A fine motor test involves the manipulation of smaller objects with fingers, hands, and wrists. This test is an integral part of the evaluation of an upper extremity function. Nine Hole Peg Test (NHPT) is one among such tests which assess the ability to manipulate pegs with the thumb and finger. There is a need to develop a fine motor assessment tool which is reproducible and mimics closely the natural movement of hands. The aim of this work is to develop an electronic pegboard which is easy to administer and efficient in terms of time. Pegboard device is modified and standardized by (1) Adding electronic circuits to custom-made pegboard and programmed using a microcontroller (ATmega2560), (2) Following a specific sequence in placing and picking the pegs from the board, and (3) Using Infrared sensor and robust algorithm to ensure one peg movement at a time. The setup is administered on 15 healthy participants (nine females, six males aged between 21 and 80) and the outcome is compared with the results of traditional NHPT. Predefined sequence in moving the pegs and electronic timer features provide reliable results for repeated measurements and facilitate storing test score in a digital repository. This data could be used as reference data during the follow-up visits. The maximum difference between the measured timing between the present setup and traditional NHPT is about 6.7%. It is important to note that, due to inherent delay (response time) in the traditional NHPT, when compared to present setup the measured timing is always on the higher side. Nondependency on the manual stopwatch to record the time and hands-free of any wearable device are the advantages of the present setup.
Detecting named entities in user generated text is a challenging task. Lab protocols specify steps in performing a lab procedure. The majority of wet lab protocols are written in noisy, dense, and domain-specific natural language. There is a growing need of automatic or semi-automatic conversion of protocols into machine-readable format to benefit biological research.The paper describes how a classifier model built using Conditional Random Field[1] detects named entities in wet lab protocols. The model 1 trained on the training data showed precision, recall and F1-score of 0.762, 0.743 and 0.752 respectively on the development set. When applied to unseen test data, the model showed 0.737, 0.640 and 0.685 respectively.
This paper describes the system submitted in the SemEval-2021 Statement Verification and Evidence Finding with Tables task. The system relies on candidate generation for logical forms on the table based on keyword matching and dependency parsing on the claim statements.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.