Your browser doesn't support javascript.
Show: 20 | 50 | 100
Results 1 - 7 de 7
Filter
1.
arxiv; 2023.
Preprint in English | PREPRINT-ARXIV | ID: ppzbmed-2301.08888v1

ABSTRACT

Annotating medical images for disease detection is often tedious and expensive. Moreover, the available training samples for a given task are generally scarce and imbalanced. These conditions are not conducive for learning effective deep neural models. Hence, it is common to 'transfer' neural networks trained on natural images to the medical image domain. However, this paradigm lacks in performance due to the large domain gap between the natural and medical image data. To address that, we propose a novel concept of Pre-text Representation Transfer (PRT). In contrast to the conventional transfer learning, which fine-tunes a source model after replacing its classification layers, PRT retains the original classification layers and updates the representation layers through an unsupervised pre-text task. The task is performed with (original, not synthetic) medical images, without utilizing any annotations. This enables representation transfer with a large amount of training data. This high-fidelity representation transfer allows us to use the resulting model as a more effective feature extractor. Moreover, we can also subsequently perform the traditional transfer learning with this model. We devise a collaborative representation based classification layer for the case when we leverage the model as a feature extractor. We fuse the output of this layer with the predictions of a model induced with the traditional transfer learning performed over our pre-text transferred model. The utility of our technique for limited and imbalanced data classification problem is demonstrated with an extensive five-fold evaluation for three large-scale models, tested for five different class-imbalance ratios for CT based COVID-19 detection. Our results show a consistent gain over the conventional transfer learning with the proposed method.

2.
Neutrosophic Sets and Systems ; 48:251-290, 2022.
Article in English | Scopus | ID: covidwho-1824444

ABSTRACT

The overarching structures like intuitionistic fuzzy sets, Pythagorean fuzzy sets, m-polar fuzzy sets, and neutrosophic fuzzy sets etc. have their own inadequacies and impediments. These models are unable to do work because of their impediments in many real life situations. To overcome these deficiencies, in this paper, we introduce a set entitled Pythagorean m-polar fuzzy neutrosophic set (PmFNS), as a hybrid model of Pythagorean fuzzy set, m-polar fuzzy set and single-valued neutrosophic set. We define some notions related to PmFNS with the help of illustrations. We also present some concept of Pythagorean m-polar fuzzy neutrosophic topology alongside its leading characteristics. We render two applications of PmFNS of scarcity of water and uplifting economy ruined due to COVID-19 using TOPSIS. © 2022, Neutrosophic Sets and Systems. All Rights Reserved.

3.
4.
International Journal of Intelligent Systems ; : 34, 2022.
Article in English | Web of Science | ID: covidwho-1680353

ABSTRACT

Modeling uncertainties with multipolar information is an important tool in computational intelligence to address complexities in real-world circumstances. An m-polar fuzzy set (mPFS) is the strong model to express multipolarity with m $m$ membership grades (MGs) in the unit closed interval [ 0 , 1 ] $[0,1]$. A q-rung orthopair fuzzy set (qROFS) is the strong model to express vague and uncertain information with MGs and nonmembership grades (NMGs). The notion of q-rung orthopair m-polar fuzzy set is a new hybrid extension of both mPFS and qROFS. An ROmPFS is a generalized concept that has the ability to deal with multipolarity with m $m$ ordered pairs of MGs and NMGs. Motivated by these robust concepts, in this article, various aggregation operators (AOs) for the aggregation of q-rung orthopair m-polar fuzzy numbers are proposed, including q-rung orthopair m-polar fuzzy weighted averaging operator, symmetric q-rung orthopair m-polar fuzzy weighted averaging operator, q-rung orthopair m-polar fuzzy weighted geometric operator, symmetric q-rung orthopair m-polar fuzzy weighted geometric operator, and q $q$-rung orthopair m-polar fuzzy Maclaurin symmetric mean operator. On the basis of proposed AOs, a robust multicriteria decision-making approach is proposed. An application of proposed AOs is presented to address economic crises during COVID-19. Furthermore, the comparison analysis is designed to discuss the validity and rationality of proposed AOs.

5.
International Journal of Biomathematics ; 13(8):32, 2020.
Article in English | Web of Science | ID: covidwho-1088306

ABSTRACT

The corona virus disease 2019 (COVID-19) has emerged as a fatal virus. This deadly virus has taken the whole world into clutches and many people have embraced death due to this invincible bug. The death toll is rising with every tick of time. The aspiration behind this article is to discover the preventive measure that should be taken to cope with this intangible enemy. We study the prime notions of novel sort of topology accredited Pythagorean m-polar fuzzy topology along with its prime attributes. We slightly amend the well-acknowledged multi-criteria decision analysis tool TOPSIS (Technique for Order of Preference by Similarity to Ideal Solution) to befit the proposed multi-criteria group decision making (MCGDM) problem of exploring the most effective method for curing from COVID-19 employing the proposed model.

6.
arxiv; 2021.
Preprint in English | PREPRINT-ARXIV | ID: ppzbmed-2102.08085v1

ABSTRACT

COVID-19 classification using chest Computed Tomography (CT) has been found pragmatically useful by several studies. Due to the lack of annotated samples, these studies recommend transfer learning and explore the choices of pre-trained models and data augmentation. However, it is still unknown if there are better strategies than vanilla transfer learning for more accurate COVID-19 classification with limited CT data. This paper provides an affirmative answer, devising a novel `model' augmentation technique that allows a considerable performance boost to transfer learning for the task. Our method systematically reduces the distributional shift between the source and target domains and considers augmenting deep learning with complementary representation learning techniques. We establish the efficacy of our method with publicly available datasets and models, along with identifying contrasting observations in the previous studies.

7.
Eur Rev Med Pharmacol Sci ; 24(17): 9172-9181, 2020 Sep.
Article in English | MEDLINE | ID: covidwho-790179

ABSTRACT

OBJECTIVE: Our objective was to find an association between exposure of a population to Middle East Respiratory Syndrome Coronavirus (MERS-CoV) and mortality rate due to Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) across different countries worldwide. MATERIALS AND METHODS: To find the relationship between exposure to MERS-CoV and mortality rate due to SARS-CoV-2, we collected and analyzed data of three possible factors that may have resulted in an exposure of a population to MERS-CoV: (1) the number of Middle East Respiratory Syndrome (MERS) cases reported among 16 countries since 2012; (2) data of MERS-CoV seroprevalence in camels across 23 countries, as working with camels increase risk of exposure to MERS-CoV; (3) data of travel history of people from 51 countries to Saudi Arabia was collected on the assumption that travel to a country where MERS is endemic, such as, Saudi Arabia, could also lead to exposure to MERS-CoV. RESULTS: We found a significantly lower number of Coronavirus disease 2019 (COVID-19) deaths per million (deaths/M) of a population in countries that are likely to be exposed to MERS-CoV than otherwise (t-stat=3.686, p<0.01). In addition, the number of COVID-19 deaths/M of a population was significantly lower in countries that reported a higher seroprevalence of MERS-CoV in camels than otherwise (t-stat=4.5077, p<0.01). Regression analysis showed that increased travelling history to Saudi Arabia is likely to be associated with a lower mortality rate due to COVID-19. CONCLUSIONS: This study provides empirical evidence that a population that was at an increased risk of exposure to MERS-CoV had a significantly lower mortality rate due to SARS-CoV-2, which might be due to cross-protective immunity against SARS-CoV-2 in that population because of an earlier exposure to MERS-CoV.


Subject(s)
Coronavirus Infections/mortality , Pneumonia, Viral/mortality , Betacoronavirus/isolation & purification , COVID-19 , Coronavirus Infections/epidemiology , Coronavirus Infections/pathology , Coronavirus Infections/virology , Humans , Pandemics , Pneumonia, Viral/epidemiology , Pneumonia, Viral/pathology , Pneumonia, Viral/virology , Prevalence , Regression Analysis , SARS-CoV-2 , Seroepidemiologic Studies , Survival Rate
SELECTION OF CITATIONS
SEARCH DETAIL