A clinically oriented approach to smooth pursuit eye movement quantitative evaluation

Contribution of cerebellum and brainstem in the control of eye movement: evidence from a functional study in a clinical model

The idiopathic cerebellar ataxias (IDCA) comprise a wide spectrum of neurodegenerative diseases with heterogeneous neuropathology, characterized by the negativity of search for any known genetic mutation. On the basis of both their clinical presentation and their magnetic resonance imaging pattern, patients with IDCA can be subdivided into patients with a purely cerebellar syndrome and atrophy of the cerebellum (IDCA-C) and patients with additional noncerebellar symptoms and atrophy of both cerebellum and brainstem (IDCA-P). The aim of the present study was to evaluate the disaggregated contribution of brainstem and cerebellum in the control of eye movements, by means of an extensive battery of quantitative tests covering most oculomotor subfunctions related to lesions of the cerebellum and the brainstem. The smooth-pursuit movement analysis showed a decrease in gain and magnitude in both subgroups of IDCA with respect to normal controls, without any significant differences in the prevalence pattern between the two subgroups; the mean values of these parameters, however, were significantly lower in IDCA-P than in IDCA-C subjects in both gain (P < 0.01) and magnitude (P < 0.001). No statistically significant difference was observed between the two subgroups in the analysis of saccadic movements or in the other parameters investigated. The distinction between IDCA-P and IDCA-C subgroups has clinical implications, as a poorer prognosis is related to brainstem involvement, which may occur late in the course of the disease. Thus, the possibility to detect the brainstem involvement, also in association with cerebellar impairment, by a relatively simple eye-movement analysis, potentially useful mainly in follow-up investigations, needs to be evaluated further.

Eye movement abnormalities in myotonic dystrophy

We studied saccade and smooth pursuit eye movements in 31 patients suffering from myotonic dystrophy (MD). On the basis of mean value comparisons, saccades were slower and hypometric and smooth pursuit eye movements performed worse in MD patients than in controls. On an individual basis, saccade duration was prolonged in 67.7%, saccades were hypometric in 19.4%, saccade latency was delayed in 9.7%, and the smooth pursuit performance index was decreased in 9.7% of patients. Eye movement abnormalities did not correlate with those detectable by visual, brain-stem auditory and somatosensory evoked potentials. We attempted to classify eye movement abnormalities as myogenic or neurogenic on the basis of differences in combination of eye movement abnormalities and the occurrence of D5/D35 dissociation; the latter consists of a prolonged duration for large (35 degrees) but not for small (5 degrees) saccades. Since D5/D35 dissociation occurred in 26/33 multiple sclerosis patients with increased saccade duration, we considered it to be a neurogenic pattern attributable to a central nervous system (CNS) dysfunction. A prolonged duration without dissociation especially in combination with saccade hypometria, is interpreted as a myogenic pattern, although the lack of dissociation may also occur with CNS impairment in case of a marked increase in saccade duration. Accordingly we classified the oculomotor abnormalities detected as neurogenic in 11 MD patients and as myogenic in another 10, but in some subjects belonging to the second group concomitant CNS impairment is not to be excluded.

Identification of input variables for feature based artificial neural networks-saccade detection in EOG recordings

Though artificial neural networks (ANN) are excellent tools for pattern recognition problems when signal to noise ratio is low, the identification of decision relevant features for ANN input data is still a crucial issue. The experience of the ANN designer and the existing knowledge and understanding of the problem seem to be the only links for a specific construction. In the present study a backpropagation ANN based on modified raw data inputs showed encouraging results. Investigating the specific influences of prototypical input patterns on a specially designed ANN led to a new sparse and efficient input data presentation. This data coding obtained by a semiautomatic procedure combining existing expert knowledge and the internal representation structures of the raw data based ANN yielded a list of feature vectors, each representing the relevant information for saccade identification. The feature based ANN produced a reduction of the error rate of nearly 40% compared with the raw data ANN. An overall correct classification of 92% of so far unknown data was realized. The proposed method of extracting internal ANN knowledge for the production of a better input data representation is not restricted to EOG recordings, and could be used in various fields of signal analysis.