Deep feature learning for soft tissue sarcoma classification in MR images via transfer learning

Hermessi, Haithem; Mourali, Olfa; Zagrouba, Ezzeddine

doi:10.1016/j.eswa.2018.11.025

Cited by 58 publications

(24 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sarcoma : To date, radiomics has been used in the classification ( 58 ) and prediction of metastasis for sarcomas ( 59 ); additionally, nomograms have been used in survival prediction ( 60 ) for sarcomas. There have been relatively few radiomic analyses of sarcomas, since the prevalence of this disease is relatively low.…”

Section: Discussionmentioning

confidence: 99%

Radiomics in Oncology: A 10-Year Bibliometric Analysis

Ding

Wu²,

Liao³

et al. 2021

Front. Oncol.

View full text Add to dashboard Cite

ObjectivesTo date, radiomics has been applied in oncology for over a decade and has shown great progress. We used a bibliometric analysis to analyze the publications of radiomics in oncology to clearly illustrate the current situation and future trends and encourage more researchers to participate in radiomics research in oncology.MethodsPublications for radiomics in oncology were downloaded from the Web of Science Core Collection (WoSCC). WoSCC data were collected, and CiteSpace was used for a bibliometric analysis of countries, institutions, journals, authors, keywords, and references pertaining to this field. The state of research and areas of focus were analyzed through burst detection.ResultsA total of 7,199 pieces of literature concerning radiomics in oncology were analyzed on CiteSpace. The number of publications has undergone rapid growth and continues to increase. The USA and Chinese Academy of Sciences are found to be the most prolific country and institution, respectively. In terms of journals and co-cited journals, Scientific Reports is ranked highest with respect to the number of publications, and Radiology is ranked highest among co-cited journals. Moreover, Jie Tian has published the most publications, and Phillipe Lambin is the most cited author. A paper published by Gillies et al. presents the highest citation counts. Artificial intelligence (AI), segmentation methods, and the use of radiomics for classification and diagnosis in oncology are major areas of focus in this field. Test-retest statistics, including reproducibility and statistical methods of radiomics research, the relation between genomics and radiomics, and applications of radiomics to sarcoma and intensity-modulated radiotherapy, are frontier areas of this field.ConclusionTo our knowledge, this is the first study to provide an overview of the literature related to radiomics in oncology and may inspire researchers from multiple disciplines to engage in radiomics-related research.

show abstract

Section: Discussionmentioning

confidence: 99%

Radiomics in Oncology: A 10-Year Bibliometric Analysis

Ding

Wu²,

Liao³

et al. 2021

Front. Oncol.

View full text Add to dashboard Cite

show abstract

“…This approach has become very popular in the domain of biology and medicine [20]- [24] where the aim is to replace human experts in the diagnostic process. Transfer learning, in the form of ImageNet pretrained feature extractors, also proves to be very useful in the field [21] despite the very different nature of the images being processed. When it comes to image-based species identification, according to Wäldchen and Mäder's review [25], deep learning-based approaches represent current state-of-the-art and pretty much revolutionized the domain.…”

Section: Related Workmentioning

confidence: 99%

Species Identification for Aquatic Biomonitoring Using Deep Residual CNN and Transfer Learning

2021

View full text Add to dashboard Cite

Aquatic insects and other benthic macroinvertebrates are mostly used as bioindicators of the ecological status of freshwaters. However, an expensive and time-consuming process of species identification represents one of the key obstacles for reliable biomonitoring of aquatic ecosystems. In this paper, we proposed a deep learning (DL) based method for species identification that we evaluated on several available public datasets (FIN-Benthic, STONEFLY9, and EPT29) along with our Chironomidae dataset (CHIRO10). The proposed method relies on three DL techniques used to improve robustness when training is done on a relatively small dataset: transfer learning, data augmentation, and feature dropout. We applied transfer learning by employing ResNet-50 deep convolutional neural network (CNN) pretrained on ImageNet 2012 dataset. The results show significant improvement compared to original contributions and confirms that there is a considerable gain when there are multiple images per specimen.

show abstract

“…The use of convolutional neural networks is justified because the dating of bruise has low accuracy rates, thus reaching 40% for bruises less than 48 hours, a percentage that decreases as the age of the bruise increases [7]. In addition, they present good results, comparable to medical specialists, for similar problems such as sarcoma [24] and melanoma [13] (they affect the skin and its diagnosis is visual based on images).…”

Section: B Ruis E Dating Modelmentioning

confidence: 99%

Bruise dating using deep learning

Tirado

Mauricio

2020

Journal of Forensic Sciences

View full text Add to dashboard Cite

According to the latest report of the World Health Organization (WHO) [1], domestic violence against women is a global scourge, with a prevalence of physical or sexual abuse against women of just over 30% globally and 29.8% in Latin America (see Figure 1). Some victims of violence report these facts to the authorities, which require forensic clinical services such as bruise dating to support complaints, since the results of dating, used as evidence, can be decisive for justice. However, many of these complaints are filed due to lack of evidence or error in the dating of bruise, which aggravates the consequences of this scourge. Bruise, also known as ecchymosis, is the internal bleeding of the skin due to the rupture of blood vessels, which is caused by an impact with an edgeless object, without tearing or cutting the skin. Therefore, blood escapes from blood vessels close to the surface of the skin being trapped under it [2]. This generates a coloration on the surface of the skin, which shows a series of colors that vary over time and can be visible for up to 30 days from its appearance, so the medical literature reports the use of color scales for dating of an bruise. This procedure consists in determining the age of this bruise; thus, the bruise is visualized in situ or through photographs, and the judgment of an expert is applied [3], who, in general, is a coroner, and its importance lies in being evidence fundamental in a trial for domestic violence or against women. There are studies on the dating of bruise, from the medical and forensic point of view, where the use of histological analysis [4], genetic [5], chromatography [6], and visual inspection techniques are

show abstract

Deep feature learning for soft tissue sarcoma classification in MR images via transfer learning

Cited by 58 publications

References 27 publications

Radiomics in Oncology: A 10-Year Bibliometric Analysis

Radiomics in Oncology: A 10-Year Bibliometric Analysis

Species Identification for Aquatic Biomonitoring Using Deep Residual CNN and Transfer Learning

Bruise dating using deep learning

Contact Info

Product

Resources

About