Retraction Watch

Tracking retractions as a window into the scientific process

Note to authors: Please don’t use the word “novel” when you plagiarize

without comments

compbiomedcoverRetraction Watch Rule 5.1, which governs ironic article titles (and does not actually exist), clearly states that researchers who plagiarize should avoid the use of words like “new” or “novel” when describing their research (or lack thereof). Failure to adhere to Rule 5.1 can lead to embarrassment — as in the case below.

A pair of electrical engineers from Islamic Azad University, in Isfahan, Iran, has lost their 2012 article in Computers in Biology and Medicine, titled “A novel real-time patient-specific seizure diagnosis algorithm based on analysis of EEG and ECG signals using spectral and spatial features and improved particle swarm optimization classifier,” because, well, it wasn’t. Turns out, the researchers lifted data from an Irish group who, several years earlier, had proposed their own “novel algorithm for neonatal seizure detection.”

As the admirably detailed retraction notice explains:

This article has been retracted at the request of the Editor-in-Chief and Author.

The authors have plagiarized part of a paper that had already appeared in Clin. Neurophysiol. 118(6), 2007, pp.1348–1359, http://dx.doi.org/10.1016/j.clinph.2007.02.015.

Furthermore, Table 1 contains values taken from a database produced by the Neonatal Brain Research Group of University College Cork, Ireland, which also correspond with Table 1 of the Clin. Neurophysiol. paper.

Table 4 uses a dataset from Physionet (A. Shoeb, and J. Guttag. Application of Machine Learning to Epileptic Seizure Onset Detection. 27th International Conference on Machine Learning (ICML), June 21–24, 2010, Haifa, Israel).

However, the Cork and Physionet datasets are very different in size and nature: the Cork dataset documents 633 seizures in 10 neonates, whereas the Physionet database contains data for 173 seizures in 23 pediatric patients and one adult patient.

The source of the data actually employed for this study remains unclear, casting doubt on the conclusions drawn in Table 4.

One of the conditions of submission of a paper for publication is that authors declare explicitly that their work is original and has not appeared in a publication elsewhere. Re-use of any data should be appropriately cited. As such this article represents a severe abuse of the scientific publishing system. The scientific community takes a very strong view on this matter and apologies are offered to readers of the journal that this was not detected during the submission process.

Comments
  • Neuroskeptic (@Neuro_Skeptic) January 17, 2013 at 10:36 am

    Notice says “The authors have plagiarized…”

    For once, a journal that doesn’t mince words! It’s shocking how many plagiarism retractions don’t actually contain “The P Word”

  • Nancy Lapid January 17, 2013 at 10:37 am

    Can’t we extend Rule 5.1 — maybe have 5.1.a — to say “Don’t use the word novel to refer to something that’s 10 years old”? Or even more radically, how about: just say “new” instead of “novel” ?

    • chirality January 17, 2013 at 11:30 am

      The American Chemical Society journals, for instance, discourage the use of the words “novel” and “new” in article titles. These words are pleonastic in this context anyway because the very fact that something has been published implies it is new.

  • yasas January 17, 2013 at 11:07 am

    “Cork and Physionet datasets are very different in size and nature: the Cork dataset documents 633 seizures in 10 neonates, whereas the Physionet database contains data for 173 seizures in 23 pediatric patients and one adult patient”.

    So why wasn’t this discrepancy detected during peer-review? Sometimes the journals & reviwers also need to shoulder part of the irresponsibility for allowing such glaring errors go undetected

  • Post a comment

    Threaded commenting powered by interconnect/it code.