At the Computational Cognition Laboratory, three projects are

advertisement
At the Computational Cognition Laboratory, three projects are currently been pursued:
1. Human Language Technologies for supporting the diagnosis and rehabilitation of
language disorders
The project aims to cross-fertilize recent advances in areas such as the neuropsychology of
language disorders, theories of meaning, computational language resources, and automatic
content extraction from corpora, in order to build an infrastructure that supports the diagnosis and
rehabilitation of language disorders.
The project has both a theoretical and a practical import. From a theoretical point of view it will
explore issues related to the coding of common sense information associated to the mental lexicon
that reflect and capture insights form the neuropsychology of language (including the role of
images in defining word meanings), and the automatic extraction thereof from corpora. The
insights gained will be used to develop a Neuropsychologist + Speech Therapy Helper, which the
clinician can use to prepare tasks and exercises for patients with neuropsychological language
deficits. After bootstrapping from existing ones, the needed resource will be built in a WEB-based,
wiki-like cooperative model, involving language therapists from all over Italy.
2.Advanced interfaces for the rehabilitation of cognitive disorders
Clinicians make use of a wide variety of stimuli (words, sentences, images, etc.) during the
diagnosis and clinical treatment of cognitive disorders. That material must be prepared in advance
before presenting it to the patient; the responses of the latter must be recorded, along with relevant
aspects of the context, for the clinician to later on assess the session’s output and prepare the
following one. This complex process in which the therapist-patient dyad is involved can be
improved and made more effective if appropriate interactive systems are provided that assist them
in the clinical process. The challenge here is to design advanced interaction modalities that suite
the greatly varying needs of the different patient-therapists dyads; we will face it by means of a
User Centred Design approach, whereby both the therapist and the patients are constantly
involved in the design process.
3.Assisted Cognition
Whatever their origins, cognitive problems in adults and the older people show a steady increase
that follows the demographic trend in advanced countries. The quality of life of those people and of
their caregivers can be greatly enhanced by the availability of intelligent systems that exploit
advanced environmental sensing capabilities (through cameras, microphones, RFID sensors, etc.),
monitor and understand the person behaviour, and provide useful functionalities to patients,
caregivers, and clinicians. For instance, if the person has forgotten to execute or complete an
important activity (e.g., tooth brushing), the system can remind her about it, or tell the caregiver
that she needs to be reminded; if tooth brushing is badly executed, instructions and guidance for
the correct execution can be provided, etc. Clinician could crucially benefit from such systems, too,
thanks to the possibility of a much more precise and refined monitoring of the patient’s conditions
and
capabilities,
as
they
are
deployed
in
her
natural
environment.
In this ambitious project (pursued in collaboration with CreateNet and the Aware Home project at
GeorgiaTec) we focus on a limited number of simple activities of daily living (dressing/undressing,
hand washing, preparing coffee), emphasising the automatic recognition of anomalies, and the
delivery of functionalities that are useful for the patient, the caregiver and the clinician. The initial
experimentation will be conducted at the CeRiN premise; extensive tests will then be conducted at
patients’ homes.
Download