management (140/164, 85.4%), or exams interpretation (150/164, 91.5%), but their approvals for AI when used by other health professionals (85/163, 52.1%) or directly by patients (82/162, 50.6%) were not as high. The main benefit would be increasing the speed for diagnosis and management (106/163, 61.3%), and the worst issue would be to over rely on AI and lose medical skills (118/163, 72.4%). Physicians believed that AI would be useful (106/163, 65%), facilitate their work (140/153, 91.5%), not alter the number of appointments (80/162, 49.4%), not interfere in their financial gain (94/162, 58%), and not replace their jobs but be an additional source of information (104/162, 64.2%). In case of disagreement between AI and physicians, most (108/159, 67.9%) answered that a third opinion should be requested. Physicians with ≤10 years since graduation would adopt AI solutions more frequently than those with >20 years since graduation (P=.04), and female physicians were more receptive to other hospital staff using AI than male physicians (P=.008). Conclusions: Physicians were shown to have good expectations regarding the use of AI in medicine when they apply it themselves, but not when used by others. They also intend to use it, as long as it was approved by a regulatory agency. Although there was hope for a beneficial impact of AI on health care, it also brings specific concerns.


  • Open Access

    Medical Expectations Survey on Artificial Intelligence Solutions in daily practice

    by Mara Giavina-Bianchi, Edson Amaro, Birajara Soares Machado

    Background: Artificial intelligence (AI) applied to Medicine has become one of the hottest topics for the past years. Although scarcely used in real practice, it brings along many expectations, doubts and fears for physicians. Surveys can help to understand this situation. Objective: To explore the degree of knowledge, expectations, fears and daily practice questions on AI use by physicians. Methods: an electronic survey was sent to physicians of a large hospital in Brazil, from August-September 2022. Results: 171 physicians responded to our survey. 54% considered themselves to have an intermediate knowledge of AI. 79% believe AI should be regulated by a Governmental Agency. If AI were reliable and available, 78% intend to use AI frequently/always for diagnosis (87%) and/or management (83%), but they were unsure about the use of AI by other health professionals (50%) or by the patients (51%). The main benefit would be increasing the speed for diagnosis and management (64%), and the worst issue, to over rely on AI and lose medical skills (71%). Physicians believe AI would be useful (94%), facilitate the work (87%), increase the number of appointments (54%), not interfere in the financial gain (58%) and not replace their jobs, but, rather, be utilized as an additional source of information (65%). In case of disagreement between AI and physicians, most answered that a third opinion should be requested (86%). There were no significant differences between the physicians’ answers according to time since graduation. Conclusions: physicians showed to have good expectations regarding the use of AI in Medicine when applied by themselves, but not so much by others. They also have intention to use it, as long as it was approved by a Regulatory Agency. Although there was hope for the beneficial impact of AI on healthcare, it also brings specific concerns.


  • Open Access

    Melanoma: implications of diagnostic failure and perspectives

    by Mara Giavina-Bianchi, Eduardo Cordioli, Birajara Soares Machado

    Melanoma accounts for only approximately 5% of all skin cancer diagnoses worldwide, but it is very important due to its metastatic potential and the fact that it can lead to death.() In Brazil, in 2019, there were 1,978 deaths from the disease, and the estimate was 8,450 new cases for 2020.() A recently published study with Brazilian data showed that the mean incidence in men increased from 2.52 to 4.84, and in women, from 1.33 to 3.22 cases/100,000 inhabitants in [...] Melanoma accounts for only approximately 5% of all skin cancer diagnoses worldwide, but it is very important due to its metastatic potential and the fact that it can lead to death.() In Brazil, in 2019, there were 1,978 deaths from the disease, and the estimate was 8,450 new cases for 2020.() A recently published study with Brazilian data showed that the mean incidence in men increased from 2.52 to 4.84, and in women, from 1.33 to 3.22 cases/100,000 inhabitants in


  • Open Access

    Brain connectivity in patients with dystonia during motor tasks

    by Carlos Arruda Baltazar, Birajara Soares Machado, Danilo Donizete Silva de Faria, Artur José Marques Paulo, Sonia Maria Cezar Azevedo Silva, Henrique Ballalai Ferraz, Patrícia Maria de Carvalho Aguiar

    The pathophysiology of dystonia has been related to abnormalities of sensorimotor integration and disorders in basal nuclei, but the role of other brain regions is still unclear. This study aims to investigate alterations of brain connectivity using multivariate electroencephalographic data to provide new insights of the brain connectivity dynamics of dystonia. We recorded electroencephalography (EEG) of patients with right upper limb idiopathic focal dystonia and paired controls during resting state, writing-from-memory, and finger-tapping tasks. We applied power spectrum analyses considering the mu, beta and gamma rhythms of the motor cortex and analyzed brain connectivity networks and microstates. The power spectra results showed that patients had a loss of desynchronization of the beta rhythm during the writing task. We observed differences in the structure of the connective core in beta rhythm, as well as, in the intensity of the patient's hubs observed with basis in path length measures in mu and beta rhythms. Abnormalities were also identified in microstates of default mode networks of patients associated with its performances during motor tasks. The EEG connectivity analyses provided interesting insights about the cortical electrophysiological patterns in dystonia, such as loss of event-related desynchronization, changes in the effective connectivity with similar signature to other neurological diseases, indications of alterations in the default-mode-network. Our findings are consistent with previously described connectivity abnormalities in neuroimaging studies confirming that dystonia is a network disorder.


  • Coronavirus Disease 2019 Deep Learning Models: Methodologic Considerations

    by Andrew Maranhão Ventura Dadário, Joselisa P. Q. de Paiva, Rodrigo Caruso Chate, Birajara Soares Machado, Gilberto Szarf


  • Open Access

    Connective Core Structures in Cognitive Networks: The Role of Hubs

    by Carlos Arruda Baltazar, Maria Isabel Barros Guinle, Cora Jirschik Caron, Edson Amaro Jr., Birajara Soares Machado

    Complex network analysis applied to the resting brain has shown that sets of highly interconnected networks with coherent activity may support a default mode of brain function within a global workspace. Perceptual processing of environmental stimuli induces architectural changes in network topology with higher specialized modules. Evidence shows that, during cognitive tasks, network topology is reconfigured and information is broadcast from modular processors to a connective core, promoting efficient information integration. In this study, we explored how the brain adapts its effective connectivity within the connective core and across behavioral states. We used complex network metrics to identify hubs and proposed a method of classification based on the effective connectivity patterns of information flow. Finally, we interpreted the role of the connective core and each type of hub on the network effectiveness. We also calculated the complexity of electroencephalography microstate sequences across different tasks. We observed that divergent hubs contribute significantly to the network effectiveness and that part of this contribution persists across behavioral states, forming an invariant structure. Moreover, we found that a large quantity of multiple types of hubs may be associated with transitions of functional networks.


  • Synchronization structure of evolving epileptic networks using cross-entropy

    by Abner Cardoso Rodrigues, Luis Otavio Sales Ferreira Caboclo, Hilda Alicia Cerdeira, Edson Amaro, Birajara Soares Machado