BMBruno Augusto Linhares Almeida Mariz

Bruno Augusto Linhares Almeida Mariz

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Bruno Augusto Linhares Almeida Mariz has a total of 178 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Machine Learning Classification of Palatal Salivary Gland Tumors Using Clinical and Demographic Descriptors

    by Luís Arthur de Melo Tassinari, Anna Luíza Damaceno Araújo, Sebastião Silvério de Sousa‐Neto, Felipe Paiva Fonseca, Bruno Augusto Linhares Almeida Mariz, Manoela Domingues Martins, Bruno Augusto Benevenuto de Andrade, Michelle Agostini, Hélder Antônio Rebelo Pontes, Ana Carolina Prado Ribeiro, et al.

    Objective: To investigate the potential of machine learning systems for the clinical classification of palatal salivary gland tumors into five diagnostic categories using only demographic and clinical data. Methods: Four machine learning models—Multilayer Perceptron (MLP), Support Vector Machine (SVM), Random Forest (RF), and XGBoost—were implemented based on demographic and clinical attributes from 100 patients. The methodology involved a four‐step process: Hyperparameter optimization using systematic search for combinations (Grid Search), fivefold cross‐validation comprising training and testing, classifier training, and tests, followed by the acquisition of standardized performance metrics. The input attributes included age, sex, location, and symptoms. Performance was evaluated using accuracy, macro‐average sensitivity, specificity, precision, and F1‐score. Results: According to the mean accuracy values, XGBoost and MLP achieved the highest performance (81%), followed by SVM (80%) and RF (79%). Nevertheless, for all models, both macro‐averaged sensitivity and F1‐score were relatively low, remaining below 75%. Specificity emerged as the most consistent metric, ranging from 85% to 90% across all classifiers. All models reached a perfect score (1.0) in the classification of PA, whereas performance declined for malignant tumors, particularly for the rarer subtypes. Conclusions: Machine learning is a feasible approach for classifying palatal salivary gland tumors, demonstrating high specificity but limited sensitivity, primarily due to the uneven distribution of tumor subclasses, particularly the malignant ones, which are underrepresented owing to their rarity. XGBoost proved to be the most robust model with a low computational cost.

  • Open Access

    Impact of CCR5-inhibitors on cancer treatment: A systematic review of preclinical evidence

    by João Lucas Corrêa de Andrade, Bruno Augusto Linhares Almeida Mariz, Nilva Karla Cervigne Furlan, Ricardo D. Coletta, Rômulo Dias Novaes, Carine Ervolino de Oliveira

    Introduction and objective: In view of the role of the CCR5 chemokine receptor in tumour development and progression, researchers have investigated the effects of its inhibition in different types of cancer. Although promising results have been reported, the efficacy, safety, and methodological...

  • Head and Neck Rhabdomyosarcoma in Pediatric Patients: An International Collaborative Study

    by Karen Patricia Domínguez Gallagher, Keith D. Hunter, Lady Paola Aristizabal Arboleda, Caique Mariano Pedroso, Bruno Augusto Linhares Almeida Mariz, Paulo Victor Mendes Penafort, Lucas Lacerda de Souza, Carla Isabelly Rodrigues‐Fernandes, Elena María José Roman Tager, Roman Carlos, et al.

    Background: Rhabdomyosarcoma (RMS), a rare malignant tumor, frequently affects pediatric patients, with 35%–40% occurring in the head and neck. This study analyzes the clinicopathologic profile of pediatric head and neck rhabdomyosarcomas from Brazil, Guatemala, Mexico, and South Africa.Methods: We reviewed 44 cases from 10 Oral and Maxillofacial Pathology services, conducting immunohistochemical analyses of desmin, myogenin, Myo‐D1, and Ki67, with quantification via QuPath software. Cases with ≥ 50% myogenin expression were tested for fusion status using AP2β, NOS‐1, and HMGA2. Statistical analyses included the Kruskal–Wallis test for age and marker expression comparisons, Fisher's exact test for categorical variables, Spearman's rank correlation for marker relationships, and multinomial logistic regression to assess fusion status likelihood.Results: Cases were predominantly from Brazil (40.9%), followed by South Africa (27.3%), Guatemala (22.7%), and Mexico (9.1%). Two‐thirds of patients were diagnosed in their first decade with no gender predilection. Nonparameningeal sites (45.5%) were more affected than parameningeal (40.9%) and orbital sites. Microscopically, embryonal RMS (77.3%) was most common, followed by alveolar (18.2%) and spindle cell (2.3%) tumors. Immunohistochemistry revealed positivity for myogenic markers, with significant differences in myogenin expression between embryonal and alveolar RMS variants (p < 0.05). Fusion status prediction identified two potential fusion‐positive alveolar RMS cases, while all embryonal RMS and one alveolar RMS case appeared fusion‐negative. Significant correlation with positive fusion status was found only between AP2β and NOS1 (p < 0.05).Conclusion: Although there are slight clinical‐demographic variations among pediatric head and neck rhabdomyosarcomas in these regions, identifying fusion status through immunohistochemistry remains a diagnostic challenge.


  • Atypical cellular neurothekeoma of the oral mucosa: A rare case report and literature review

    by Hélen Kaline Farias Bezerra, Gabriele Prospero Nakamura, Bruno Augusto Linhares Almeida Mariz, Rodrigo Nascimento Lopes, José Divaldo Prado, Ana Luiza Oliveira Corrêa Roza, Pablo Agustin Vargas

    Cellular neurothekeoma (CN) is a benign dermal neoplasm that most often affects the head and neck region and rarely occurs in the oral mucosa. We report a rare case of CN with atypical features on the floor of the mouth and summarize the reported cases of oral CN in English‐language literature. A 62‐year‐old woman presented with a 6‐month history of a painless mass on the floor of the mouth. Histopathological analysis of the excised specimen revealed a proliferation of neoplastic cells with oval to spindle morphology arranged in a vaguely nested and multinodular architecture separated by scarce hyaline collagen within a predominantly myxoid‐rich stroma. The tumor cells were positive for NSE, and CD63 (NKI/C3), and negative for S100 protein, CD34, and SMA. Thus, the final diagnosis was CN. In addition, we summarized all clinicopathological data on oral CNs reported in the English‐language literature. Nineteen cases were reviewed. Among them, only one case affected the floor of the mouth of a young girl, in contrast to the present case that occurred in an elderly woman. CN is particularly rare in this location and may be a diagnostic challenge for oral pathologists due to its rarity and morphological similarity with other lesions.

  • The use of deep learning state‐of‐the‐art architectures for oral epithelial dysplasia grading: A comparative appraisal

    by Anna Luíza Damaceno Araújo, Viviane Mariano da Silva, Matheus Cardoso Moraes, Henrique Alves de Amorim, Felipe Paiva Fonseca, Maria Sissa Pereira Sant'ANa, Ricardo Alves Mesquita, Bruno Augusto Linhares Almeida Mariz, Hélder Antônio Rebelo Pontes, Lucas Lacerda de Souza, et al.

    Background: Dysplasia grading systems for oral epithelial dysplasia are a source of disagreement among pathologists. Therefore, machine learning approaches are being developed to mitigate this issue.Methods: This cross‐sectional study included a cohort of 82 patients with oral potentially malignant disorders and correspondent 98 hematoxylin and eosin‐stained whole slide images with biopsied‐proven dysplasia. All whole‐slide images were manually annotated based on the binary system for oral epithelial dysplasia. The annotated regions of interest were segmented and fragmented into small patches and non‐randomly sampled into training/validation and test subsets. The training/validation data were color augmented, resulting in a total of 81,786 patches for training. The held‐out independent test set enrolled a total of 4,486 patches. Seven state‐of‐the‐art convolutional neural networks were trained, validated, and tested with the same dataset.Results: The models presented a high learning rate, yet very low generalization potential. At the model development, VGG16 performed the best, but with massive overfitting. In the test set, VGG16 presented the best accuracy, sensitivity, specificity, and area under the curve (62%, 62%, 66%, and 65%, respectively), associated with the higher loss among all Convolutional Neural Networks (CNNs) tested. EfficientB0 has comparable metrics and the lowest loss among all convolutional neural networks, being a great candidate for further studies.Conclusion: The models were not able to generalize enough to be applied in real‐life datasets due to an overlapping of features between the two classes (i.e., high risk and low risk of malignization).


  • DENTAL CARE IN CRITICALLY ILL PATIENTS: THE EXPERIENCE OF A MULTIDISCIPLINARY TEAM

    by Bruno Augusto Linhares Almeida Mariz, Aljomar José Vechiato Filho, Patrícia Verónica Aulestia Vieira, André Caroli Rocha, Rafael Alves Franco, Thiago Gomes Romano, Thaís Bianca Brandão

  • IMMUNE MICROENVIRONMENT AND PROGNOSIS OF TONGUE SQUAMOUS CELL CARCINOMAS

    by Raísa Sales de Sá, Marisol Miranda-Galvis, Bruno Augusto Linhares Almeida Mariz, Luciana Schultz, Pablo Agustin Vargas, Kenneth Jonh Gollob, Luiz Paulo Kowalski

  • The impact of radiation caries on morbidity and mortality outcomes of head and neck squamous cell carcinoma patients

    by Natália Rangel Palmier, Ana Carolina Prado‐Ribeiro, Bruno Augusto Linhares Almeida Mariz, Letícia Rodrigues‐Oliveira, Mariana de Pauli Paglioni, Juliana Trindade Clemente Napimoga, Caique Mariano Pedroso, Karina Morais‐Faria, Maria Cecília Querido de Oliveira, Aljomar José Vechiato‐Filho, et al.

    AimsRadiation caries (RC) is a highly prevalent and chronic complication of head and neck radiotherapy (HNRT) and presents a challenge for clinicians and patients. The present study aimed to assess the impact of RC on the morbidity and mortality outcomes of head and neck squamous cell carcinoma (HNSCC) patients. Methods and ResultsPatients were divided into three groups: (1) RC (n = 20), (2) control (n = 20), and (3) edentulous (n = 20). Information regarding the number of appointments, dental procedures, osteoradionecrosis (ORN), prescriptions, and hospital admissions were collected. Mortality outcomes were assessed through disease-free survival (DFS) and overall survival (OS) rates. RC patients required more dental appointments (p < .001), restorations (p < .001), extractions (p = .001), and antibiotic and analgesic prescriptions (p < .001). Kaplan-Meier subgroup analyses showed a significantly increased risk of ORN in RC compared to edentulous patients (p = .015). RC patients presented lower DFS rates (43.2 months) than the control and edentulous groups (55.4 and 56.1 months, respectively). ConclusionsRC impacts morbidity outcomes among cancer survivors due to increased demand for medication prescriptions, multiple specialized dental appointments, invasive surgical treatments, increased risk of ORN, and increased need for hospital admissions.


  • FGFR1 is an important prognostic factor in oral leukoplakia and tongue squamous cell carcinoma

    by Bruno Augusto Linhares Almeida Mariz, Raísa Sales de Sá, Anna Luiza Damaceno Araújo, Carla Isabelly Rodrigues Fernandes, Fernanda Viviane Mariano, Alan Roger Santos‐Silva, Marcio Ajudarte Lopes, Pablo Agustin Vargas, Oslei Paes de Almeida, Luiz Paulo Kowalski, et al.

    Background: Fibroblast growth factor receptor 1 is a potential prognostic factor for tongue squamous cell carcinoma and is associated with oral epithelial dysplasia grade in oral leukoplakia.Methods: Thirty cases of tongue squamous cell carcinoma and 30 cases of oral leukoplakia were analyzed. Fibroblast growth factor receptor 1 and phosphorylated Akt protein expression were analyzed by immunohistochemistry and quantified using a digital algorithm. Fibroblast growth factor receptor 1 gene amplification was analyzed by fluorescent in situ hybridization in the tongue squamous cell carcinoma cases.Results: Clinical appearance and dysplasia grade were correlated with oral leukoplakia malignant transformation. Oral leukoplakia cases presenting high fibroblast growth factor receptor 1 expression showed a higher risk of malignant transformation (p = 0.016, HR: 7.3, 95% CI: 1.4–37.4). Phosphorylated Akt showed faint to no expression in oral leukoplakia, which did not correlate with dysplasia grade or malignant transformation. High expression of fibroblast growth factor receptor 1 and phosohorylated Akt were associated with poor overall survival and disease‐free survival in tongue squamous cell carcinoma, although only fibroblast growth factor receptor 1 expression was significantly associated with poor overall survival (p = 0.024; HR: 4.9, 95% CI: 1.2–19.9). Cases presenting double fibroblast growth factor receptor 1/phosphorylated Akt overexpression (n = 8) showed markedly impaired overall survival (p = 0.020; HR: 6.4, 95% CI: 1.3–31.1) and disease‐free survival (p = 0.001, HR: 13.0, 95% CI: 3.0–55.7). Fibroblast growth factor receptor 1 amplification was observed in 16.6% of tongue squamous cell carcinoma cases, being correlated with vascular and neural invasion (p = 0.001 and 0.017, respectively), but not with fibroblast growth factor receptor 1 protein expression, overall survival, or disease‐free survival.Conclusion: Fibroblast growth factor receptor 1 protein expression is an important prognostic factor in oral leukoplakia and tongue squamous cell carcinoma.