Ensembles of data-efficient vision transformers as a new paradigm for automated classification in ecology.

Monitoring biodiversity is paramount to manage and protect natural resources. Collecting images of organisms over large temporal or spatial scales is a promising practice to monitor the biodiversity of natural ecosystems, providing large amounts of data with minimal interference with the environment. Deep learning models are currently used to automate classification of organisms into taxonomic units. However, imprecision in these classifiers introduces a measurement noise that is difficult to control and can significantly hinder the analysis and interpretation of data. We overcome this limitation through ensembles of Data-efficient image Transformers (DeiTs), which not only are easy to train and implement, but also significantly outperform the previous state of the art (SOTA). We validate our results on ten ecological imaging datasets of diverse origin, ranging from plankton to birds. On all the datasets, we achieve a new SOTA, with a reduction of the error with respect to the previous SOTA ranging from 29.35% to 100.00%, and often achieving performances very close to perfect classification. Ensembles of DeiTs perform better not because of superior single-model performances but rather due to smaller overlaps in the predictions by independent models and lower top-1 probabilities. This increases the benefit of ensembling, especially when using geometric averages to combine individual learners. While we only test our approach on biodiversity image datasets, our approach is generic and can be applied to any kind of images.

Does superficial fat affect metabolite concentrations determined by MR spectroscopy with water referencing?

It has recently been reported in this journal that local fat depots produce a sizable frequency-dependent signal attenuation in magnetic resonance spectroscopy (MRS) of the brain. If of a general nature, this effect would question the use of internal reference signals for quantification of MRS and the quantitative use of MRS as a whole. Here, it was attempted to verify this effect and pinpoint the potential causes by acquiring data with various acquisition settings, including two field strengths, two MR scanners from different vendors, different water suppression sequences, RF coils, localization sequences, echo times, and lipid/metabolite phantoms. With all settings tested, the reported effect could not be reproduced, and it is concluded that water referencing and quantitative MRS per se remain valid tools under common acquisition conditions.