As you may noticed, it is some time since the last "Paper of the Day". The main point of it, was a decreasing motivation in reading old papers. Of course there are some hidden diamonds between them, but I am not a person who lives in the past. So I changed my focus to newly published papers. I created email alerts for a lot of different journals and I must say: I love it. I will not say that the quality is better as in papers from around 1950... . The only benefit of reading new published paper is the knowledge that the problem discussed in the paper is a topic/question/problem which brings restless nights to at least one scientist in the world in this moment. This is a comfort thought, at least when you are a PhD student like I am :D Sometimes, problems which experimentalists are dealing with, are so general that it is sort of funny to write about it. This is why this paper from Peter Elliott et al., published in PLOS ONE (5. July), caught my attention. It describes a new method to detect mouse scratching. For all the readers which like me think “What? Mouse scratching?”, let me explain: yes indeed, I mean the behaviour mice (and other animals including humans) are doing when they feel itchy. There are a lot of illnesses in humans, which can lead to (chronic) itching. And of course there are some scientists doing animal experiments (with mice) in order to find better treatment for affected individuals. For such experiments it is important to analyse the scratching behaviour of the mice, because it is the strongest clue for itchy feeling in the mice. Normally these means a lot of videos and a lot of work analysing them… by a human who watches the videos or paw movement tracking algorithms. But videos need light and as mice are nocturnal animals this creates the problem that the light induces stress in the mice which may affect the scratching behaviour. So the question is: How can we detect a mice scratching without light? One possibility is “to place a metal ring around the mouse’s hind paws and track perturbations in an electromagnetic field in the cage”, but this would detect ANY paw movement. The paper I present today, investigates the value of sound analyzation in the problem of detecting mouse scratches. A mouse scratching consists of a small set of rapid paw movements (of around 50 ms). These typical rhythmic pattern of the often 200- 500ms long sets of scratches, can be identified in by applying a filter to a time-frequency representation of the audio recording. Together with some supervised supervised learning algorithm, which helps to distinguish scratching sound from environmental noise and other animal behaviour sounds, they managed to quantify the chloroquine-induced scratching behaviour of a certain mice model (C57BL/6) with reasonable accuracy (85% sensitivity, 75% positive predictive value). Congratulation. "Automated acoustic detection of mouse scratching."
Peter Elliott et al. PloS one 12.7 (2017): e0179662.
0 Kommentare
Hinterlasse eine Antwort. |
IdeaI love to increase my general science knowledge by reading papers from different fields of science. Here I share some of them. Archiv
März 2018
Kategorien
Alle
|