A quantitative performance study of two automatic methods for the diagnosis of ovarian cancer

Vázquez, Manuel A.; Mariño, Inés P.; Blyuss, Oleg; Ryan, Andy; Gentry‐Maharaj, Aleksandra; Kalsi, Jatinderpal; Manchanda, Ranjit; Jacobs, Ian; Menon, Usha; Zaikin, Alexey

doi:10.1016/j.bspc.2018.07.001

Cited by 20 publications

(25 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With that being said, sample collection limited to a single time point, restricts statistical analysis to the comparison with population characteristics as some individual’s previous information cannot be taken into account. Potential approaches to be used may include mixed models, which are commonly applied to the analysis of repeated measurements [52], methods that analyse trend indices to evaluate the dynamics [51], as well as more sophisticated techniques, including Bayesian changepoint models [53,54,55], and recurrent neural networks [56]—one of the most prominent deep learning techniques utilizing serial measurements. The listed approaches could be used to describe serial patterns in infants, with or without allergic diseases, thus enabling improved discrimination.…”

Section: Discussionmentioning

confidence: 99%

Statistical Approaches in the Studies Assessing Associations between Human Milk Immune Composition and Allergic Diseases: A Scoping Review

et al. 2019

Self Cite

View full text Add to dashboard Cite

A growing number of studies are focusing on the associations between human milk (HM) immunological composition and allergic diseases. This scoping review aims to identify statistical methods applied in the field and highlight pitfalls and unmet needs. A comprehensive literature search in MEDLINE and Embase retrieved 13,607 unique records. Following title/abstract screening, 29 studies met the selection criteria and were included in this review. We found that definitions of colostrum and mature milk varied across the studies. A total of 17 out of 29 (59%) studies collected samples longitudinally, but only 12% of these used serial (longitudinal) analyses. Multivariable analysis was used in 45% of the studies, but statistical approaches to modelling varied largely across the studies. Types of variables included as potential confounding factors differed considerably between models. Discrimination analysis was absent from all studies and only a single study reported classification measures. Outcomes of this scoping review highlight lack of standardization, both in data collection and handling, which remains one of the main challenges in the field. Improved standardization could be obtained by a consensus group of researchers and clinicians that could recommend appropriate methods to be applied in future prospective studies, as well as already existing datasets.

show abstract

Section: Discussionmentioning

confidence: 99%

Statistical Approaches in the Studies Assessing Associations between Human Milk Immune Composition and Allergic Diseases: A Scoping Review

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…In particular, recurrent neural networks can integrate information of multiple biomarkers without the need to construct explicit probabilistic models, as opposed to Bayesian analysis methods. This has been recently shown in (Vázquez et al, 2018), where a quantitative performance study of these two approaches for the diagnosis of ovarian cancer from longitudinal biomarker data has been carried out.…”

Section: Emerging Strategies For Early Diagnosis Of Age-related Diseamentioning

confidence: 96%

The Human Body as a Super Network: Digital Methods to Analyze the Propagation of Aging

Whitwell

Bacalini

Blyuss

et al. 2020

Front. Aging Neurosci.

Self Cite

View full text Add to dashboard Cite

Biological aging is a complex process involving multiple biological processes. These can be understood theoretically though considering them as individual networks-e.g., epigenetic networks, cell-cell networks (such as astroglial networks), and population genetics. Mathematical modeling allows the combination of such networks so that they may be studied in unison, to better understand how the so-called "seven pillars of aging" combine and to generate hypothesis for treating aging as a condition at relatively early biological ages. In this review, we consider how recent progression in mathematical modeling can be utilized to investigate aging, particularly in, but not exclusive to, the context of degenerative neuronal disease. We also consider how the latest techniques for generating biomarker models for disease prediction, such as longitudinal analysis and parenclitic analysis can be applied to as both biomarker platforms for aging, as well as to better understand the inescapable condition. This review is written by a highly diverse and multidisciplinary team of scientists from across the globe and calls for greater collaboration between diverse fields of research.

show abstract

“…In recent years machine learning (ML) techniques have become important tools in addressing classification tasks that involve medical problems. As examples, we can mention the use of long short-term memory recurrent neural networks (RNNs) to classify diagnoses from pediatric intensive care unit data (Lipton et al, 2015), the use of RNNs and Bayesian models to discriminate patients with ovarian cancer (Mariño et al, 2017;Vázquez et al, 2018), the use of support vector machines (SVMs) for attention deficit hyperactivity disorder prediction (Dai et al, 2012), the application of convolutional neural networks (CNNs) to classifying electroencephalogram (EEG) signals for emotion recognition (Luo et al, 2020), or the combination of multilayer perceptrons and SVMs to diagnose major depressive disorders (Saeedi et al, 2020b).…”

Section: Introductionmentioning

confidence: 99%

An Interpretable Machine Learning Method for the Detection of Schizophrenia Using EEG Signals

Vázquez

Maghsoudi

Mariño

2021

Front. Syst. Neurosci.

Self Cite

View full text Add to dashboard Cite

In this work we propose a machine learning (ML) method to aid in the diagnosis of schizophrenia using electroencephalograms (EEGs) as input data. The computational algorithm not only yields a proposal of diagnostic but, even more importantly, it provides additional information that admits clinical interpretation. It is based on an ML model called random forest that operates on connectivity metrics extracted from the EEG signals. Specifically, we use measures of generalized partial directed coherence (GPDC) and direct directed transfer function (dDTF) to construct the input features to the ML model. The latter allows the identification of the most performance-wise relevant features which, in turn, provide some insights about EEG signals and frequency bands that are associated with schizophrenia. Our preliminary results on real data show that signals associated with the occipital region seem to play a significant role in the diagnosis of the disease. Moreover, although every frequency band might yield useful information for the diagnosis, the beta and theta (frequency) bands provide features that are ultimately more relevant for the ML classifier that we have implemented.

show abstract

A quantitative performance study of two automatic methods for the diagnosis of ovarian cancer

Cited by 20 publications

References 26 publications

Statistical Approaches in the Studies Assessing Associations between Human Milk Immune Composition and Allergic Diseases: A Scoping Review

Statistical Approaches in the Studies Assessing Associations between Human Milk Immune Composition and Allergic Diseases: A Scoping Review

The Human Body as a Super Network: Digital Methods to Analyze the Propagation of Aging

An Interpretable Machine Learning Method for the Detection of Schizophrenia Using EEG Signals

Contact Info

Product

Resources

About