Medical case retrieval from a committee of decision trees

Gwénolé Quellec (1, 2) , Mathieu Lamard (2, 3) , Lynda Bekri (4) , Guy Cazuguel (1, 2) , Christian Roux (1, 2) , Béatrice Cochener (2, 3, 4)

1 ITI - Département Image et Traitement Information
2 LaTIM - Laboratoire de Traitement de l'Information Medicale
3 UBO - Université de Brest
4 CHRU Brest - Centre Hospitalier Régional Universitaire de Brest

A novel content-based information retrieval framework, designed to cover several medical applications, is presented in this paper. The presented framework allows the retrieval of possibly incomplete medical cases consisting of several images together with semantic information. It relies on a committee of decision trees, decision support tools well suited to process this type of information. In our proposed framework, images are characterized by their digital content. It was applied to two heterogeneous medical datasets for computer aided diagnosis: a diabetic retinopathy follow-up dataset (DRD) and a mammography screening dataset (DDSM). Measure of precision among the top five retrieved results of 0.788+-0.137 and 0.869+-0.161 was obtained on DRD and DDSM, respectively. On DRD for instance, it increases by half the retrieval of single images.

Mots clés: Information retrieval; Decision trees; CBIR; CAD; Medical databases

Domaines: Apprentissage [cs.LG]; Base de données [cs.DB]; Traitement du signal et de l'image [eess.SP]; Machine Learning [stat.ML]; Traitement du signal et de l'image [eess.SP]

Gwénolé Quellec, Mathieu Lamard, Lynda Bekri, Guy Cazuguel, Christian Roux, et al.. Medical case retrieval from a committee of decision trees. IEEE Transactions on Information Technology in Biomedicine, 2010, 14 (5), pp.1227-1235. ⟨hal-00515356⟩