Automatic eye fixations identification based on analysis of variance and covariance

Abstract:

Eye movement is the simplest and repetitive movement that enables humans to interact with the environment. The common daily activities, such as reading a book or watching television, involve this natural activity, which consists of rapidly shifting our gaze from one region to another. In clinical application, the identification of the main components of eye movement during visual exploration, such as fixations and saccades, is the objective of the analysis of eye movements: however, in patients affected by motor control disorder the identification of fixation is not banal.

This work [download] presents a new fixation identification algorithm based on the analysis of variance and covariance: the main idea was to use bivariate statistical analysis to compare variance over x and y to identify fixation. We describe the new algorithm, and we compare it with the common fixations algorithm based on dispersion. To demonstrate the performance of our approach, we tested the algorithm in a group of healthy subjects and patients affected by motor control disorder.



Comments:

In the last decade a large effort has been made to identify fixations [1-4], however it is not yet easy to provide a formal mathematical definition of fixation: some authors have demonstrated that fixation’s parameters depend strictly by the type of task [5-8].

We suggested a formal definition of fixations based on analysis of variance between x axis and y axis; the implemented algorithm is based on the dispersion algorithm I-DT developed by Salvucci and Goldberg [9] and integrates it with a statistical test (F-test) and covariance.

The main advantage of the proposed technique is to provide a new definition of fixation which does not require the setting of any critical parameter or threshold, and provides a probability value of correctness.

Relationship between the modified Rankin Scale and the Barthel Index in the process of functional recovery after stroke

The modified Rankin Scale (mRS) and the Barthel Index (BI) are the most  
common clinimetrical instruments for measuring disability after stroke. 

This study [here] investigated the relationship between the BI and the 

mRS at multiple time points after stroke. 

The BI, which is a widely used instrument for longitudinal follow-up post-

stroke, was used as  reference to determine the effect of time on the 

sensitivity of the mRS in differentiating  functional recovery. 


Methods: Ninety-two patients with first stroke and hemispheric brain 

lesion were evaluated using the BI and mRS at 10 days, 3 and 6 months. 

The Kruskal-Wallis test was applied to examine median differences in BI 

among the mRS levels at 10 days, 3 and 6 months with Dunn's correction 

for multigroup comparison. The Mann and Whitney test was used to 

compare median differences in BI scores between two aggregations of 

mRS grades (mRS=0-2, mRS=3-5) at the same time periods after 

stroke. 


Results: BI score distribution amongst mRS grades overlapped at 10 

days, differentiating only between extreme grades (no disability vs severe 

disability). At 3 months, independent patients with slight disability could be 

distinguished from dependent patients with marked disability. At 6 

months, grade 2 and 3 overlapped no more, differentiating independence 

(class 0-2) from dependence (class 3-5). The largest transition to an 

independent functional status occurred from grade 4, at 3 months.


Conclusion: Maximum sensitivity of mRS in differentiating functional 

recovery is reached at six months post-stroke.