Recent advances in Apertium, a free/open-source rule-based machine translation platform for low-resource languages

This paper presents an overview of Apertium, a free and open-source rule-based machine translation platform. Translation in Apertium happens through a pipeline of modular tools, and the platform continues to be improved as more language pairs are added. Several advances have been implemented since the last publication, including some new optional modules: a module that allows rules to process recursive structures at the structural transfer stage, a module that deals with contiguous and discontiguous multi-word expressions, and a module that resolves anaphora to aid translation. Also highlighted is the hybridisation of Apertium through statistical modules that augment the pipeline, and statistical methods that augment existing modules. This includes morphological disambiguation, weighted structural transfer, and lexical selection modules that learn from limited data. The paper also discusses how a platform like Apertium can be a critical part of access to language technology for so-called low-resource languages, which might be ignored or deemed unapproachable by popular corpus-based translation technologies. Finally, the paper presents some of the released and unreleased language pairs, concluding with a brief look at some supplementary Apertium tools that prove valuable to users as well as language developers. All Apertium-related code, including language data, is free/open-source and available at https://github.com/apertium.

Neurophysiological analysis of the clinical pull test

Postural reflexes are impaired in conditions such as Parkinson’s disease, leading to difficulty walking and falls. In clinical practice, postural responses are assessed using the “pull test,” where an examiner tugs the prewarned standing patient backward at the shoulders and grades the response. However, validity of the pull test is debated, with issues including scaling and variability in administration and interpretation. It is unclear whether to assess the first trial or only subsequent repeated trials. The ecological relevance of a forewarned backward challenge is also debated. We therefore developed an instrumented version of the pull test to characterize responses and clarify how the test should be performed and interpreted. In 33 healthy participants, “pulls” were manually administered and pull force measured. Trunk and step responses were assessed with motion tracking. We probed for the StartReact phenomenon (where preprepared responses are released early by a startling stimulus) by delivering concurrent normal or “startling” auditory stimuli. We found that the first pull triggers a different response, including a larger step size suggesting more destabilization. This is consistent with “first trial effects,” reported by platform translation studies, where movement execution appears confounded by startle reflex-like activity. Thus, first pull test trials have clinical relevance and should not be discarded as practice. Supportive of ecological relevance, responses to repeated pulls exhibited StartReact, as previously reported with a variety of other postural challenges, including those delivered with unexpected timing and direction. Examiner pull force significantly affected the postural response, particularly the size of stepping. NEW & NOTEWORTHY We characterized postural responses elicited by the clinical “pull test” using instrumentation. The first pull triggers a different response, including a larger step size suggesting more destabilization. Thus, first trials likely have important clinical and ecological relevance and should not be discarded as practice. Responses to repeated pulls can be accelerated with a startling stimulus, as reported with a variety of other challenges. Examiner pull force was a significant factor influencing the postural response.

THEORETICAL PREDICTION OF THE ROLE OF CO-CONTRACTION DURING BALANCE RECOVERY IN ELDERLY

This study aimed to determine the effect of muscle co-contraction on balance recovery by using a simulation model. The muscle-driven forward simulation model included an inverted pendulum with two ankle muscles, a plantar flexor muscle (PF), and a dorsal flexor muscle (DF). The model was created based on experimental data obtained from balance recovery after applying backward platform translation to a standing elderly woman. Baseline simulation was performed using this model. Additionally, we performed two simulations with increased DF excitation at the same level of simulation and at the same pattern of simulation. The same level of simulation had the same PF excitation level as the baseline simulation with increased DF excitation. The same pattern simulation had the same increased or decreased PF excitation pattern but with a constant increase in PF excitation level to offset the increased DF excitation. Our results revealed that the same pattern simulation decreased the maximum dorsal flexion angle after platform translation. During the same level of simulation, the insufficient PF force used to recover balance resulted in a forward fall. These results imply that co-contraction is an effective strategy for recovering balance at the expense of additional muscle excitation in the elderly.

Sensory reweighting of proprioceptive information of the left and right leg during human balance control

To keep balance, information from different sensory systems is integrated to generate corrective torques. Current literature suggests that this information is combined according to the sensory reweighting hypothesis, i.e., more reliable information is weighted more strongly than less reliable information. In this approach, no distinction has been made between the contributions of both legs. In this study, we investigated how proprioceptive information from both legs is combined to maintain upright stance. Healthy subjects maintained balance with eyes closed while proprioceptive information of each leg was perturbed independently by continuous rotations of the support surfaces (SS) and the human body by platform translation. Two conditions were tested: perturbation amplitude of one SS was increased over trials while the other SS 1) did not move or 2) was perturbed with constant amplitude. With the use of system identification techniques, the response of the ankle torques to the perturbation amplitudes (i.e., the torque sensitivity functions) was determined and how much each leg contributed to stabilize stance (i.e., stabilizing mechanisms) was estimated. Increased amplitude of one SS resulted in a decreased torque sensitivity. The torque sensitivity to the constant perturbed SS showed no significant differences. The properties of the stabilizing mechanisms remained constant during perturbations of each SS. This study demonstrates that proprioceptive information from each leg is weighted independently and that the weight decreases with perturbation amplitude. Weighting of proprioceptive information of one leg has no influence on the weight of the proprioceptive information of the other leg. According to the sensory reweighting hypothesis, vestibular information must be up-weighted, because closing the eyes eliminates visual information.

Age-Related Changes in Postural Responses to Backward Platform Translation

The aim of the study was to investigate age-related changes in postural responses to platform translation with 3 various velocities. We focused on the influence of linear velocity using the smoothed profile of platform acceleration (till 100 cm.s−2). Eleven healthy young (20-31 years) and eleven healthy elderly (65-76 years) subjects were examined. The subjects stood on the force platform with their eyes closed. Each trial (lasting for 8 sec) with different velocity (10, 15, 20 cm.s−1) of 20 cm backward platform translation was repeated 4 times. We have recorded displacements of the centre of pressure (CoP) and the EMG activity of gastrocnemius muscle (GS) and tibialis anterior muscle (TA). The results showed increased maximal values of CoP responses to the platform translation. There was also observed a scaling delay of CoP responses to platform translation with different velocities in elderly. The EMG activity of GS muscle during backward platform translation was of about similar shape in both groups during the slowest platform velocity, but it increased depending on rising velocity. EMG activity of TA was not related to the platform velocity. Early parts of postural responses showed significant co-activation of TA and GS muscles of elderly. It is likely that elderly increased body stiffening in order to help their further balance control.