Ontology-based automatic generation of computerized cognitive - - PowerPoint PPT Presentation

▶
ontology based automatic generation of computerized
SMART_READER_LITE
LIVE PREVIEW

Ontology-based automatic generation of computerized cognitive - - PowerPoint PPT Presentation

Dec 26, 2023 357 likes •597 views

Oral presentation at MIE2011 Ontology-based automatic generation of computerized cognitive exercises Giorgio Leonardi a,c , Silvia Panzarasa b and Silvana Quaglini a Laboratory for Biomedical a Dept. Computer Eng. & Systems Science,

slide-1
SLIDE 1

Laboratory for Biomedical Informatics “Mario Stefanelli”

Ontology-based automatic generation of computerized cognitive exercises

Giorgio Leonardi a,c, Silvia Panzarasa b and Silvana Quaglini a

a Dept. Computer Eng. & Systems Science, University of Pavia, b CBIM, Pavia c Dept. Computer Science, University Piemonte Orientale

Italy

Mie2011, Quaglini 1

Oral presentation at MIE2011

Oslo, 27-31 August 2011

slide-2
SLIDE 2

Cognitive rehabilitation exercises: examples

Choose the correct category

slide-3
SLIDE 3

… examples

Complete the logical similarity

Spaghetti : PASTA = Hamburger : ?

MEAT VEGETABLE CEREAL

slide-4
SLIDE 4

…examples

Mie2011, Quaglini 4

Ordering

  • f scenes
slide-5
SLIDE 5

Background

l Several (research/commercial) systems exist

for moving cognitive rehabilitation

l from paper to computer l from face-to-face encounters to homecare

l Not to replace the therapist but to enforce and

intensify the rehabilitation

l Among open problems

l Automatic patient-tailoring

l According to patient’s skill/performances l According to patient’s preferences

l Generating ever new exercises

Mie2011, Quaglini 5

?

COGREHAB REHACOM ANASTASIS

Serious games

slide-6
SLIDE 6

Aim of the work

Mie2011, Quaglini 6

  • To build an architecture allowing easy maintainance and

updating of a repository of stimuli (images, words, sounds, etc.) to be used for the exercise generation, with

  • Hierarchical classification of stimuli
  • Full description of every stimulus
  • Relationships among stimuli
  • Relationships between stimuli and specific patients
  • Using the repository to generate personalised exercises within

the E-prime tool

  • Maintaining the separation between the stimuli repository and the

exercise-generator software, for sake of re-usability

slide-7
SLIDE 7

Methods: Protégé for the stimuli ontology

Mie2011, Quaglini 7

An ontology is composed by a hierarchy of classes (containing the domain concepts), attributes (defining the intrinsic properties of a class) and relationships (defining semantic links between different classes).

slide-8
SLIDE 8

Export from Protégé

Mie2011, Quaglini 8

slide-9
SLIDE 9

From Protégé to a Relational DB

Mie2011, Quaglini 9

Translator tool

slide-10
SLIDE 10

Adding patient’s data at local level

Mie2011, Quaglini 10

Patient’s data Patient’s profiling

slide-11
SLIDE 11

Exploiting ontological relationships

Mie2011, Quaglini 11

slide-12
SLIDE 12

To create ever new exercises

Mie2011, Quaglini 12

These exercises exploit the relationships between a course and its ingredients and between an animal and its habitations

slide-13
SLIDE 13

To make exercise more/less difficult

Spaghetti : PASTA = Hamburger : ?

MEAT VEGETABLE CEREAL

Spaghetti : PASTA = Hamburger : ?

MEAT VEGETABLE CEREAL FISH CHEESE

slide-14
SLIDE 14

To tailor exercise according to a patient’s profile

ü Showing the patient’s dog instead of a generic dog ü Using classes corresponding to the patient’s hobbies and interests ü Using easy or difficult stimuli according to the patient’s scholarity ü ...

slide-15
SLIDE 15

Using the same relationship “OPPOSITE”

... less difficult

slide-16
SLIDE 16

... More difficult

Using different relationships uses Made of Lives in

slide-17
SLIDE 17

Food Animal Habitation Food First Second Dessert … Animals Mammals Birds Insects ... Habitation For humans

House Floor

… For animals

Stable Nest …

Different Complexity for the same exercise

Find the right category

slide-18
SLIDE 18

Increasing complexity by going into subclasses

Food First Second Dessert … Animals Mammals Birds Insects ... Habitation For humans

House Floor

… For animals

Stable Nest …

Find the right category

slide-19
SLIDE 19

Conclusions

l Every update to the common stimuli repository is made at

the ontology level to maintain consistency

l At the local level

l The patient’s data are integrated l the concept table may be enriched with patient-specific stimuli l The “interest” relationship between concepts and patient may be

valued

l This keep knowledge level and data level separate,

allowing

l easy personalisation l re-using the stimuli ontology in different contexts

Mie2011, Quaglini 19

slide-20
SLIDE 20

The final architecture

Mie2011, Quaglini 20

stimuli

slide-21
SLIDE 21

Acknowledgments

l The medical staff of

l the Neurological Hospital “C. Mondino”, Pavia

(Prof. Sandrini, Dr. Sinforiani, Dr. Zucchella)

l the Rehabilitation Hospital “S. Maugeri”, Pavia

(Dr. Pistarini, Dr. Cattani)

Mie2011, Quaglini 21

slide-22
SLIDE 22

Thank you

www.labmedinfo.org silvana.quaglini@unipv.it

Mie2011, Quaglini 22

Laboratory of Biomedical Informatics “Mario Stefanelli” Department of Computer Science and Systems, University of Pavia, Italy

slide-23
SLIDE 23

One-month report