Benchmarking Scientific Image Forgery Detectors

Cardenuto, João Phillipe; Rocha, Armando Freitas da

doi:10.1007/s11948-022-00391-4

Cited by 4 publications

(3 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While these methods have shown some effectiveness, further research and development are still needed to improve their performance and make them more robust for scientific integrity assessment 25 .…”

Section: Discussionmentioning

confidence: 99%

Unveiling Scientific Articles from Paper Mills with Provenance Analysis

Cardenuto

Moreira

Rocha

2023

Preprint

View full text Add to dashboard Cite

The increasing prevalence of fake publications created by paper mills poses a significant challenge to maintaining scientific integrity. While integrity analysts typically rely on textual and visual clues to identify fake articles, determining which papers merit further investigation can be akin to searching for a needle in a haystack, as these fake publications have non-related authors and are published on non-related venues. To address this challenge, we developed a new methodology for provenance analysis, which automatically tracks and groups suspicious figures and documents. Our approach groups manuscripts from the same paper mill by analyzing their figures and identifying duplicated and manipulated regions. These regions are linked and organized in a provenance graph, providing evidence of systematic production. We tested our solution on a paper mill dataset of hundreds of documents and also on a larger version of the dataset that deliberately included thousands of documents intentionally selected to distract our method. Our approach successfully identified and linked systematically produced articles on both datasets by pinpointing the figures they reused and manipulated from one another. The technique herein proposed offers a promising solution to identify fraudulent manuscripts, and it could be a valuable tool for supporting scientific integrity. All datasets, annotations, trained models, and implementations from this research are freely available at https://github.com/phillipecardenuto/upm.

show abstract

Section: Discussionmentioning

confidence: 99%

Unveiling Scientific Articles from Paper Mills with Provenance Analysis

Cardenuto

Moreira

Rocha

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…One example of this kind is the detection of image manipulation including copy-move forgery. Utilizing computer vision algorithms point descriptors it's possible to identify areas of similarity within e.g., microscopy images that are indicative of image manipulation ( 27 ). AI can be used to combat the surge of mass-produced content often released by paper mills ( 28 ).…”

Section: Ai In the Publishing Domainmentioning

confidence: 99%

Current and future applications of artificial intelligence in surgery: implications for clinical practice and research

Morris,

Fiocco,

Caneva

et al. 2024

Front. Surg.

View full text Add to dashboard Cite

Surgeons are skilled at making complex decisions over invasive procedures that can save lives and alleviate pain and avoid complications in patients. The knowledge to make these decisions is accumulated over years of schooling and practice. Their experience is in turn shared with others, also via peer-reviewed articles, which get published in larger and larger amounts every year. In this work, we review the literature related to the use of Artificial Intelligence (AI) in surgery. We focus on what is currently available and what is likely to come in the near future in both clinical care and research. We show that AI has the potential to be a key tool to elevate the effectiveness of training and decision-making in surgery and the discovery of relevant and valid scientific knowledge in the surgical domain. We also address concerns about AI technology, including the inability for users to interpret algorithms as well as incorrect predictions. A better understanding of AI will allow surgeons to use new tools wisely for the benefit of their patients.

show abstract

“…The risk of false positives and difficulties in determining the true number of duplications should not discourage research into scientific misconduct, encouragingly, three corrigenda have been published by Toxicology Reports which were initiated by the authors after corresponding comments were posted to PubPeer [6][7][8] . Alternative approaches, for example generating synthetic datasets containing manipulated images will likely prove useful to benchmark the performance of automated tools 9 .…”

Section: Risk Of False Positives and False Negativesmentioning

confidence: 99%

A Quantitative Study of Inappropriate Image Duplication in the JournalToxicology Reports

David

2023

Preprint

View full text Add to dashboard Cite

Inappropriate image duplication is a type of scientific error that can be detected by examining published literature. Few estimates of the frequency of this problem have been published. This study aimed to quantify the rate of image duplication in the journal Toxicology Reports. In total 1540 unique articles (identified by DOI) were checked for the presence of research related images (microscopy, photography, western blot scans, etc). Each research paper containing at least one such image was scrutinized for the presence of inappropriate duplications, first by manual review only, and subsequently with the assistance of an AI tool (ImageTwin.ai). Overall, Toxicology Reports published 715 papers containing relevant images, and 115 of these papers contained inappropriate duplications (16%). Screening papers with the use of ImageTwin.ai increased the number of inappropriate duplications detected, with 41 of the 115 being missed during the manual screen and subsequently detected with the aid of the software. In summary, the rate of inappropriate image duplication in this journal has been quantified at 16%, most of these errors could have been detected at peer review by careful reading of the paper and related literature. The use of ImageTwin.ai was able to increase the number of detected problematic duplications.

show abstract

Benchmarking Scientific Image Forgery Detectors

Cited by 4 publications

References 31 publications

Unveiling Scientific Articles from Paper Mills with Provenance Analysis

Unveiling Scientific Articles from Paper Mills with Provenance Analysis

Current and future applications of artificial intelligence in surgery: implications for clinical practice and research

A Quantitative Study of Inappropriate Image Duplication in the JournalToxicology Reports

Contact Info

Product

Resources

About