Fuzzy Decision Tree for the Objective Explanation of Subjective Functional Evaluation: Application to the Upper Limb

Emmanuel Roux (1) , Patrice Caulier (1) , Anne-Pascale Godillon-Maquinghen (1) , Stéphane Bouilland (2) , Denis Bouttens (3)

1 LAMIH - Laboratoire d'Automatique, de Mécanique et d'Informatique industrielles et Humaines - UMR 8201
2 Fondation Hopale/institut du handicap de Berck-sur-mer (IHB)
3 Institut Calot

The purpose of this paper is to contribute to the kinematics data interpretation in the domain of the shoulder arthroplasty and the associated rehabilitation process. Indeed this practice suffers from a lack of hindsight and from a lack of objective evaluation of its results. At first, the measurement protocol is briefly described and the data characterizing and coding methods are presented. An objective explanation of the functional evaluation of the upper limb is derived from the fuzzy rules generated by fuzzy decision trees. Then the method is evaluated.

Mots clés: Machine learning; reuse of knowledge; fuzzy decision trees

Domaines: Intelligence artificielle [cs.AI]; Ingénierie biomédicale; Anatomie, Histologie, Anatomopathologie [q-bio.TO]

Emmanuel Roux, Patrice Caulier, Anne-Pascale Godillon-Maquinghen, Stéphane Bouilland, Denis Bouttens. Fuzzy Decision Tree for the Objective Explanation of Subjective Functional Evaluation: Application to the Upper Limb. Workshop Intelligent Data Analysis in Medicine and Pharmacology (IDAMAP), held during the European Conference on Artificial Intelligence (ECAI), 2002, Lyon, France. ⟨hal-01372740⟩