Nicole Fernandes

Centro Hospitalar de Vila Nova de Gaia

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Uma Sriram

Temple University

Raghava Potula

Temple University

Adam M. Staffaroni

University of California, San Francisco

Joel H. Kramer

University of California, San Francisco

Fanny M. Elahi

Icahn School of Medicine at Mount Sinai

Bruce L. Miller

University of California, San Francisco

Kaitlin B. Casaletto

University of California, San Francisco

Nivetha Brathaban

University of California, San Francisco

Viviana Zuluaga‐Ramirez

Temple University

Yuri Persidsky

Temple University

Cooperative Institutions

University of California, San Francisco

University of Cape Town

Temple University

University Memory and Aging Center

Groote Schuur Hospital

University College London

Alzheimer’s Disease Neuroimaging Initiative

Universidade Federal do Rio de Janeiro

Cairo University

University of Birmingham

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Can Tree Based Approaches Surpass Deep Learning in Anomaly Detection? A Benchmarking Study

arXiv (Cornell University) (2024)

Santonu Sarkar Shanay Mehta Nicole Fernandes Jyotirmoy Sarkar Snehanshu Saha

Detection of anomalous situations for complex mission-critical systems holds paramount importance when their service continuity needs to be ensured. A major challenge in detecting anomalies from the operational data arises due to the imbalanced class distribution problem since the anomalies are supposed to be rare events. This paper evaluates a diverse array of machine learning-based anomaly detection algorithms through a comprehensive benchmark study. The paper contributes significantly by conducting an unbiased comparison of various anomaly detection algorithms, spanning classical machine learning including various tree-based approaches to deep learning and outlier detection methods. The inclusion of 104 publicly available and a few proprietary industrial systems datasets enhances the diversity of the study, allowing for a more realistic evaluation of algorithm performance and emphasizing the importance of adaptability to real-world scenarios. The paper dispels the deep learning myth, demonstrating that though powerful, deep learning is not a universal solution in this case. We observed that recently proposed tree-based evolutionary algorithms outperform in many scenarios. We noticed that tree-based approaches catch a singleton anomaly in a dataset where deep learning methods fail. On the other hand, classical SVM performs the best on datasets with more than 10% anomalies, implying that such scenarios can be best modeled as a classification problem rather than anomaly detection. To our knowledge, such a study on a large number of state-of-the-art algorithms using diverse data sets, with the objective of guiding researchers and practitioners in making informed algorithmic choices, has not been attempted earlier.

Benchmarking

Anomaly (physics)

Tree (set theory)

10.48550/arxiv.2402.07281

Cite

Citations (0)

Plasma Glial Fibrillary Acidic Protein Levels Differ Along the Spectra of Amyloid Burden and Clinical Disease Stage1

Journal of Alzheimer s Disease (2020)

Breton M. Asken Fanny M. Elahi Renaud La Joie Amelia Strom Adam M. Staffaroni

Background: Measuring plasma glial fibrillary acidic protein (GFAP) alongside cortical amyloid-β (Aβ) may shed light on astrocytic changes in aging and Alzheimer’s disease (AD). Objective: To examine associations between plasma GFAP and cortical Aβ deposition in older adults across the typical aging-to-AD dementia spectrum. Methods: We studied two independent samples from UCSF (Cohort 1, N = 50; Cohort 2, N = 37) covering the spectra of clinical severity (CDR Sum of Boxes; CDR-SB) and Aβ-PET burden. Aβ-PET was completed with either florbetapir or Pittsburgh Compound B and standardized uptake value ratios were converted to the Centiloid (CL) scale for analyses. All participants with CDR-SB > 0 were Aβ-PET positive, while clinically normal participants (CDR-SB = 0) were a mix of Aβ-PET positive and negative. Regression analyses evaluated main effect and interaction associations between plasma GFAP, Aβ-PET, and clinical severity. Results: In both cohorts, plasma GFAP increased linearly with Aβ-PET CLs in clinically normal older adults. In Cohort 2, which included participants with more severe clinical dysfunction and Aβ-PET burden, the association between Aβ and GFAP became curvilinear (inverted U-shape; quadratic model R2 change = 0.165, p = 0.009), and Aβ-PET interacted with CDR-SB (R2 change = 0.164, p = 0.007): older adults with intermediate functional impairment (CDR-SB = 0.5–4.0) showed a weak (negative) association between Aβ-PET CLs and plasma GFAP, while older adults with dementia (CDR-SB > 4.0) showed a strong, negative association of higher Aβ-PET CLs with lower plasma GFAP. Conclusion: The relationship between astrocytic integrity and cortical Aβ may be highly dynamic, with linear, positive associations early in disease that diverge in more severe disease stages.

Amyloid (mycology)

10.3233/jad-200755

Cite

Citations (61)

The Value of Cardiopulmonary Exercise Testing As A Prognostic Indicator For Head And Neck Cancer Patients Considered For Primary Surgery

British Journal of Oral and Maxillofacial Surgery (2018)

Nicole Fernandes Mohammed Hussain Michael Ho Anastasios Kanatas Gillon Fabbroni

Risk Stratification

10.1016/j.bjoms.2018.10.177

Cite

Citations (1)

Bilateral pneumothorax complicated by extraperitoneal air

BMJ Case Reports (2014)

Dianne E.C. van Beek Nicole Fernandes Mark G. Hazekamp Karin Hogenbirk

An 11-month-old male child with Down syndrome was admitted to the pediatric intensive care unit (PICU) after an uncomplicated correction of his complete atrioventricular septal defect. On postoperative day 5, there was an acute incident after administration of an enema which started with a decrease in tidal volumes. Eventually, there was no air entry, resulting in desaturation and subsequently a bradycardia with no cardiac output. Cardiopulmonary resuscitation was …

10.1136/bcr-2014-205463

Cite

Citations (0)

Lower White Matter Volume and Worse Executive Functioning Reflected in Higher Levels of Plasma GFAP among Older Adults with and Without Cognitive Impairment

Journal of the International Neuropsychological Society (2021)

Breton M. Asken Lawren VandeVrede Julio C. Rojas Corrina Fonseca Adam M. Staffaroni

Abstract Objective: There are minimal data directly comparing plasma neurofilament light (NfL) and glial fibrillary acidic protein (GFAP) in aging and neurodegenerative disease research. We evaluated associations of plasma NfL and plasma GFAP with brain volume and cognition in two independent cohorts of older adults diagnosed as clinically normal (CN), mild cognitive impairment (MCI), or Alzheimer’s dementia. Methods: We studied 121 total participants (Cohort 1: n = 50, age 71.6 ± 6.9 years, 78% CN, 22% MCI; Cohort 2: n = 71, age 72.2 ± 9.2 years, 45% CN, 25% MCI, 30% dementia). Gray and white matter volumes were obtained for total brain and broad subregions of interest (ROIs). Neuropsychological testing evaluated memory, executive functioning, language, and visuospatial abilities. Plasma samples were analyzed in duplicate for NfL and GFAP using single molecule array assays (Quanterix Simoa). Linear regression models with structural MRI and cognitive outcomes included plasma NfL and GFAP simultaneously along with relevant covariates. Results: Higher plasma GFAP was associated with lower white matter volume in both cohorts for temporal (Cohort 1: β = −0.33, p = .002; Cohort 2: β = −0.36, p = .03) and parietal ROIs (Cohort 1: β = −0.31, p = .01; Cohort 2: β = −0.35, p = .04). No consistent findings emerged for gray matter volumes. Higher plasma GFAP was associated with lower executive function scores (Cohort 1: β = −0.38, p = .01; Cohort 2: β = −0.36, p = .007). Plasma NfL was not associated with gray or white matter volumes, or cognition after adjusting for plasma GFAP. Conclusions: Plasma GFAP may be more sensitive to white matter and cognitive changes than plasma NfL. Biomarkers reflecting astroglial pathophysiology may capture complex dynamics of aging and neurodegenerative disease.

Neuropsychological Assessment

Cognitive Decline

Executive dysfunction

Voxel-based morphometry

10.1017/s1355617721000813

Cite

Citations (24)

Author Correction: Elevated complement mediator levels in endothelial-derived plasma exosomes implicate endothelial innate inflammation in diminished brain function of aging humans

Scientific Reports (2022)

Fanny M. Elahi Danielle Harvey Marie Altendahl Nivetha Brathaban Nicole Fernandes

Mediator

Complement

Inflammatory mediator

Brain Function

10.1038/s41598-022-05977-3

Cite

Citations (0)

An outpatient approach to home mechanical ventilation follow-up

Luís Lázaro Ferreira Nicole Fernandes Carla Nogueira Daniela Ferreira Sara Conde

The COVID-19 pandemic brought the outpatient management to the spotlight, especially in what home mechanical ventilation (HMV) is regarded. Our goal was to assess the main complaints/problems and the adjustments made in the appointment. We performed a transversal retrospective analysis of patients on HMV for at least a month, followed in the outpatient clinic of a tertiary hospital, in 2019’s 2nd semester. The HMV outpatient clinic consists of a pulmonologist, a nurse and a technician from the home respiratory care company (provider of HMV in Portugal). In a day-hospital regimen, patients are monitored on HMV with their equipment for at least 30 minutes with blood gas analysis and/or capnography. Ventilator data is observed in real time and also collected from the previous 3 months. A total of 301 patients were analyzed. No changes were made in 138 cases (45.8%). A total of 212 changes were made in the remainder 163 patients. Most detected problems were found in HMV software data (33.5%), such as usage, leakage and volumes. HMV parameters suffered the most adjustments (36.3%). Only 5 problems (2.4%) lead to stop HMV. All results are shown in this table: Almost half of the patients needed no changes. On the other hand, our results show how an outpatient approach to HMV follow-up allows clinicians to detect a diverse amount of patients’ complaints or problems regarding the treatment itself and, at the same time, address changes to try to fix them.

Pulmonologists

Technician

Outpatient clinic

Regimen

Capnography

10.1183/13993003.congress-2021.pa3577

Cite

Citations (0)

Is there a pattern of compliance to home mechanical ventilation depending on disease?

Luís Lázaro Ferreira Nicole Fernandes Carla Nogueira Daniela Ferreira Sara Conde

Outcomes of home mechanical ventilation (HMV) are related to treatment compliance. Our goal was to evaluate the adherence to HMV across different diseases in general and then according to groups of compliance. We performed a transversal retrospective analysis of patients on HMV for at least a month, followed in the outpatient clinic of a tertiary hospital, in 2019’s 2nd semester. Compliance data were obtained from ventilator software. A total of 301 patients were analyzed with no tracheostomized patients. The daily average usage was <8h in 159 patients (52.8%), 8-16h in 135 (44.9%) and >16h in 7 (2.3%). There was no difference in compliance between gender (female 7.2 [5.5-9.3] vs male 8.0 [6.2-9.4]; p=0.115). Patients on HMV for >6 months had a higher compliance (8.0 [6.0-9.5] vs 6.2 [3.3-8.1; p <0.001]. All other results are described in the following table: Overall, patients are highly adherent to HMV. Chest wall disease patients were statistically longer on HMV than chronic obstructive pulmonary disease, neuromuscular disease and interstitial lung disease patients. Almost half of the patients are on HMV between 8 and 16 hours and 2.3% are almost dependent on it. There was no statistically significant difference in compliance to HMV across different diseases.

Outpatient clinic

Noninvasive Ventilation

Pulmonary compliance

10.1183/13993003.congress-2021.pa3573

Cite

Citations (0)

Who are the chronic pulmonary obstructive disease patients highly compliant to home mechanical ventilation?

Luís Lázaro Ferreira Nicole Fernandes Carla Nogueira Daniela Ferreira Sara Conde

Adherence to treatment in chronic pulmonary obstructive disease (COPD), including home mechanical ventilation (HMV), is essential to optimize management. Our goal was to characterize COPD patients according to their HMV adherence. We performed a transversal retrospective analysis of COPD patients in HMV for at least a month, followed in the outpatient clinic of a tertiary hospital, between July and December 2019. Compliance data were obtained from ventilator software, recent lung function test data from clinical records and blood gas analysis data from the appointment day. A total of 139 COPD patients were analyzed. There were 114 (82%) patients with average daily usage ≥5 hours and 68 (48.9%) with ≥8 hours. We defined high compliance as ≥8 hours/daily according to our mean average (7.7±3.1). Results are described in the table: Results were similar with a 5 hours/daily cut-off. Overall, COPD patients are highly adherent to HMV. The high compliant group was older, longer on HMV and had less hospitalizations in the previous year; whether this might represent a cause or a consequence of their compliance is uncertain. Although the patients with higher usage had slightly better blood gas analysis, this was not statistically significant, neither were the ventilatory parameters. Mortality was significant in this group, interestingly higher in the high compliance group (although non statistical).

Obstructive lung disease

Outpatient clinic

10.1183/13993003.congress-2021.pa3570

Cite

Citations (0)

Tropical rhodolith beds are a major and belittled reef fish habitat

Scientific Reports (2021)

Rodrigo L. Moura Maria L. Abieri Guilherme M. Castro Lélis Antonio Carlos-Júnior Pamela M. Chiroque-Solano

Abstract Understanding habitat-level variation in community structure provides an informed basis for natural resources’ management. Reef fishes are a major component of tropical marine biodiversity, but their abundance and distribution are poorly assessed beyond conventional SCUBA diving depths. Based on a baited-video survey of fish assemblages in Southwestern Atlantic’s most biodiverse region we show that species composition responded mainly to the two major hard-bottom megahabitats (reefs and rhodolith beds) and to the amount of light reaching the bottom. Both megahabitats encompassed typical reef fish assemblages but, unexpectedly, richness in rhodolith beds and reefs was equivalent. The dissimilar fish biomass and trophic structure in reefs and rhodolith beds indicates that these systems function based on contrasting energy pathways, such as the much lower herbivory recorded in the latter. Rhodolith beds, the dominant benthic megahabitat in the tropical Southwestern Atlantic shelf, play an underrated role as fish habitats, and it is critical that they are considered in conservation planning.

Marine protected area

Scuba diving

10.1038/s41598-020-80574-w

Cite

Citations (60)