Bio


Mateus A. Esmeraldo is a Postdoctoral Research Fellow in the Department of Radiology at Stanford University, currently working under the mentorship of Dr. Bruno P. Soares. His current research focuses on pediatric neuroradiology and the integration of artificial intelligence (AI) tools to enhance diagnostic and healthcare delivery in neuropediatric populations, with a particular emphasis on magnetic resonance imaging.

Originally from Brazil, Mateus earned his medical degree Magna cum Laude from the University of Ceará - Sobral. He completed his residency in Radiology at the University of São Paulo, where he was awarded the Guerbet-InRad Best Resident Award and received the CBR/ESOR Europe Scholarship for complementary training in Neuroradiology at Addenbrooke’s Hospital, University of Cambridge, during his final year.

Following his residency, Mateus served as a Radiologist in the Ultrasound Section of the Institute of Radiology (InRad) at the University of São Paulo. In this role, he was engaged in clinical and research activities involving Doppler ultrasound and transcranial ultrasound in adult patients. His work was closely aligned with innovation-driven projects in neuroimaging and artificial intelligence, and he also participated in academic and teaching initiatives for medical students and residents.

Honors & Awards


  • Bayer - CBR Award (2024), Bayer, Brazilian College of Radiology (CBR) and European School of Radiology (2024)
  • Best Resident InRad - Guerbet Award, InRad (University of São Paulo) and Guerbet (2024)
  • Magna Cum Laude, University of Ceará - School of Medicine (2021)

Stanford Advisors


All Publications


  • External Validation of an Artificial Intelligence Triaging System for Chest X-Rays: A Retrospective Independent Clinical Study DIAGNOSTICS Castilla, A., D'Amorim, I., Wanderley, M., Esmeraldo, M., Yoshida, A., Eigier, A., Sawamura, M. 2025; 15 (22)

    Abstract

    Background: Chest radiography (CXR) is the most frequently performed radiological exam worldwide, but reporting backlogs, caused by a shortage of radiologists, remain a critical challenge in emergency care. Artificial intelligence (AI) triage systems can help alleviate this challenge by differentiating normal from abnormal studies and prioritizing urgent cases for review. This study aimed to externally validate TRIA, a commercial AI-powered CXR triage algorithm (NeuralMed, São Paulo, Brazil). Methods: TRIA employs a two-stage deep learning approach, comprising an image segmentation module that isolates the thoracic region, followed by a classification model trained to recognize common cardiopulmonary pathologies. We trained the system on 275,399 CXRs from multiple public and private datasets. We performed external validation retrospectively on 1045 CXRs (568 normal and 477 abnormal) from a teaching university hospital that was not used for training. We established ground truth using a large language model (LLM) to extract findings from original radiologist reports. An independent radiologist review of a 300-report subset confirmed the reliability of this method, achieving an accuracy of 0.98 (95% CI 0.978-0.988). We compared four ensemble decision strategies for abnormality detection. Performance metrics included sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUROC) with 95% CI. Results: The general abnormality classifier achieved strong performance (AUROC 0.911). Individual pathology models for cardiomegaly, pneumothorax, and effusion showed excellent results (AUROC of 0.968, 0.955, and 0.935, respectively). The weighted ensemble demonstrated the best balance, with an accuracy of 0.854 (95% CI, 0.831-0.874), a sensitivity of 0.845 (0.810-0.875), a specificity of 0.861 (0.830-0.887), and an AUROC of 0.927 (0.911-0.940). Sensitivity-prioritized methods achieving sensitivity >0.92 produced lower specificity (<0.69). False negatives were mainly subtle or equivocal cases, although many were still flagged as abnormal by the general classifier. Conclusions: TRIA achieved robust and balanced accuracy in distinguishing normal from abnormal CXRs. Integrating this system into clinical workflows has the potential to reduce reporting delays, prioritize urgent cases, and improve patient safety. These findings support its clinical utility and warrant prospective multicenter validation.

    View details for DOI 10.3390/diagnostics15222899

    View details for Web of Science ID 001625925700001

    View details for PubMedID 41300923

    View details for PubMedCentralID PMC12651339

  • Prenatal Diagnosis of VLDLR-associated Cerebellar Hypoplasia via Fetal MRI. Neuropediatrics Povlow, M. R., Esmeraldo, M. A., Dahmoush, H. M., Halverson, M. R., Barsh, G. R., Soares, B. P. 2025

    View details for DOI 10.1055/a-2736-4758

    View details for PubMedID 41213595

  • Intersection of Brain Complexity, Functional Connectivity, and Neuropsychology: A Systematic Review. Cureus de Moura Targino, R., Esmeraldo, M. A., Cajazeiras Falcão, I., Macêdo Melo, K. 2025; 17 (3): e80719

    Abstract

    The definition of brain complexity is based on the principal property of the coexistence of a high degree of integration and differentiation within a single neural system. Despite the fruitful scope of emerging studies involving the applicability of brain complexity metrics, there is a notable scarcity of research focusing on the qualitative characteristics of conscious systems, which are recognized for their high complexity. These qualitative characteristics are expressed in complex cognitive processes, reflecting the interaction between distinct neuropsychological domains, such as attention, memory, language, and executive functions (EFs). Cognitive flexibility and inhibitory control, for instance, emerge from the dynamic integration of distributed neural networks, underscoring the interdependence between brain complexity and cognitive functioning. In light of this, the present study aimed to evaluate how studies addressing measures of functional connectivity and brain complexity, obtained through resting-state functional magnetic resonance imaging (rs-fMRI), relate to neuropsychological aspects. To achieve this, a systematic review was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and based on the PICO (Patient, Intervention, Comparison, Outcome) strategy. Studies were searched in PubMed, CAPES Periodicals, and Virtual Health Library databases to identify relevant studies published between 2019 and March 2024. Articles were included based on study type, sample characteristics, methodological aspects, and specific listed variables. Exclusion criteria encompassed theoretical studies, animal research, and studies involving children/adolescents, as well as those addressing psychiatric conditions, psychoactive substance use, intervention evaluations (e.g., transcranial magnetic stimulation), and disorders of consciousness, due to limitations in applying neuropsychological measures. Possible limitations include the exclusion of studies with specific populations and clinical conditions, which may limit the generalizability of findings to broader, more diverse groups. After applying the selection criteria, 30 articles were chosen and fully analyzed. The results allowed for the establishment of characteristics of the research landscape in this area, initially highlighting a greater number of studies focused on functional connectivity compared to those directed at brain complexity. Additionally, EFs were identified as the most frequently addressed neuropsychological domain in the studies, consistent with the most commonly used evaluative measures in the research: Trail Making Test (TMT), Symbol Digit Modalities Test (SDMT), and verbal fluency tasks. The findings suggest that this is an area of study still in its early stages of development, with notable gaps in the in-depth understanding of the relationships between neural network complexity metrics and neuropsychological functioning.

    View details for DOI 10.7759/cureus.80719

    View details for PubMedID 40242669

    View details for PubMedCentralID PMC12002407

  • Mevalonic Aciduria in a Pediatric Patient: A Case Report and Literature Review of Neuroimaging Findings. Cureus Esmeraldo, M. A., Prates, I. R., Lucato, L. T., Barbosa Junior, A. A. 2024; 16 (7): e65209

    Abstract

    Mevalonic aciduria is a rare autosomal recessive disorder resulting from mevalonate kinase deficiency. Neuroimaging findings associated with the disease have been documented in only a few case reports. We present a case of mevalonic aciduria with both already reported and novel neuroimaging findings and conduct a literature review regarding the role of neuroimaging in the understanding and diagnosis of mevalonate kinase deficiency disorders. The brain magnetic resonance imaging of the reported case revealed several notable findings, including polymicrogyric cortical thickening, an interhypothalamic adhesion or small hypothalamic hamartoma (findings not classically associated with mevalonic aciduria), and mild cerebellar atrophy. This case underscores the significance of recognizing the diverse spectrum of neuroimaging findings associated with the disease, encompassing both well-documented features and those that have not been traditionally reported.

    View details for DOI 10.7759/cureus.65209

    View details for PubMedID 39176373

    View details for PubMedCentralID PMC11340854

  • Intracranial Chondrosarcoma Originating from the Sella Turcica JBNC - Jornal Brasileiro de Neurocirurgia Carlos de Sousa Oliveira, S., Carolina Ferretti Wisenfad, A., Aragão Esmeraldo, M., Medina Guimarães, W., Hiromi Otani, M. 2024
  • Complete C4-C5 Dislocation Secondary to Shallow Water Diving in a Child: A Case-Based Update BRAZILIAN NEUROSURGERY-ARQUIVOS BRASILEIROS DE NEUROCIRURGIA Matos, L., Figueiredo, G., Monteiro, M., Esmeraldo, M., da Ponte, K., Cristino-Filho, G., Leal, P. 2023
  • Late mesenteric ischemia after Sars-Cov-2 infection: case report JORNAL VASCULAR BRASILEIRO Oliveira de Macedo, V., Moreira, G., Fiuza de Albuquerque, A., de Sousa Oliveira, S., Esmeraldo, M., Barroso Barbosa, F. 2021; 20: e20200105

    Abstract

    The purpose of this article is to report the case of a 53-year-old black man, with no previous comorbidities, who presented 48 days after a confirmed diagnosis of COVID-19, complaining of an initially insidious epigastric pain that had progressed to severe pain radiating to the interscapular vertebral region, with hyporexia and episodes of projectile vomiting, with no nausea or fever. Laboratory tests revealed no signs of acute infection or pancreatic injury. Abdominal computed tomography showed dilated, fluid-filled small bowel loops with thickened walls. After clinical treatment, the patient developed persistent abdominal pain. An exploratory laparotomy was performed, finding two sites of small bowel stenosis, with no extrinsic cause, and signs of local ischemia and considerable distension of jejunal and ileal loops. After enterectomy and side-to-side enteroanastomosis, the patient recovered satisfactorily and was discharged with a prescription for oral anticoagulants for outpatient use.

    View details for DOI 10.1590/1677-5449.200105

    View details for Web of Science ID 001003546500017

    View details for PubMedID 34093678

    View details for PubMedCentralID PMC8147701

  • Macroscopic Hematuria as the Initial Presentation of Polycythemia Vera. Cureus Oliveira, S. e., Santos, L. T., Esmeraldo, M. A., Neto, O. s., da Ponte, M. F. 2020; 12 (10): e10800

    Abstract

    Polycythemia vera (PV) is a chronic myeloproliferative neoplasm (CMN) characterized by the accumulation of red blood cells, granulocytes and platelets in the peripheral blood. The main complications of PV are an increased risk of thrombosis, bleeding and transformation to myelodysplasia or acute leukemia. The authors report the case of a 28-year-old man with a complaint of macroscopic hematuria, low back pain and edema of the left arm associated with elevated hemoglobin, hematocrit and lactic dehydrogenase, leukocytosis and increased renal volume. Computed tomography of the chest and abdomen with contrast showed venous ectasia in the left upper limb and thrombosis of the right renal vein with extension to the inferior vena cava. A diagnosis of PV was confirmed by the presence of the JAK2 mutation and a bone marrow biopsy that showed panmyelosis. The patient was anticoagulated and treatment for PV was started with aspirin, phlebotomy and hydroxyurea. Then, the patient was discharged for outpatient follow-up with a hematologist. The case emphasizes the importance of clinical suspicion for atypical presentation of the disease in an unusual age range and of adequate etiological investigation of thrombosis in unusual sites.

    View details for DOI 10.7759/cureus.10800

    View details for PubMedID 33163304

    View details for PubMedCentralID PMC7641481

  • Pseudoaneurysm of the superficial temporal artery: Report of a case JOURNAL OF ORAL AND MAXILLOFACIAL SURGERY Webber, C. M., Wind, G. G., Burton, R. G. 1997; 55 (2): 166-169

    View details for DOI 10.1016/S0278-2391(97)90236-6

    View details for Web of Science ID A1997WG08200016

    View details for PubMedID 9024354